'89 Jordan vs '18 LeBron

[Heej]

Yes, I agree, but the post of mine you responded to was specifically about the 1989 Bulls being significantly bigger underdogs compared to the 1989 Cavs than the 2018 Cavs were compared to the 2018 Raptors. Which was a point I made in response to someone basically arguing that 2018 LeBron was better than 1989 Jordan because the 1989 Bulls had more trouble with the 1989 Cavs than the 2018 Cavs did with the 2018 Raptors. That argument makes no sense if the 1989 Bulls were way bigger underdogs against the 1989 Cavs than the 2018 Cavs were against the 2018 Raptors—hence why I’ve pointed out that that is precisely the case. And, given the title odds I mentioned, my point is clearly correct even when accounting for the fact that the 2018 Cavs were in “complete disarray.” And this is because of some combination of the fact that the 1989 Bulls were an even worse team than the 2018 Cavs (even despite the Cavs’ disarray), and the 1989 Cavs were better than the 2018 Raptors. In other words, I basically agree with what you’re saying, but don’t think it is at all mutually exclusive with what I’m saying.

Again, I think you guys just went way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff and it ruined your guys' ability to see the thread for what it is. Betting odds have almost everything to do with perception, that's what you don't seem to be fully getting here. Vegas is in the business of making money more than deciding which team is better which is why betting odds fluctuate based on which team gets more bets. So using who Vegas favored is not some infallible tool to bring up in a bb forum. I would think you are old enough to understand this stuff. LeBron's aura of invincibility vs the east in 2018 was on a completely different level than MJ's was back in 1989 when he hadn't even made it out of the 2nd rd yet. Re 89 Bulls roster vs 18 Cavs, it's not really clear which roster was better. If the Cavs was better it’s simply because of having some guys who could hit the 3 around that version of LeBron. The Bulls otoh were built on defense and letting MJ be the offense. At the very least the Bulls team had way more experience having played together which counts for something.

Yes, the fact that betting odds fluctuate based on which team gets more bets is actually precisely the reason they’re informative, because it means the betting odds end up substantially reflecting contemporaneous consensus views of the public regarding the teams (especially in a highly-bet-on betting market such as NBA title odds). It’s akin to a publicly-traded company’s stock market value (i.e. market cap) reflecting the public’s consensus view regarding a company’s value. Of course, the public’s consensus view can be wrong, but it is still a relevant data point here. Notably, though, my argument didn’t just rely on that anyways. For instance, I also presented data showing that the 1989 Bulls got completely destroyed by the Cavs when Jordan was off the court, while the 2018 Cavs outscored the 2018 Raptors with LeBron off the court. And I also presented data regarding the historical playoff SRS of the Daugherty-era Cavs against non-Bulls teams compared to the historical playoff SRS of the pre-Kawhi Raptors against non-Cavs teams. I and others presented data about random variance with regards to three-point shooting and FT shooting in those series. I presented a lot of information on this that all pretty clearly points to the fact that it is silly to draw a conclusion that 2018 LeBron > 1989 Jordan simply because the 1989 Bulls had a tougher time beating the 1989 Cavs than the 2018 Cavs did in beating the 2018 Raptors. A comparison of title odds is just one of those data points. I think the overall picture is pretty inarguable.

As for going “way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff,” I completely agree with you. It was a silly point that someone else led with as a major argument, and my point from the beginning was that the comparison between the two series was being wildly overblown in its importance. Others have simply continued trying to assert that that’s not the case. I’d be happy to stop discussing it, since I think it’s manifestly silly and unimportant, but others seem to think it is a crucially important point in favor of 2018 LeBron, so I’ve continued to address it (albeit somewhat reluctantly and nearing the end of my patience with it).

Eh the stock market point is precisely why using Vegas betting odds doesn't work. Few understand this but what actually moves markets is inventory. People are told and believe that market movements are based on real time reflections of supply and demand when in reality it's more akin to exchanges moving lines to entice the most bets that they can offload inventory (people's orders that need to be matched with opposing orders; aka every buyer needs a seller) into. It's never real time and can never perfectly reflect fundamental value because exchanges themselves exist and need to make money scalping fees off of transactions which inherently introduces variance and erratic fluctuation which is then amplified by the psychological variance of participants in the market.

So in stock terms it seems you think that ultimately the closing price was the one most representative of fair value when in reality it was the mass of orders put in above or below the closing price that actually drove the market to where it ended up. What Cavsfan is trying to explain to you and you're failing to understand is that the lines moved more for LeBron because the market itself hit a psychological level where despite the Cavs being an inferior team there was enough volume betting on LeBron for exchanges to keep accumulating orders while guiding the lines upward and trapping everyone from lower numbers that "shorted the hole". There's absolutely no reason or evidence to believe that such a move is purely representative of the absolute qualitative change in the goodness of both teams. That doesn't even make sense if you understand markets. Influenced by it? Sure. A 1:1 near perfect measure of relative strength like you seem to believe? Hell to the nah

It's also an extreme logical fallacy to think that stocks are more an accurate representation of a company's fundamental worth than that they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps . This same deification of the all-knowing perfectly up to date eye of the market is why so many geniuses blew up these past 2 years thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value in real time. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines informs you of what you think it does chief.

[Tim Lehrbach]

Again, I think you guys just went way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff and it ruined your guys' ability to see the thread for what it is. Betting odds have almost everything to do with perception, that's what you don't seem to be fully getting here. Vegas is in the business of making money more than deciding which team is better which is why betting odds fluctuate based on which team gets more bets. So using who Vegas favored is not some infallible tool to bring up in a bb forum. I would think you are old enough to understand this stuff. LeBron's aura of invincibility vs the east in 2018 was on a completely different level than MJ's was back in 1989 when he hadn't even made it out of the 2nd rd yet. Re 89 Bulls roster vs 18 Cavs, it's not really clear which roster was better. If the Cavs was better it’s simply because of having some guys who could hit the 3 around that version of LeBron. The Bulls otoh were built on defense and letting MJ be the offense. At the very least the Bulls team had way more experience having played together which counts for something.

Yes, the fact that betting odds fluctuate based on which team gets more bets is actually precisely the reason they’re informative, because it means the betting odds end up substantially reflecting contemporaneous consensus views of the public regarding the teams (especially in a highly-bet-on betting market such as NBA title odds). It’s akin to a publicly-traded company’s stock market value (i.e. market cap) reflecting the public’s consensus view regarding a company’s value. Of course, the public’s consensus view can be wrong, but it is still a relevant data point here. Notably, though, my argument didn’t just rely on that anyways. For instance, I also presented data showing that the 1989 Bulls got completely destroyed by the Cavs when Jordan was off the court, while the 2018 Cavs outscored the 2018 Raptors with LeBron off the court. And I also presented data regarding the historical playoff SRS of the Daugherty-era Cavs against non-Bulls teams compared to the historical playoff SRS of the pre-Kawhi Raptors against non-Cavs teams. I and others presented data about random variance with regards to three-point shooting and FT shooting in those series. I presented a lot of information on this that all pretty clearly points to the fact that it is silly to draw a conclusion that 2018 LeBron > 1989 Jordan simply because the 1989 Bulls had a tougher time beating the 1989 Cavs than the 2018 Cavs did in beating the 2018 Raptors. A comparison of title odds is just one of those data points. I think the overall picture is pretty inarguable.

As for going “way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff,” I completely agree with you. It was a silly point that someone else led with as a major argument, and my point from the beginning was that the comparison between the two series was being wildly overblown in its importance. Others have simply continued trying to assert that that’s not the case. I’d be happy to stop discussing it, since I think it’s manifestly silly and unimportant, but others seem to think it is a crucially important point in favor of 2018 LeBron, so I’ve continued to address it (albeit somewhat reluctantly and nearing the end of my patience with it).

Eh the stock market point is precisely why using Vegas betting odds doesn't work. Few understand this but what actually moves markets is inventory. People are told and believe that market movements are based on real time reflections of supply and demand when in reality it's more akin to exchanges moving lines to entice the most bets that they can offload inventory (people's orders that need to be matched with opposing orders; aka every buyer needs a seller) into. It's never real time and can never perfectly reflect fundamental value because exchanges themselves exist and need to make money scalping fees off of transactions which inherently introduces variance and erratic fluctuation which is then amplified by the psychological variance of participants in the market.

So in stock terms it seems you think that ultimately the closing price was the one most representative of fair value when in reality it was the mass of orders put in above or below the closing price that actually drove the market to where it ended up. What Cavsfan is trying to explain to you and you're failing to understand is that the lines moved more for LeBron because the market itself hit a psychological level where despite the Cavs being an inferior team there was enough volume betting on LeBron for exchanges to keep accumulating orders while guiding the lines upward and trapping everyone from lower numbers that "shorted the hole". There's absolutely no reason or evidence to believe that such a move is purely representative of the absolute qualitative change in the goodness of both teams. That doesn't even make sense if you understand markets. Influenced by it? Sure. A 1:1 near perfect measure of relative strength like you seem to believe? Hell to the nah

It's an extremely fallacy to think that stocks are more an accurate representation of a company's fundamental worth than they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps and why so many geniuses blew up this year thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines is what you think it is chief.

Great post that should put some silliness to rest.

At the same time, could the respondents to the pro-Jordan side acknowledge the anecdotal, contemporaneous accounts of 2018 LeBron? And could the pro-Jordan side accept their burden to show why these accounts should affect our view of LeBron's season? The narrative is not proven by the impact data, but it remains a valid point that the narrative, based on observation, was live during the season.

When data are absent we frequently rely on contemporary accounts from journalists and other observers to inform our understanding of what people said at the time. They can be wrong, of course, but we take it as the state of discourse at the time. Why should it count for Chamberlain but not for James? In any case, impact data are not positive confirmation of quality of play; there are too many variables involved in what yields positive or negative impact data. Impact data tell us what happened to the score when a player played, not how they played.

Rather, the data serve helpfully to disprove false narratives. In this case, I think the data soften any blows LeBron's detractors might land as to the consequences of his defense. He simply wasn't killing the Cavs out there, the way it might be imagined. What the data do not and cannot tell us definitively is whether LeBron "played good (or acceptable) defense."

What the detractors are on the hook for, in short, is demonstrating why their complaints matter. What did LeBron do to hurt Cleveland's chances at winning? If the impact data do not tell your story, what does? It cannot be strictly eye test that LeBron took plays off or wasn't trying because it's entirely possible that was schemed for, perhaps successfully. The team results, which were pitiful, do count here. But you need to be able to explain how LeBron was responsible. I haven't seen it.

[lessthanjake]

They were better, and then their best player missed a close loss before rushing a visibly hampered return — something you laughably tried to brush away, because agendas trump sincere assessment.

Mark Price was genuinely probably the Cavs’ best player in one of the games the Cavs won. He missed one game, played in four games, and was good in two of them (which the Cavs split) and pretty bad in one of them (which the Cavs still won) and very bad in another (which the Cavs lost). That very bad game was not his first game back from injury—it was largely just a poor game from him.

… So players get one game of recovery and then are officially at their normal level?

Not sure which is funnier: the idea that you think people will buy that, or that you are even willing to try to sell it.

And the Bulls closed out the series with Price playing well. Obviously, Price’s injury helped the Bulls in the series, but it’s laughable to suggest a 7.95 SRS, +400 title odds team becomes like a 1.28 SRS, +10000 title odds team because one of their top few players misses one game and has an inconsistent series after coming back.

Thank you for emphasising how much you are willing to abandon basic analysis to push a dishonest narrative.

You do not believe the 1988 Cavaliers were a 1.3 SRS team that postseason. You do not believe one offseason of mild improvement across their key players produced a 6.6 SRS elevation. You do not believe that Price played anywhere near as well in the 1989 postseason as he did in the 1988 postseason. Given your commitment to pushing the 1990-93 Cavaliers, you are clearly not a Ron Harper true believer (the only guy who did elevate his performance from the prior year). Yet here you are, pretending all that may actually be true, because in this thread that happens to be the thing that would reflect best on Jordan.

And in the alternative where this is a matter of sincere belief, then I suppose we all really have been wasting our time trying to communicate with beliefs that disconnected from the realities of the sport.

And to suggest that while simultaneously accusing others of having “agendas” and being “invested in mythmaking” is the height of irony. Anyways, look, there’s always something available to cling to if you feel a need to come to a certain conclusion. Convincing yourself on this one is Olympic level mental gymnastics, but one can always find something. But it’s also not really worth discussing any further, because it all just goes back to an argument that is just manifestly dumb in its conception

Now this is the height of irony.

saying that a player was better than another player because that player’s team won a playoff series more easily than the other player’s team won a playoff series against a completely different team.

No one said this. What was said is that one was the more impressive accomplishment, and the only reason that even needed to be said was because of people exalting Jordan for a worse performance against a worse performing version of the team he had already beaten. No one would have anything positive to say about the 2018 Raptors sweep if Lowry had been injured that year instead of in 2017, but I suppose that is the downside of having real standards.

On its face that’s just an absurdly reductionist argument that no reasonable person would find convincing. If you’re convinced by it, then you just desperately want to be convinced of any conclusion that you like, and there’s therefore obviously no point in discussing with you. And if you’re not convinced by it, then there’s no point in discussing nitty-gritty things that merely relate to an argument neither of us think is valid. Either way, it’s not worth discussing.

Yet you have spent pages desperately pushing it, and not for the first time either.

As always, the immediate vindication is appreciated.

This is a joke of a post, in which you don’t really try to make any substantive defense of the underlying argument that you’re aiming to defend, and instead just opt to be as insulting as possible. Of course, that’s not uncommon from you, so I shouldn’t be surprised. In any event, if your position is that no one actually said or meant what I was arguing against (despite me very clearly summarizing what I took the argument to be, and the poster repeatedly responding without correcting me), then I imagine you are also of the position that no one thinks it either. And if that’s the case then we’re in the zone I described earlier where “there’s no point in discussing nitty-gritty things that merely relate to an argument neither of us think is valid.” Either no one made the argument, in which case there’s no point discussing counterarguments to it, or someone did make the argument and you are very clearly unwilling to substantively defend that argument, in which case there’s still no point in discussing it with you.

Again, I think you guys just went way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff and it ruined your guys' ability to see the thread for what it is. Betting odds have almost everything to do with perception, that's what you don't seem to be fully getting here. Vegas is in the business of making money more than deciding which team is better which is why betting odds fluctuate based on which team gets more bets. So using who Vegas favored is not some infallible tool to bring up in a bb forum. I would think you are old enough to understand this stuff. LeBron's aura of invincibility vs the east in 2018 was on a completely different level than MJ's was back in 1989 when he hadn't even made it out of the 2nd rd yet. Re 89 Bulls roster vs 18 Cavs, it's not really clear which roster was better. If the Cavs was better it’s simply because of having some guys who could hit the 3 around that version of LeBron. The Bulls otoh were built on defense and letting MJ be the offense. At the very least the Bulls team had way more experience having played together which counts for something.

Yes, the fact that betting odds fluctuate based on which team gets more bets is actually precisely the reason they’re informative, because it means the betting odds end up substantially reflecting contemporaneous consensus views of the public regarding the teams (especially in a highly-bet-on betting market such as NBA title odds). It’s akin to a publicly-traded company’s stock market value (i.e. market cap) reflecting the public’s consensus view regarding a company’s value. Of course, the public’s consensus view can be wrong, but it is still a relevant data point here. Notably, though, my argument didn’t just rely on that anyways. For instance, I also presented data showing that the 1989 Bulls got completely destroyed by the Cavs when Jordan was off the court, while the 2018 Cavs outscored the 2018 Raptors with LeBron off the court. And I also presented data regarding the historical playoff SRS of the Daugherty-era Cavs against non-Bulls teams compared to the historical playoff SRS of the pre-Kawhi Raptors against non-Cavs teams. I and others presented data about random variance with regards to three-point shooting and FT shooting in those series. I presented a lot of information on this that all pretty clearly points to the fact that it is silly to draw a conclusion that 2018 LeBron > 1989 Jordan simply because the 1989 Bulls had a tougher time beating the 1989 Cavs than the 2018 Cavs did in beating the 2018 Raptors. A comparison of title odds is just one of those data points. I think the overall picture is pretty inarguable.

As for going “way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff,” I completely agree with you. It was a silly point that someone else led with as a major argument, and my point from the beginning was that the comparison between the two series was being wildly overblown in its importance. Others have simply continued trying to assert that that’s not the case. I’d be happy to stop discussing it, since I think it’s manifestly silly and unimportant, but others seem to think it is a crucially important point in favor of 2018 LeBron, so I’ve continued to address it (albeit somewhat reluctantly and nearing the end of my patience with it).

Eh the stock market point is precisely why using Vegas betting odds doesn't work. Few understand this but what actually moves markets is inventory. People are told and believe that market movements are based on real time reflections of supply and demand when in reality it's more akin to exchanges moving lines to entice the most bets that they can offload inventory (people's orders that need to be matched with opposing orders; aka every buyer needs a seller) into. It's never real time and can never perfectly reflect fundamental value because exchanges themselves exist and need to make money scalping fees off of transactions which inherently introduces variance and erratic fluctuation which is then amplified by the psychological variance of participants in the market.

So in stock terms it seems you think that ultimately the closing price was the one most representative of fair value when in reality it was the mass of orders put in above or below the closing price that actually drove the market to where it ended up. What Cavsfan is trying to explain to you and you're failing to understand is that the lines moved more for LeBron because the market itself hit a psychological level where despite the Cavs being an inferior team there was enough volume betting on LeBron for exchanges to keep accumulating orders while guiding the lines upward and trapping everyone from lower numbers that "shorted the hole". There's absolutely no reason or evidence to believe that such a move is purely representative of the absolute qualitative change in the goodness of both teams. That doesn't even make sense if you understand markets. Influenced by it? Sure. A 1:1 near perfect measure of relative strength like you seem to believe? Hell to the nah

It's an extremely fallacy to think that stocks are more an accurate representation of a company's fundamental worth than they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps and why so many geniuses blew up this year thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines is what you think it is chief.

I think this is getting more in the weeds than is necessary here. Yes, neither betting markets nor stock markets are a truly perfect market. But that doesn’t mean that a highly-bet-on betting market isn’t quite informative regarding what the contemporaneous consensus perception of teams was. Essentially nothing I’ve said is really dependent on the precise numbers or requiring some sort of “1:1 near perfect measure” to make the point. It’s largely based on odds that weren’t even close. They don’t need to be perfect to support my point. I don’t think there’s any reason to think that such significant differences aren’t generally reflective of consensus perception in those years. Of course, perception can be wrong and markets aren’t perfect, but it’s certainly still a relevant data point regarding the strength of different teams (and was also far from the only data point I raised in that discussion).

As for the thing about people essentially betting on LeBron basically out of trust in him, I get the point, but another way of putting that is that that’s the market internalizing that LeBron was a historically great player. And they weren’t wrong! I don’t think you’re identifying actual overvaluing of the Cavs. Rather, you’re just identifying the pretty obvious fact that LeBron was a huge reason their odds were where they were. That’s no surprise! But markets of course internalized Jordan’s greatness too. I think if you wanted to make an argument along these lines, it’d probably be better articulated as that betting markets didn’t fully internalize Jordan’s greatness before he’d won a title. That’s probably true to some degree. But betting markets were fully capable of installing the Bulls as the pre-playoffs title favorite in 1991, before Jordan had won a title (and also as one of the top few favorites in 1990). And Jordan wasn’t really substantially better in 1990 or 1991 than in 1989—these were all peak years for him. So the Bulls’ long odds in 1989 really was more about the team not being good than about Jordan not being trusted until he won a title.

It's an extremely fallacy to think that stocks are more an accurate representation of a company's fundamental worth than they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps and why so many geniuses blew up this year thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines is what you think it is chief.

Yes, of course market cap is simply a representation of what the public thinks a company is worth (and not a perfect representation since markets aren’t perfect). But there’s wisdom in crowds, and so the public’s consensus view of what something is worth is certainly a good indicator of what it actually is worth (and more likely to be correct that just one individual person’s assessment). It is definitely a relevant data point! Similarly, betting odds are simply a representation of how good the public thinks teams are (and not a perfect one since markets aren’t perfect). But there’s wisdom in crowds, and so the public’s consensus view of how good different teams are is certainly a good indicator of how good they are. It’s obviously not perfect, but a data point doesn’t have to be perfect in order to be relevant and persuasive. If it did, then no one could make virtually any argument about anything. Ultimately, I’m not really sure what the objection is here. Are you of the view that title odds in past years don’t tell us anything about the strength of teams? Or are you just arguing that betting odds aren’t a perfect measure? It feels like it’s the latter, and, if so, then that’s just arguing against a straw man—I never suggested anything was a perfect measure (and, more generally, I very often say on these forums that essentially all data we have is flawed—something that this same crew of people has somehow criticized me for), and providing data does not mean that someone is saying that that data is perfect or infallible.

One other aside: For anyone reading this who may need to hear this, Heej is absolutely right that sports betting just takes money from people. Don’t bet on sports. It has negative expected value, and there are many places to put your money that have positive expected value. I am discussing what betting odds can tell us, but would never actually bet on sports because doing that is objectively a bad use of money.

[Tim Lehrbach]

lessthanjake, don't wanna keep quoting long posts, but if we acknowledge that betting markets are not about predictions but about taking people's money, isn't the game up about their predictive meaningfulness? I know you're arguing that it's not, but saying that the masses have an idea about team quality is inconclusive at best. We don't know what informs bets, but we do know that what informs markets is taking the maximum amount of action for the most money.

[lessthanjake]

lessthanjake, don't wanna keep quoting long posts, but if we acknowledge that betting markets are not about predictions but about taking people's money, isn't the game up about their predictive meaningfulness? I know you're arguing that it's not, but saying that the masses have an idea about team quality is inconclusive at best. We don't know what informs bets, but we do know that what informs markets is taking the maximum amount of action for the most money.

The reason sports betting is a bad use of money is because it’s designed to have negative expected value for bettors, which is achieved by the odds being a bit shorter than the actual implicit chances, in order to account for the oddsmakers’ cut. So let’s say the consensus was that a game is a legitimate 50/50 coin flip—the odds for both teams would move such that the two teams’ odds are the same, but they’d end up implying a slightly higher than 50% chance for each. So, for instance, they might end up being -120 for both teams (implying both teams have a 54.5% chance of winning). If you bet on one of those teams, you’d on average lose like 8 cents for every dollar you bet. It’s similar to why it is a bad use of money to play roulette. As it applies to title odds, the upshot is that if you bet money on every team, you’d lose money, because the odds are all slightly shorter than they would be in the absence of the oddsmakers’ basically charging a premium. But that doesn’t really suggest we can’t gain information from looking at the *relative* odds of two teams—which is what I was doing in this discussion. To use my first example for the sake of simplicity, if two teams are playing and there ends up being -120 odds for both teams in a highly-bet-on game, we can simultaneously say that it is a bad use of money to bet on that game, but also that the public consensus is that the two teams are evenly matched. I think the latter is relevant information that is informative regarding a discussion of those teams, because there’s wisdom in crowds, and the fact that there’s negative expected value on bets is only ancillary to that.

You say that it is “inconclusive at best” that “the masses have an idea about team quality,” but I’d generally disagree—IMO this is basically an application of price efficiency, in which an asset price (or, in this case, betting price) ends up reflecting all available information in the public domain (assuming it is a market that is highly-traded, which I think NBA title odds are). And I think that’s therefore a good data point to look at. And, actually, there’s probably a good argument that sports betting lines reflect even more than just all information publicly available. This is because there’s few real checks on quasi-insider-trading in sports betting, so if a betting line accurately reflects publicly available information but there’s non-public information that meaningfully contradicts that, there’s a good chance that people who have access to that non-public information will basically exploit that informational asymmetry to reap some arbitrage until the betting line eventually moves to more accurately reflect that non-public information.

Great post that should put some silliness to rest.

At the same time, could the respondents to the pro-Jordan side acknowledge the anecdotal, contemporaneous accounts of 2018 LeBron? And could the pro-Jordan side accept their burden to show why these accounts should affect our view of LeBron's season? The narrative is not proven by the impact data, but it remains a valid point that the narrative, based on observation, was live during the season.

When data are absent we frequently rely on contemporary accounts from journalists and other observers to inform our understanding of what people said at the time. They can be wrong, of course, but we take it as the state of discourse at the time. Why should it count for Chamberlain but not for James? In any case, impact data are not positive confirmation of quality of play; there are too many variables involved in what yields positive or negative impact data. Impact data tell us what happened to the score when a player played, not how they played.

Rather, the data serve helpfully to disprove false narratives. In this case, I think the data soften any blows LeBron's detractors might land as to the consequences of his defense. He simply wasn't killing the Cavs out there, the way it might be imagined. What the data do not and cannot tell us definitively is whether LeBron "played good (or acceptable) defense."

What the detractors are on the hook for, in short, is demonstrating why their complaints matter. What did LeBron do to hurt Cleveland's chances at winning? If the impact data do not tell your story, what does? It cannot be strictly eye test that LeBron took plays off or wasn't trying because it's entirely possible that was schemed for, perhaps successfully. The team results, which were pitiful, do count here. But you need to be able to explain how LeBron was responsible. I haven't seen it.

I’m confused why you say “the data soften any blows LeBron’s detractors might land as to the consequences of his defense” and that “the impact data does not tell [LeBron detractors’] story.” The Cavs’ defensive rating with LeBron on the court was *a lot* worse than with LeBron off the court. And, notably, there was no player on the Cavs that the Cavs defense was better without than LeBron. See below (a quote of me in an earlier post on this thread) for some more data on this:

Spoiler:

But the 2018 Cavs defense was ranked 2nd worst in the league! And it was actually an above-average defense in the minutes LeBron was off the floor! There’s nobody on the 2018 Cavs team that the defense did better without than LeBron. Depending on if we look at basketball-reference numbers or NBA.com numbers, the 2018 Cavs either had a 6.0 or 8.3 worse DRTG with LeBron on the court than off the court. Of course, sample sizes for a single season aren’t enormous (especially the “off” samples) so in theory this could just be random bad luck rather than reflecting real defensive weakness. But it comports with the eye test, in which LeBron wasn’t putting much effort in on the defensive end. And, FWIW, it also is consistent with the playoff data too, where the Cavs had either a 6.6 or 13.0 worse DRTG with LeBron on the court than off the court. Single-year playoff data is super low sample size so I don’t think that is worth much of anything on its own, but the fact that the playoff data on this was awful too lends more credence to the fact that LeBron really was just weak defensively that year, by adding even more to the sample size of the awfulness.

Those are big differences in DRTG. And, for what it’s worth, RAPM also rates LeBron’s defense quite badly that year—for instance, Engelmann’s single season RAPM had LeBron in the bottom 10% of the league in DRAPM. Similarly, the GitHub RAPM has LeBron’s DRAPM ranked in the bottom 5% of the league (and bottom 10% of the league in the playoffs too). I agree with you that impact data isn’t exactly *conclusive* about how any individual player played. But this data certainly doesn’t “disprove” any narrative about LeBron’s defensive weakness that year. Rather, it is supportive of those arguments. And the eye test is supportive of it too IMO.

[Tim Lehrbach]

Spoiler:

lessthanjake, don't wanna keep quoting long posts, but if we acknowledge that betting markets are not about predictions but about taking people's money, isn't the game up about their predictive meaningfulness? I know you're arguing that it's not, but saying that the masses have an idea about team quality is inconclusive at best. We don't know what informs bets, but we do know that what informs markets is taking the maximum amount of action for the most money.

The reason sports betting is a bad use of money is because it’s designed to have negative expected value for bettors, which is achieved by the odds being a bit shorter than the actual implicit chances, in order to account for the oddsmakers’ cut. So let’s say the consensus was that a game is a legitimate 50/50 coin flip—the odds for both teams would move such that the two teams’ odds are the same, but they’d end up implying a slightly higher than 50% chance for each. So, for instance, they might end up being -120 for both teams (implying both teams have a 54.5% chance of winning). If you bet on one of those teams, you’d on average lose like 8 cents for every dollar you bet. It’s similar to why it is a bad use of money to play roulette. As it applies to title odds, the upshot is that if you bet money on every team, you’d lose money, because the odds are all slightly shorter than they would be in the absence of the oddsmakers’ basically charging a premium. But that doesn’t really suggest we can’t gain information from looking at the *relative* odds of two teams—which is what I was doing in this discussion. To use my first example for the sake of simplicity, if two teams are playing and there ends up being -120 odds for both teams in a highly-bet-on game, we can simultaneously say that it is a bad use of money to bet on that game, but also that the public consensus is that the two teams are evenly matched. I think the latter is relevant information that is informative regarding a discussion of those teams, because there’s wisdom in crowds, and the fact that there’s negative expected value on bets is only ancillary to that.

You say that it is “inconclusive at best” that “the masses have an idea about team quality,” but I’d generally disagree—IMO this is basically an application of price efficiency, in which an asset price (or, in this case, betting price) ends up reflecting all available information in the public domain (assuming it is a market that is highly-traded, which I think NBA title odds are). And I think that’s therefore good information to look at. And, actually, there’s probably a good argument that sports betting lines reflect even more than just all information publicly available. This is because there’s few real checks on quasi-insider-trading in sports betting, so if a betting line accurately reflects publicly available information but there’s non-public information that meaningfully contradicts that, there’s a good chance that people who have access to that non-public information will basically exploit that informational asymmetry to reap some arbitrage until the betting line eventually moves to more accurately reflect that non-public information.

This is all a fair response. It may sound like a cop-out to reply by repeating myself, but accounting for the above, I yet find the wisdom of betting markets "inconclusive at best" in contributing to any meaningful measure of team quality, even relatively across teams or years. Your suppositions in the second paragraph are plausible. I am just not convinced, though I admit my preconception matches the broad outlines of what Heej posted, and it's entirely possible I misunderstand one or both of you.

Also, I should not have represented the discussion about odds as "silly." That is not helpful.

Spoiler:

Great post that should put some silliness to rest.

At the same time, could the respondents to the pro-Jordan side acknowledge the anecdotal, contemporaneous accounts of 2018 LeBron? And could the pro-Jordan side accept their burden to show why these accounts should affect our view of LeBron's season? The narrative is not proven by the impact data, but it remains a valid point that the narrative, based on observation, was live during the season.

When data are absent we frequently rely on contemporary accounts from journalists and other observers to inform our understanding of what people said at the time. They can be wrong, of course, but we take it as the state of discourse at the time. Why should it count for Chamberlain but not for James? In any case, impact data are not positive confirmation of quality of play; there are too many variables involved in what yields positive or negative impact data. Impact data tell us what happened to the score when a player played, not how they played.

Rather, the data serve helpfully to disprove false narratives. In this case, I think the data soften any blows LeBron's detractors might land as to the consequences of his defense. He simply wasn't killing the Cavs out there, the way it might be imagined. What the data do not and cannot tell us definitively is whether LeBron "played good (or acceptable) defense."

What the detractors are on the hook for, in short, is demonstrating why their complaints matter. What did LeBron do to hurt Cleveland's chances at winning? If the impact data do not tell your story, what does? It cannot be strictly eye test that LeBron took plays off or wasn't trying because it's entirely possible that was schemed for, perhaps successfully. The team results, which were pitiful, do count here. But you need to be able to explain how LeBron was responsible. I haven't seen it.

I’m confused why you say “the data soften any blows LeBron’s detractors might land as to the consequences of his defense” and that “the impact data does not tell [LeBron detractors’] story.” The Cavs’ defensive rating with LeBron on the court was *a lot* worse than with LeBron off the court. And, notably, there was no player on the Cavs that the Cavs defense was better without than LeBron. See below (a quote of me in an earlier post on this thread) for some more data on this:

Spoiler:

But the 2018 Cavs defense was ranked 2nd worst in the league! And it was actually an above-average defense in the minutes LeBron was off the floor! There’s nobody on the 2018 Cavs team that the defense did better without than LeBron. Depending on if we look at basketball-reference numbers or NBA.com numbers, the 2018 Cavs either had a 6.0 or 8.3 worse DRTG with LeBron on the court than off the court. Of course, sample sizes for a single season aren’t enormous (especially the “off” samples) so in theory this could just be random bad luck rather than reflecting real defensive weakness. But it comports with the eye test, in which LeBron wasn’t putting much effort in on the defensive end. And, FWIW, it also is consistent with the playoff data too, where the Cavs had either a 6.6 or 13.0 worse DRTG with LeBron on the court than off the court. Single-year playoff data is super low sample size so I don’t think that is worth much of anything on its own, but the fact that the playoff data on this was awful too lends more credence to the fact that LeBron really was just weak defensively that year, by adding even more to the sample size of the awfulness.

Yes, I had read that post, and yes, those are big disparities. I think you have already anticipated my responses: off samples are small, on samples are subject to collinearity, mostly. To the latter, you're right to point out the improved impact values of his teammates without LeBron, although still, I'd argue that lineups are dynamic and drawing conclusions about one input (player) will always be messy, hence the need for regression and a reasonable "adjustment." No need to belabor the raw on/off because I am detecting some degree of agreement that it's inconclusive (though supportive of your claims -- I'll come back to this). But there's more...

Those are big differences in DRTG. And, for what it’s worth, RAPM also rates LeBron’s defense quite badly that year—for instance, Engelmann’s single season RAPM had LeBron in the bottom 10% of the league in DRAPM. Similarly, the GitHub RAPM has LeBron’s DRAPM ranked in the bottom 5% of the league (and bottom 10% of the league in the playoffs too). I agree with you that impact data isn’t exactly *conclusive* about how any individual player played. But this data certainly doesn’t “disprove” any narrative about LeBron’s defensive weakness that year. Rather, it is supportive of those arguments.

...[bleep]. For the second time this irritatingly long, sleepless night, let me just admit I've spouted off without enough information to make the claims that I did. I did not see that RAPM rated LeBron's regular season defensive impact so poorly and was in too big of a hurry to get to my skepticism about making affirmative claims using impact data to realize that I'd run right past the actual data and committed either the same sin or its inverse, I'm not sure. That is, I assured myself that the data are not only inconclusive by the nature of the metric but that these data point in the direction opposite the LeBron "detractors'" claims about his defense. So, like Primedeion in the Kobe thread, you've caught me being inexcusably lazy. I relied on faulty memory in order to advance a claim that the data "disprove" your arguments, when the specific data points in question do nothing of the sort. Rather, you may use them in support of your case, though I think you understand that I am indeed skeptical.

Anyway, don't mistake my laziness for dishonesty, I'd ask. The flub with the RAPM exposes me as reverting to (poor) memory over doing the requisite research to participate in making any worthwhile arguments here. But I'm not here to talk about LeBron, anyway. I think you get my point regarding impact data generally. I do mean when I say that the burden of explanation is on the person trying to reach a broad conclusion using impact data. I suppose this isn't any different from claims supported by the eye test, or anecdotes, or any evidence, really.

I react especially strongly against affirmative claims from impact data, even adjusted for teammates and opponents (and whatever else the creator chooses) because, as I'll keep repeating on this board until somebody explains why I'm wrong about this too ( ), "impact" just means what happens to the scoreboard when a player or lineup is on the court. Raw plus-minus alone cannot tell us why one input (player) or one combination of inputs (lineup, opposing lineup) causes the score to change however it does. The "adjustment" (e.g., from box score, other priors, literally anything) is doing the explanatory work that we attribute to a given APM value. I realize this is an extremely elementary understanding of a diverse family of metrics.

I also realize I am not up on the state of the RAPM art. (For that ignorance, I blame the Wages of Wins debacle of the mid-2000s, in which one brand of "adjusted" stat was pitted against all others. I remember Dan Rosenbaum, in particular, being brutally honest about the role that chosen priors have in all of these metrics.) So maybe this has been significantly untangled in the past 15 years, and folks really do understand how they are using data when they draw strong inferences about a player or lineup's quality of play from impact data.

What I should have said is that I'm left wanting more from the arguments for LeBron's bad defense. And you'd fairly ask me what more I could possibly want beyond matching contemporary accounts, NBA possession tracking stats, and impact data. I'm not actually sure. I usually say that we're all just telling stories about the game using the best evidence and reasoning we have, and you've done that. Maybe, in this case, it's my own eye test, or my pro-LeBron bias, that needs to be checked. I also have a pitiful memory, so I'll need to re-re-read the thread to even remind myself what inspired my above post from earlier tonight in the first place (other than my ax to grind about impact data becoming claims about individual quality of play).

I've wasted everybody's time. You're welcome.

[AEnigma]

Mark Price was genuinely probably the Cavs’ best player in one of the games the Cavs won. He missed one game, played in four games, and was good in two of them (which the Cavs split) and pretty bad in one of them (which the Cavs still won) and very bad in another (which the Cavs lost). That very bad game was not his first game back from injury—it was largely just a poor game from him.

… So players get one game of recovery and then are officially at their normal level?

Not sure which is funnier: the idea that you think people will buy that, or that you are even willing to try to sell it.

And the Bulls closed out the series with Price playing well. Obviously, Price’s injury helped the Bulls in the series, but it’s laughable to suggest a 7.95 SRS, +400 title odds team becomes like a 1.28 SRS, +10000 title odds team because one of their top few players misses one game and has an inconsistent series after coming back.

Thank you for emphasising how much you are willing to abandon basic analysis to push a dishonest narrative.

You do not believe the 1988 Cavaliers were a 1.3 SRS team that postseason. You do not believe one offseason of mild improvement across their key players produced a 6.6 SRS elevation. You do not believe that Price played anywhere near as well in the 1989 postseason as he did in the 1988 postseason. Given your commitment to pushing the 1990-93 Cavaliers, you are clearly not a Ron Harper true believer (the only guy who did elevate his performance from the prior year). Yet here you are, pretending all that may actually be true, because in this thread that happens to be the thing that would reflect best on Jordan.

And in the alternative where this is a matter of sincere belief, then I suppose we all really have been wasting our time trying to communicate with beliefs that disconnected from the realities of the sport.

And to suggest that while simultaneously accusing others of having “agendas” and being “invested in mythmaking” is the height of irony. Anyways, look, there’s always something available to cling to if you feel a need to come to a certain conclusion. Convincing yourself on this one is Olympic level mental gymnastics, but one can always find something. But it’s also not really worth discussing any further, because it all just goes back to an argument that is just manifestly dumb in its conception

Now this is the height of irony.

saying that a player was better than another player because that player’s team won a playoff series more easily than the other player’s team won a playoff series against a completely different team.

No one said this. What was said is that one was the more impressive accomplishment, and the only reason that even needed to be said was because of people exalting Jordan for a worse performance against a worse performing version of the team he had already beaten. No one would have anything positive to say about the 2018 Raptors sweep if Lowry had been injured that year instead of in 2017, but I suppose that is the downside of having real standards.

On its face that’s just an absurdly reductionist argument that no reasonable person would find convincing. If you’re convinced by it, then you just desperately want to be convinced of any conclusion that you like, and there’s therefore obviously no point in discussing with you. And if you’re not convinced by it, then there’s no point in discussing nitty-gritty things that merely relate to an argument neither of us think is valid. Either way, it’s not worth discussing.

Yet you have spent pages desperately pushing it, and not for the first time either.

As always, the immediate vindication is appreciated.

This is a joke of a post, in which you don’t really try to make any substantive defense of the underlying argument that you’re aiming to defend, and instead just opt to be as insulting as possible. Of course, that’s not uncommon from you, so I shouldn’t be surprised. In any event, if your position is that no one actually said or meant what I was arguing against (despite me very clearly summarizing what I took the argument to be, and the poster repeatedly responding without correcting me), then I imagine you are also of the position that no one thinks it either. And if that’s the case then we’re in the zone I described earlier where “there’s no point in discussing nitty-gritty things that merely relate to an argument neither of us think is valid.” Either no one made the argument, in which case there’s no point discussing counterarguments to it, or someone did make the argument and you are very clearly unwilling to substantively defend that argument, in which case there’s still no point in discussing it with you.

Right, there is nothing to discuss. You try to make a show of not caring but clearly care and are committed to saying whatever you possibly can to make Jordan look impressive. I do not care that much aside from pointing out the misrepresentations you need to run to in order to selectively care about SRS, selectively care about health, selectively care about perception, selectively care about surrounding signals, selectively care about actual performance… So sure, it is a joke, but apparently you cannot distinguish whether the laughter is with, or at.

If you want to treat the 1989 Cavaliers series as a big accomplishment, then the 2018 Raptors series is an easy and obvious retort of a bigger one — albeit one without the same immortality, because sweeps are much less interesting and by that point Lebron had won 22 consecutive series in the conference. If you are willing to be honest that Jordan had an individually more impressive series the prior year, then no one is going to have trouble acknowledging that Lebron was individually more impressive against both the Pacers and the Celtics. But when you instead start trying to say 7.9 (injured) minus 2.1 > 7.3 minus 0.6, this is what happens.

I think if you wanted to make an argument along these lines, it’d probably be better articulated as that betting markets didn’t fully internalize Jordan’s greatness before he’d won a title. That’s probably true to some degree. But betting markets were fully capable of installing the Bulls as the pre-playoffs title favorite in 1991, before Jordan had won a title (and also as one of the top few favorites in 1990). And Jordan wasn’t really substantially better in 1990 or 1991 than in 1989—these were all peak years for him. So the Bulls’ long odds in 1989 really was more about the team not being good than about Jordan not being trusted until he won a title.

Yes, so interesting how the Bulls could radically change their odds from the 1989 offseason to the 1990 offseason without any significant roster changes. Must be all the training reports that Pippen was going to be an all-NBA contender that year, and definitely not people seeing a team make a 6-seed conference finals run with notable contributions from two young and highly drafted co-stars before making a positively received coaching change.

In the 1989 offseason, the Bulls traded away their second-best player for a faded Bill Cartwright, and Pippen and Grant were not known quantities. Jordan to that point had won 5 playoff games in his career. Whatever bettors think of all that is not a meaningful reflection on what the team was for the 1989 playoffs, which is what would primarily be informing the 1990 markets and is — absurd that this apparently even needs to be said — the most pertinent time for assessing what the teams would be expected to do in the playoffs or in any given series.

Again, though, that would be less convenient to our hagiographies, so instead we need to make a big deal of the 2018 preseason Cavaliers, complete with a theoretical Isaiah Thomas replacement for Kyrie and a newly signed Dwyane Wade and Derrick Rose fresh off decent enough seasons, being much more impressive than a 2018 Raptors team that surprised everyone. Relevant information if anyone here had been trying to argue the public at no point believed in the team, comically irrelevant to their play entering the second round series. But you do not have betting information for the 1989 Cavaliers/Bulls; all you have are the preseason odds, and those preseason odds offer the opportunity to push the exact narrative you want, so that inevitably becomes the standard of assessment.

[lessthanjake]

Yes, so interesting how the Bulls could radically change their odds from the 1989 offseason to the 1990 offseason without any significant roster changes. Must be all the training reports that Pippen was going to be an all-NBA contender that year, and definitely not people seeing a team make a 6-seed conference finals run with notable contributions from two young and highly drafted co-stars before making a positively received coaching change.

In the 1989 offseason, the Bulls traded away their second-best player for a faded Bill Cartwright, and Pippen and Grant were not known quantities. Jordan to that point had won 5 playoff games in his career. Whatever bettors think of all that is not a meaningful reflection on what the team was for the 1989 playoffs, which is what would primarily be informing the 1990 markets and is — absurd that this apparently even needs to be said — the most pertinent time for assessing what the teams would be expected to do in the playoffs or in any given series.

Again, though, that would be less convenient to our hagiographies, so instead we need to make a big deal of the 2018 preseason Cavaliers, complete with a theoretical Isaiah Thomas replacement for Kyrie and a newly signed Dwyane Wade and Derrick Rose fresh off decent enough seasons, being much more impressive than a 2018 Raptors team that surprised everyone. Relevant information if anyone here had been trying to argue the public at no point believed in the team, comically irrelevant to their play entering the second round series. But you do not have betting information for the 1989 Cavaliers/Bulls; all you have are the preseason odds, and those preseason odds offer the opportunity to push the exact narrative you want, so that inevitably becomes the standard of assessment.

This entire post seems premised on the idea that I’m talking about preseason odds—which is not the case at all, and I’ve been very explicit over and over again about what I’ve been talking about, so the confusion here from you is baffling and makes all the more clear that it is not a discussion worth having.

[lessthanjake]

.

I think you’ve made a perfectly reasonable response to my post. Given our exchange, at this point, I think it’s pretty clear that we both understand and acknowledge the points that the other is making. There’s a lot of data out there in basketball, and all of it has serious flaws, so I think it’s natural for people to not be in agreement as to how much they value certain data points. So, for instance, you’re certainly free to find teams’ title odds to be less persuasive an indicator of team strength than I do, despite the points I made about why I value them the way I do. After all, it’s certainly not a perfect measure! And I can’t really have any objection to you not being convinced on LeBron’s 2018 defense, when you’re simultaneously acknowledging that the argument uses the best evidence and reasoning we have. If you acknowledge the data that has been provided but are not convinced because your eye test disagrees with that evidence, then that’s fine! For anyone, their eye test is a major data point too! And I wouldn’t say any statistical data in basketball is infallible to the point of rendering eye test irrelevant. My own eye test differs from you on this one, which helps lead me to a different conclusion than you, but I’d be wrong to suggest that any data I provide can conclusively prove your eye test is wrong. I think data can allow me to make a probabilistic judgment that I think I am more likely than not correct, but there’s always uncertainty because data is flawed, and so if you respectfully disagree with my conclusion (which you do), then that’s fine!

[Djoker]

Given that betting is a billion dollar industry and there is nothing people (bettors and companies) want to do more than make money, it's hard for me to believe that betting odds are far off in terms of assessing team quality. If these huge companies set the odds incorrectly, you better believe bettors will exploit that and profit at their expense consistently. So there is STRONG INCENTIVE to set the odds just right!

[OhayoKD]

Finally, referring to the 2019 Cavs as “role player minute swaps” is an interesting (and pretty misleading) way of describing a situation where the team’s remaining best player played virtually zero minutes all season with the 7 other players that had played the most minutes the prior season (as I showed).

Idk, I think it's an accurate way to describe a team of role-players who still look terrible when we cherrypick minutes and games with the one role player who has ever evidenced being anything more...

There's also the cavs with kyrie and love (and no lebron) looking bad by both spot minute samples and actual games...weren't you saying 2016 Lebron's "help" was as talented as the 2016 Steph's?

And then there is the matter of the 6 point defensive drop for cleveland...the 2019 lakers becoming one of the best defenses(and improving to a 50-win pace) until Lebron **** up his groin vs the warriors(and tbf, bi and lonzo got injured shortly after)...

No, no. definitely not noise.

.

And I can’t really have any objection to you not being convinced on LeBron’s 2018 defense, when you’re simultaneously acknowledging that the argument uses the best evidence and reasoning we have.

Tip:

He who must say he is a user the best evidence and reasoning we have is no true user of the best reasoning and evidence we have.

[Heej]

Yes, the fact that betting odds fluctuate based on which team gets more bets is actually precisely the reason they’re informative, because it means the betting odds end up substantially reflecting contemporaneous consensus views of the public regarding the teams (especially in a highly-bet-on betting market such as NBA title odds). It’s akin to a publicly-traded company’s stock market value (i.e. market cap) reflecting the public’s consensus view regarding a company’s value. Of course, the public’s consensus view can be wrong, but it is still a relevant data point here. Notably, though, my argument didn’t just rely on that anyways. For instance, I also presented data showing that the 1989 Bulls got completely destroyed by the Cavs when Jordan was off the court, while the 2018 Cavs outscored the 2018 Raptors with LeBron off the court. And I also presented data regarding the historical playoff SRS of the Daugherty-era Cavs against non-Bulls teams compared to the historical playoff SRS of the pre-Kawhi Raptors against non-Cavs teams. I and others presented data about random variance with regards to three-point shooting and FT shooting in those series. I presented a lot of information on this that all pretty clearly points to the fact that it is silly to draw a conclusion that 2018 LeBron > 1989 Jordan simply because the 1989 Bulls had a tougher time beating the 1989 Cavs than the 2018 Cavs did in beating the 2018 Raptors. A comparison of title odds is just one of those data points. I think the overall picture is pretty inarguable.

As for going “way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff,” I completely agree with you. It was a silly point that someone else led with as a major argument, and my point from the beginning was that the comparison between the two series was being wildly overblown in its importance. Others have simply continued trying to assert that that’s not the case. I’d be happy to stop discussing it, since I think it’s manifestly silly and unimportant, but others seem to think it is a crucially important point in favor of 2018 LeBron, so I’ve continued to address it (albeit somewhat reluctantly and nearing the end of my patience with it).

Eh the stock market point is precisely why using Vegas betting odds doesn't work. Few understand this but what actually moves markets is inventory. People are told and believe that market movements are based on real time reflections of supply and demand when in reality it's more akin to exchanges moving lines to entice the most bets that they can offload inventory (people's orders that need to be matched with opposing orders; aka every buyer needs a seller) into. It's never real time and can never perfectly reflect fundamental value because exchanges themselves exist and need to make money scalping fees off of transactions which inherently introduces variance and erratic fluctuation which is then amplified by the psychological variance of participants in the market.

So in stock terms it seems you think that ultimately the closing price was the one most representative of fair value when in reality it was the mass of orders put in above or below the closing price that actually drove the market to where it ended up. What Cavsfan is trying to explain to you and you're failing to understand is that the lines moved more for LeBron because the market itself hit a psychological level where despite the Cavs being an inferior team there was enough volume betting on LeBron for exchanges to keep accumulating orders while guiding the lines upward and trapping everyone from lower numbers that "shorted the hole". There's absolutely no reason or evidence to believe that such a move is purely representative of the absolute qualitative change in the goodness of both teams. That doesn't even make sense if you understand markets. Influenced by it? Sure. A 1:1 near perfect measure of relative strength like you seem to believe? Hell to the nah

It's an extremely fallacy to think that stocks are more an accurate representation of a company's fundamental worth than they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps and why so many geniuses blew up this year thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines is what you think it is chief.

I think this is getting more in the weeds than is necessary here. Yes, neither betting markets nor stock markets are a truly perfect market. But that doesn’t mean that a highly-bet-on betting market isn’t quite informative regarding what the contemporaneous consensus perception of teams was. Essentially nothing I’ve said is really dependent on the precise numbers or requiring some sort of “1:1 near perfect measure” to make the point. It’s largely based on odds that weren’t even close. They don’t need to be perfect to support my point. I don’t think there’s any reason to think that such significant differences aren’t generally reflective of consensus perception in those years. Of course, perception can be wrong and markets aren’t perfect, but it’s certainly still a relevant data point regarding the strength of different teams (and was also far from the only data point I raised in that discussion).

As for the thing about people essentially betting on LeBron basically out of trust in him, I get the point, but another way of putting that is that that’s the market internalizing that LeBron was a historically great player. And they weren’t wrong! I don’t think you’re identifying actual overvaluing of the Cavs. Rather, you’re just identifying the pretty obvious fact that LeBron was a huge reason their odds were where they were. That’s no surprise! But markets of course internalized Jordan’s greatness too. I think if you wanted to make an argument along these lines, it’d probably be better articulated as that betting markets didn’t fully internalize Jordan’s greatness before he’d won a title. That’s probably true to some degree. But betting markets were fully capable of installing the Bulls as the pre-playoffs title favorite in 1991, before Jordan had won a title (and also as one of the top few favorites in 1990). And Jordan wasn’t really substantially better in 1990 or 1991 than in 1989—these were all peak years for him. So the Bulls’ long odds in 1989 really was more about the team not being good than about Jordan not being trusted until he won a title.

It's an extremely fallacy to think that stocks are more an accurate representation of a company's fundamental worth than they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps and why so many geniuses blew up this year thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines is what you think it is chief.

Yes, of course market cap is simply a representation of what the public thinks a company is worth (and not a perfect representation since markets aren’t perfect). But there’s wisdom in crowds, and so the public’s consensus view of what something is worth is certainly a good indicator of what it actually is worth (and more likely to be correct that just one individual person’s assessment). It is definitely a relevant data point! Similarly, betting odds are simply a representation of how good the public thinks teams are (and not a perfect one since markets aren’t perfect). But there’s wisdom in crowds, and so the public’s consensus view of how good different teams are is certainly a good indicator of how good they are. It’s obviously not perfect, but a data point doesn’t have to be perfect in order to be relevant and persuasive. If it did, then no one could make virtually any argument about anything. Ultimately, I’m not really sure what the objection is here. Are you of the view that title odds in past years don’t tell us anything about the strength of teams? Or are you just arguing that betting odds aren’t a perfect measure? It feels like it’s the latter, and, if so, then that’s just arguing against a straw man—I never suggested anything was a perfect measure (and, more generally, I very often say on these forums that essentially all data we have is flawed—something that this same crew of people has somehow criticized me for), and providing data does not mean that someone is saying that that data is perfect or infallible.

One other aside: For anyone reading this who may need to hear this, Heej is absolutely right that sports betting just takes money from people. Don’t bet on sports. It has negative expected value, and there are many places to put your money that have positive expected value. I am discussing what betting odds can tell us, but would never actually bet on sports because doing that is objectively a bad use of money.

Well it does indeed sound to me like a both/and situation where Cavs as a whole were overrated given the context of their health situation and the Bulls from the top down were underrated. Meanwhile in Cavs-Raptors, the perceived market value is naturally gonna be more towards the middle (maybe Cavs were slightly overrated vs Raps underrated like Cavsfan was saying) due to years of data. But using lines like this to compare and contrast teams is akin to granularly using PER imo.

The 89 Bulls were a budding market inefficiency across the board, 18 Cavs and especially LeBron were known quantities; so you're better off looking at betting lines within ranges in my honest opinion. Tbh just a top down fundamental analysis of the ownership, front office, coaching, health, stats matchups in a series probably gets you a lot closer to real value than market value does.

I think the point is that what LeBron did in that series was more dominant and against a better opponent; and I just wouldn't appeal to wisdom of the crowds to let market value tell me which team was more legit. That's why many are wrong. There must be more losers than winners to pay for it all.

My objection is that I believe the perceived difference in market value between both superstars' teams diverges greatly (for myriad contextual reasons) from real value added. So invoking wisdom of the crowds is fallacious both in this example and in general.

Personally I think this is just one sticking point albeit a relevant one to their careers (LeBron doing more vs better comp). With basketball being a game of exploiting coverage inefficiencies LeBron as a player just mitigates more inefficiencies on both ends; and it's reflected in signals like this where people expect more out of LeBron and he outperforms expectations.

[Heej]

Given that betting is a billion dollar industry and there is nothing people (bettors and companies) want to do more than make money, it's hard for me to believe that betting odds are far off in terms of assessing team quality. If these huge companies set the odds incorrectly, you better believe bettors will exploit that and profit at their expense consistently. So there is STRONG INCENTIVE to set the odds just right!

Companies and betters make money in different ways. Betters make money in a zero sum game, betting companies make money from transactions and supplying liquidity to the betting market (and I'm sure they supply betting data to investors to rig the game further).

Their endgame is facilitating more and more bets. To that end, lines are placed precisely where they'll receive the most transactions and adjusted in real time. There is zero obligation or reasoning to assume lines will gravitate towards true value. In fact, the exchanges are incentivized to move lines away from true value to trap the most losers.

Can they be an approximation? Sure. That's why I say you need to think of them as within confidence intervals from true value realistically and the better assessor of true value imo is a fundamental top down analysis of matchups between ownership groups, front offices, coaching staffs, health, and results.

[lessthanjake]

Finally, referring to the 2019 Cavs as “role player minute swaps” is an interesting (and pretty misleading) way of describing a situation where the team’s remaining best player played virtually zero minutes all season with the 7 other players that had played the most minutes the prior season (as I showed).

Idk, I think it's an accurate way to describe a team of role-players who still look terrible when we cherrypick minutes and games with the one role player who has ever evidenced being anything more...

There's also the cavs with kyrie and love (and no lebron) looking bad by both spot minute samples and actual games...weren't you saying 2016 Lebron's "help" was as talented as the 2016 Steph's?

And then there is the matter of the 6 point defensive drop for cleveland...the 2019 lakers becoming one of the best defenses(and improving to a 50-win pace) until Lebron **** up his groin vs the warriors(and tbf, bi and lonzo got injured shortly after)...

No, no. definitely not noise.

This is basically just flailing around. An argument was made that, despite impact data, LeBron’s defense in 2018 wasn’t actually that bad because the 2019 Cavs defense was worse than the 2018 Cavs defense. I pointed out that the 8 non-LeBron players that had the most minutes for the 2018 Cavaliers played a grand total of just 92 games for the 2019 Cavaliers, so it is just a silly apples to oranges comparison. You responded and said that the team was really bad even in just the Kevin Love minutes. I pointed out Kevin Love spent virtually no minutes in 2019 on the court with the 7 other non-LeBron players that had the most minutes for the 2018 Cavaliers. These facts obviously show it is really silly to attempt to use the 2019 Cavaliers defensive numbers to rehabilitate LeBron’s bad 2018 defense. The teams were completely different. You might as well just compare the 2018 Cavs defense to a random other franchise’s defense in 2019—it wouldn’t be much less relevant. Your response here is to keep trying to make that argument without acknowledging that it has been completely dismantled, while throwing out other things that are not relevant to this question of whether the 2019 Cavs defense is a meaningful data point or to whether LeBron was a bad defender in 2018 specifically (note: no one is saying LeBron was a bad defender in general—both the data and eye test tell us he was a bad defender specifically in that year).

.

And I can’t really have any objection to you not being convinced on LeBron’s 2018 defense, when you’re simultaneously acknowledging that the argument uses the best evidence and reasoning we have.

Tip:

He who must say he is a user the best evidence and reasoning we have is no true user of the best reasoning and evidence we have.

LOL, I was literally quoting Tim Lehrbach’s description. I was responding to a post in which Tim had said “I usually say that we're all just telling stories about the game using the best evidence and reasoning we have, and you've done that.” Stop trying to be insulting just for the sake of it. It’s highly unpleasant.

[Djoker]

Given that betting is a billion dollar industry and there is nothing people (bettors and companies) want to do more than make money, it's hard for me to believe that betting odds are far off in terms of assessing team quality. If these huge companies set the odds incorrectly, you better believe bettors will exploit that and profit at their expense consistently. So there is STRONG INCENTIVE to set the odds just right!

Companies and betters make money in different ways. Betters make money in a zero sum game, betting companies make money from transactions and supplying liquidity to the betting market (and I'm sure they supply betting data to investors to rig the game further).

Their endgame is facilitating more and more bets. To that end, lines are placed precisely where they'll receive the most transactions and adjusted in real time. There is zero obligation or reasoning to assume lines will gravitate towards true value. In fact, the exchanges are incentivized to move lines away from true value to trap the most losers.

Can they be an approximation? Sure. That's why I say you need to think of them as within confidence intervals from true value realistically and the better assessor of true value imo is a fundamental top down analysis of matchups between ownership groups, front offices, coaching staffs, health, and results.

Their goal is facilitate more bets but if they set the lines away from the true value, then bettors will notice that and exploit it. Explain to me a scenario where a betting company would be advantaged to place a line away from true value. It doesn't make sense.

If where millions of people place their money is not an indication of "true value", then what is? It's noteworthy that even in politics, polls are wrong more often than betting markets.

[lessthanjake]

The 89 Bulls were a budding market inefficiency across the board, 18 Cavs and especially LeBron were known quantities; so you're better off looking at betting lines within ranges in my honest opinion. Tbh just a top down fundamental analysis of the ownership, front office, coaching, health, stats matchups in a series probably gets you a lot closer to real value than market value does.

I think the betting odds inherently account for, among other things, “fundamental analysis of the ownership, front office, coaching, health, [and] stats matchup.” In a highly-traded public market, we can generally assume that prices take into account all available information. Indeed, one reason I think pre-playoffs odds are helpful is that it inherently accounts for things like this that people knew about contemporaneously but that we are likely to at least partly forget about now.

I think the point is that what LeBron did in that series was more dominant and against a better opponent; and I just wouldn't appeal to wisdom of the crowds to let market value tell me which team was more legit. That's why many are wrong. There must be more losers than winners to pay for it all.

There doesn’t actually need to be more winners than losers for oddsmakers to make money. As I noted in a response to Tim, what happens is they set the odds in a way that just shortens everything’s odds a little in order to make them money. In the example I gave Tim, I talked about a hypothetical game where the two teams have a legitimate coin-flip 50/50 matchup. The odds for each team would end up implying a slightly higher than 50% chance of winning (for instance, maybe -120 for both teams). So, if you won, you’d get a little less than double your money. If there were equal numbers of bets on both sides of that game (as you might expect in a 50/50 matchup where the odds are actually the same), the oddsmakers would make money, even though there are as many “winners” as there are “losers.” It’s just that the winners win less than the losers lose, and oddsmakers pocket that difference.

Anyways, it’s probably not worth discussing the betting-odds thing specifically much more, since I think I’ve made my point, and if you still don’t agree then we’d probably both just be wasting our time to keep going back and forth on it.

[AEnigma]

I think Kawhi Leonard was actually a bad defender in 2017 because the Spurs were historically good when he was on the bench and over 8 points worse when he was on the court. I also was not visibly impressed by Kawhi — I saw some people say he played bad team defence, and come to think of it, I only rely remember him focusing on his assignment — so therefore he was probably a bad defender. Some people on the Spurs Reddit said he was lazy fighting through screens, and lazy is a bad word which does not apply to good defenders. I remember national media talking about it too. Here is an article from CBS Sports in December of 2016.

The Spurs' defense gives up the most points per 100 possession, is at its worst statistically, when Leonard is on the floor. I know that sounds crazy, but those are the numbers. San Antonio is significantly worse -- 14.8 points per 100 possessions, to be exact -- when he's on the floor. In other words, they go from downright bad on defense to elite when Leonard goes to the bench.

I can hear you all screaming: "This is the problem with analytics! Defensive metrics are flawed! It's about his teammates! This stat means nothing!"

… And yet, the numbers just get worse the deeper you look.

In addition to that really bad overall defensive rating, opponents are shooting 3.8 percent better than their average on the 9.5 shots that Leonard is contesting per game, per NBA.com's tracking data. Furthermore, the defensive rating of each one of the Spurs' eight rotation players gets worse with Leonard on the court, often by double-digits.

And Leonard's personal numbers are not appreciably affected by who he plays with. In fact, he gets better while paired with four of the eight, and only marginally worse with the other four (he is 2.2 points per 100 worse alongside LaMarcus Aldridge, for instance). All this suggests that it is not Leonard's teammates who are dragging him down, but the other way around.

As a viewer, it was sad to see that in one offseason Kawhi could drop 7.5 points of defensive impact. No idea how he managed to finished top three in DPoY voting that year; the numbers are just so obvious that he sucked, and the Spurs barely missed him in 2018. Thankfully I am a very serious person who really understands the game of basketball, so now I can tell people that we all should know how bad Kawhi was on defence in 2017.

[Heej]

Given that betting is a billion dollar industry and there is nothing people (bettors and companies) want to do more than make money, it's hard for me to believe that betting odds are far off in terms of assessing team quality. If these huge companies set the odds incorrectly, you better believe bettors will exploit that and profit at their expense consistently. So there is STRONG INCENTIVE to set the odds just right!

Companies and betters make money in different ways. Betters make money in a zero sum game, betting companies make money from transactions and supplying liquidity to the betting market (and I'm sure they supply betting data to investors to rig the game further).

Their endgame is facilitating more and more bets. To that end, lines are placed precisely where they'll receive the most transactions and adjusted in real time. There is zero obligation or reasoning to assume lines will gravitate towards true value. In fact, the exchanges are incentivized to move lines away from true value to trap the most losers.

Can they be an approximation? Sure. That's why I say you need to think of them as within confidence intervals from true value realistically and the better assessor of true value imo is a fundamental top down analysis of matchups between ownership groups, front offices, coaching staffs, health, and results.

Their goal is facilitate more bets but if they set the lines away from the true value, then bettors will notice that and exploit it. Explain to me a scenario where a betting company would be advantaged to place a line away from true value. It doesn't make sense.

If where millions of people place their money is not an indication of "true value", then what is? It's noteworthy that even in politics, polls are wrong more often than betting markets.

And that's when it self corrects the other way back into the range. At the end of the day if the ultimate outcome of a game was in the 6-10 range, it's absolutely in a company's advantage to get a bunch of people betting in the 11-15 range because those losers pay the winners.

Money is made on the margins, and to that end expansive movement in either direction is most advantageous to get the most participants both playing and on the wrong side of an outcome. All you've done is serve to agree that betting lines should really only be used in ranges and not as absolute value predictors when you're comparing it to wildly inexact sciences like polling.

The 89 Bulls were a budding market inefficiency across the board, 18 Cavs and especially LeBron were known quantities; so you're better off looking at betting lines within ranges in my honest opinion. Tbh just a top down fundamental analysis of the ownership, front office, coaching, health, stats matchups in a series probably gets you a lot closer to real value than market value does.

I think the betting odds inherently account for, among other things, “fundamental analysis of the ownership, front office, coaching, health, [and] stats matchup.” In a highly-traded public market, we can generally assume that prices take into account all available information. Indeed, one reason I think pre-playoffs odds are helpful is that it inherently accounts for things like this that people knew about contemporaneously but that we are likely to at least partly forget about now.

I think the point is that what LeBron did in that series was more dominant and against a better opponent; and I just wouldn't appeal to wisdom of the crowds to let market value tell me which team was more legit. That's why many are wrong. There must be more losers than winners to pay for it all.

There doesn’t actually need to be more winners than losers for oddsmakers to make money. As I noted in a response to Tim, what happens is they set the odds in a way that just shortens everything’s odds a little in order to make them money. In the example I gave Tim, I talked about a hypothetical game where the two teams have a legitimate coin-flip 50/50 matchup. The odds for each team would end up implying a slightly higher than 50% chance of winning (for instance, maybe -120 for both teams). So, if you won, you’d get a little less than double your money. If there were equal numbers of bets on both sides of that game (as you might expect in a 50/50 matchup where the odds are actually the same), the oddsmakers would make money, even though there are as many “winners” as there are “losers.” It’s just that the winners win less than the losers lose, and oddsmakers pocket that difference.

Anyways, it’s probably not worth discussing the betting-odds thing specifically much more, since I think I’ve made my point, and if you still don’t agree then we’d probably both just be wasting our time to keep going back and forth on it.

Because markets are forward looking they're prone to extreme fluctuations, especially when uncertainty is introduced. This all-seeing eye of the market concept you're subscribing to simply doesn't work on a granular level like you're trying to imply. There's too much fluctuation that can occur within a playoff run which is almost an entirely different game from the regular season due to the level of detailed scheming that it creates with more prep time. All it takes is a team stumbling into a pet action like the Warriors did in 2015 short-rolling David Lee vs the Cavs in the Finals to break the series open or a matchup/scheme that works best like LeBron switching onto Tony Parker pick and rolls in 2013 and using Bron as a screener.

Sure you can believe the outcomes are 50/50 like it's supposed to be on paper but in reality the number of winners that actually make it to the end vs the number of losers tacked on before the match ends are always gonna skew towards creating more losers because psychologically people psychologically love hoping losers turn around and fear winners continuing.

I'll leave you with a paraphrase of a legendary Wall Street trader named Larry Williams. "Markets have a path of destiny they tend to follow, with erratic fluctuations along the way." I believe you're making too much of the erratic fluctuations in lines thinking it's the path of destiny. But sure I'm good to drop this, twas a fun sidebar tho as it's been a topic of interest to me as a stonk enthusiast seeing so many of my bball run buddies being heavy into betting. I've seen way too many people blow up thinking the market should represent real life.

Also just doesn't make sense to me to value predictive odds over the actual series results and later data on how the teams developed over time. We know now how big a Mark Price injury is because we saw how the Cavs waned over the years as he waned and have generations of archetypal proof of concept where we can see that his type of shooting, ballhandling, and playmaking guard skillset is the most scalable and repeatable archetype because they scale well with and amplify any other type of player. Pre-playoff odds aren't gonna capture that nearly as well.

[Peregrine01]

Again, I think you guys just went way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff and it ruined your guys' ability to see the thread for what it is. Betting odds have almost everything to do with perception, that's what you don't seem to be fully getting here. Vegas is in the business of making money more than deciding which team is better which is why betting odds fluctuate based on which team gets more bets. So using who Vegas favored is not some infallible tool to bring up in a bb forum. I would think you are old enough to understand this stuff. LeBron's aura of invincibility vs the east in 2018 was on a completely different level than MJ's was back in 1989 when he hadn't even made it out of the 2nd rd yet. Re 89 Bulls roster vs 18 Cavs, it's not really clear which roster was better. If the Cavs was better it’s simply because of having some guys who could hit the 3 around that version of LeBron. The Bulls otoh were built on defense and letting MJ be the offense. At the very least the Bulls team had way more experience having played together which counts for something.

Yes, the fact that betting odds fluctuate based on which team gets more bets is actually precisely the reason they’re informative, because it means the betting odds end up substantially reflecting contemporaneous consensus views of the public regarding the teams (especially in a highly-bet-on betting market such as NBA title odds). It’s akin to a publicly-traded company’s stock market value (i.e. market cap) reflecting the public’s consensus view regarding a company’s value. Of course, the public’s consensus view can be wrong, but it is still a relevant data point here. Notably, though, my argument didn’t just rely on that anyways. For instance, I also presented data showing that the 1989 Bulls got completely destroyed by the Cavs when Jordan was off the court, while the 2018 Cavs outscored the 2018 Raptors with LeBron off the court. And I also presented data regarding the historical playoff SRS of the Daugherty-era Cavs against non-Bulls teams compared to the historical playoff SRS of the pre-Kawhi Raptors against non-Cavs teams. I and others presented data about random variance with regards to three-point shooting and FT shooting in those series. I presented a lot of information on this that all pretty clearly points to the fact that it is silly to draw a conclusion that 2018 LeBron > 1989 Jordan simply because the 1989 Bulls had a tougher time beating the 1989 Cavs than the 2018 Cavs did in beating the 2018 Raptors. A comparison of title odds is just one of those data points. I think the overall picture is pretty inarguable.

As for going “way off the deep end in this thread with the 89 Cavs v 18 Raptors stuff,” I completely agree with you. It was a silly point that someone else led with as a major argument, and my point from the beginning was that the comparison between the two series was being wildly overblown in its importance. Others have simply continued trying to assert that that’s not the case. I’d be happy to stop discussing it, since I think it’s manifestly silly and unimportant, but others seem to think it is a crucially important point in favor of 2018 LeBron, so I’ve continued to address it (albeit somewhat reluctantly and nearing the end of my patience with it).

Eh the stock market point is precisely why using Vegas betting odds doesn't work. Few understand this but what actually moves markets is inventory. People are told and believe that market movements are based on real time reflections of supply and demand when in reality it's more akin to exchanges moving lines to entice the most bets that they can offload inventory (people's orders that need to be matched with opposing orders; aka every buyer needs a seller) into. It's never real time and can never perfectly reflect fundamental value because exchanges themselves exist and need to make money scalping fees off of transactions which inherently introduces variance and erratic fluctuation which is then amplified by the psychological variance of participants in the market.

So in stock terms it seems you think that ultimately the closing price was the one most representative of fair value when in reality it was the mass of orders put in above or below the closing price that actually drove the market to where it ended up. What Cavsfan is trying to explain to you and you're failing to understand is that the lines moved more for LeBron because the market itself hit a psychological level where despite the Cavs being an inferior team there was enough volume betting on LeBron for exchanges to keep accumulating orders while guiding the lines upward and trapping everyone from lower numbers that "shorted the hole". There's absolutely no reason or evidence to believe that such a move is purely representative of the absolute qualitative change in the goodness of both teams. That doesn't even make sense if you understand markets. Influenced by it? Sure. A 1:1 near perfect measure of relative strength like you seem to believe? Hell to the nah

It's also an extreme logical fallacy to think that stocks are more an accurate representation of a company's fundamental worth than that they are simply a representation of what other people think a company is worth. Surely there must be something more systematically reliable than letting Vegas tell you what to think LOL. They're literally in the business of taking money from average chumps . This same deification of the all-knowing perfectly up to date eye of the market is why so many geniuses blew up these past 2 years thinking they were getting out ahead of everyone and shorting the upcoming recession. Markets may be influenced by fundamental value but they will never for as long as humans participate in them accurately reflect true intrinsic fundamental value in real time. But also I will say, in the long run markets and human psychology can never outrun fundamentals; much like there was nothing the Raptors could do to escape the massacre of LeBronto. Just don't think using betting lines informs you of what you think it does chief.

Not to get even more off track here, but there is a massive difference in the stock markets and the betting markets in that stocks are traded constantly while bets aren't. Stocks are often bought with the anticipation of selling it for a quick profit which makes prices even more detached from a company's fundamentals or even market participants' perception of fundamentals. At least in the short-term, stock prices are a reflection of what people think the stock price does and not what the underlying company does. Bets are a more accurate reflection of what bettors' perception of probabilities are.

[lessthanjake]

Companies and betters make money in different ways. Betters make money in a zero sum game, betting companies make money from transactions and supplying liquidity to the betting market (and I'm sure they supply betting data to investors to rig the game further).

Their endgame is facilitating more and more bets. To that end, lines are placed precisely where they'll receive the most transactions and adjusted in real time. There is zero obligation or reasoning to assume lines will gravitate towards true value. In fact, the exchanges are incentivized to move lines away from true value to trap the most losers.

Can they be an approximation? Sure. That's why I say you need to think of them as within confidence intervals from true value realistically and the better assessor of true value imo is a fundamental top down analysis of matchups between ownership groups, front offices, coaching staffs, health, and results.

Their goal is facilitate more bets but if they set the lines away from the true value, then bettors will notice that and exploit it. Explain to me a scenario where a betting company would be advantaged to place a line away from true value. It doesn't make sense.

If where millions of people place their money is not an indication of "true value", then what is? It's noteworthy that even in politics, polls are wrong more often than betting markets.

And that's when it self corrects the other way back into the range. At the end of the day if the ultimate outcome of a game was in the 6-10 range, it's absolutely in a company's advantage to get a bunch of people betting in the 11-15 range because those losers pay the winners.

Money is made on the margins, and to that end expansive movement in either direction is most advantageous to get the most participants both playing and on the wrong side of an outcome. All you've done is serve to agree that betting lines should really only be used in ranges and not as absolute value predictors when you're comparing it to wildly inexact sciences like polling.

As I said, I don’t think it’s worth either of us discussing this much more at this point (though I agree it’s been a fun sidebar), but one thing I do want to point out again is that the point I was making applies even if you merely think “betting lines should really only be used in ranges.” The difference in odds between the 1989 Bulls and 1989 Cavs is just completely different from the difference in odds between the 2018 Cavs and 2018 Raptors (for instance, the 1989 Cavs had +400 pre-playoff title odds, compared to +2000 for the 1989 Bulls, while the 2018 Raptors had +1075 pre-playoffs title odds, compared to +800 for the 2018 Cavs). My point didn’t hinge at all on minor differences in the relative odds. To use the polling analogy you refer to, this isn’t like saying a candidate is ahead because a poll has them ahead by 3%. It’s like saying a candidate is ahead because a poll has them ahead by 25%. You can think it’s an inexact science and still say that this is beyond the realm of that being the explanation. I think what we are looking at is a genuinely real and significant difference in how these teams were contemporaneously perceived relative to each other. One can of course argue that perception doesn’t equal reality (and I’ve provided other data that isn’t rooted in perception), but I think it’d be tough to argue that betting odds are so inexact that this sort of difference isn’t reflective of a real difference in perception.