Your Unpopular Basketball Opinions? [PC Board Edition]

[bondom34]

No, I am correct. You are wrong. The definition of the stat is included already in my analysis.

On/off or +/- is a metric that looks at how teams perform with a certain player on the court, how they perform with a certain player off the court, and calculates the overall impact that player has on team success.

And my answer to this flawed metric is there are FOUR other players on the court when the player is ON THE COURT, and FIVE other players on the court when that player is OFF THE COURT. Actually, check that, it's actually more than that. If you count the other team's players, it's NINE other players on the court when the player is ON THE COURT, and TEN other player on the court when that player is OFF THE COURT.

There is so much variability regarding players and events on the court that the on/off metric is almost entirely useless.

A gave the example of 2-player sports. If a doubles tennis Player A plays 10 games (or minutes, or whatever) alongside Player B, and then another 10 games alongside Player C, and all of those games were played against Players D & E, you may be able to measure the impact of Player B versus Player C based on the success they had playing with Player A. In this context, the impact statistic may hold some degree of weight.

You don't know that what happens when Player X is ON THE COURT is due whatsoever to the presence of Player X, because there are NINE OTHER PLAYERS who also impact that statistic. You also don't know that what happens when Player X is OFF THE COURT is due whatsoever to the absence of Player X, because there are TEN OTHER PLAYERS who impact that statistic.

I stand by my comment. It is a useless metric.

On/off and plus/minus are separate and distinct statistics.

On/off compares how a player's team performs when he's on court vs. off court.

Plus/minus looks at how a player's team performs vs the opponent only when he's on court.

[THKNKG]

No, I am correct. You are wrong. The definition of the stat is included already in my analysis.

On/off or +/- is a metric that looks at how teams perform with a certain player on the court, how they perform with a certain player off the court, and calculates the overall impact that player has on team success.

And my answer to this flawed metric is there are FOUR other players on the court when the player is ON THE COURT, and FIVE other players on the court when that player is OFF THE COURT. Actually, check that, it's actually more than that. If you count the other team's players, it's NINE other players on the court when the player is ON THE COURT, and TEN other player on the court when that player is OFF THE COURT.

There is so much variability regarding players and events on the court that the on/off metric is almost entirely useless.

A gave the example of 2-player sports. If a doubles tennis Player A plays 10 games (or minutes, or whatever) alongside Player B, and then another 10 games alongside Player C, and all of those games were played against Players D & E, you may be able to measure the impact of Player B versus Player C based on the success they had playing with Player A. In this context, the impact statistic may hold some degree of weight.

You don't know that what happens when Player X is ON THE COURT is due whatsoever to the presence of Player X, because there are NINE OTHER PLAYERS who also impact that statistic. You also don't know that what happens when Player X is OFF THE COURT is due whatsoever to the absence of Player X, because there are TEN OTHER PLAYERS who impact that statistic.

I stand by my comment. It is a useless metric.

It's useless just like field goal percentage outside of 3 feet right? Both are single stats that, if used in isolated circumstances to prove a point, are useless.

However, you're talking purely about raw +/-. You're ignoring all of the various other stats in the +/- family (that make efforts to deal with the idea of multicolinearity), not to mention that raw plus minus even has value. Statistics are ways for us to see tangible realities in numerical expression. They are always context dependent, because the idea of statistics is that they are intrinsically "other"-dependent.

But what do I know, keep giving 3+ ft fg% stats while giving a reductionist definition of plus minus.

The other guy is the only one who has to be consistent, right?

[Goudelock]

No, I am correct. You are wrong. The definition of the on/off +/- stat was wholly assumed and subsumed within my analysis.

On/off or +/- is a metric that looks at how teams perform with a certain player on the court, how they perform with a certain player off the court, and calculates the overall impact that player has on team success.

And my answer to this flawed metric is there are FOUR other players on the court when the player is ON THE COURT, and FIVE other players on the court when that player is OFF THE COURT. Actually, check that, it's actually more than that. If you count the other team's players, it's NINE other players on the court when the player is ON THE COURT, and TEN other player on the court when that player is OFF THE COURT.

There is so much variability regarding players and events on the court that the on/off metric is almost entirely useless.

A gave the example of 2-player sports. If a doubles tennis Player A plays 10 games (or minutes, or whatever) alongside Player B, and then another 10 games alongside Player C, and all of those games were played against Players D & E, you may be able to measure the impact of Player B versus Player C based on the success they had playing with Player A. In this context, the impact statistic may hold some degree of weight.

You don't know that what happens when Player X is ON THE COURT is due whatsoever to the presence of Player X, because there are NINE OTHER PLAYERS and lots of random events which also impact that statistic. You also don't know that what happens when Player X is OFF THE COURT is due whatsoever to the absence of Player X, because there are TEN OTHER PLAYERS and lots of random events which also impact that statistic.

I stand by my comment. It is a useless metric.

While I appreciate your thorough post, I'd like to give you some advice. It's probably best not to start your posts with things like "I am correct. You are wrong". It sounds very rude and condescending, and isn't necessary.

[ardee]

Garnett is a top 15 player at best.

Kobe is very clearly a top 10 player.

Nash is the 3rd best PG of all time.

Nowitzki is the second best PF of all time.

Wilt ranks higher on the GOAT list than anyone except Russell and Jordan.

[THKNKG]

Garnett is a top 15 player at best.

Kobe is very clearly a top 10 player.

Nash is the 3rd best PG of all time.

Nowitzki is the second best PF of all time.

Wilt ranks higher on the GOAT list than anyone except Russell and Jordan.

This is a solid list, because I disagree with 2 points, agree with 2 points, and am undecided on 1.

[ardee]

Garnett is a top 15 player at best.

Kobe is very clearly a top 10 player.

Nash is the 3rd best PG of all time.

Nowitzki is the second best PF of all time.

Wilt ranks higher on the GOAT list than anyone except Russell and Jordan.

This is a solid list, because I disagree with 2 points, agree with 2 points, and am undecided on 1.

Which are which?

[GeneralManager]

No, I am correct. You are wrong. The definition of the stat is included already in my analysis.

On/off or +/- is a metric that looks at how teams perform with a certain player on the court, how they perform with a certain player off the court, and calculates the overall impact that player has on team success.

And my answer to this flawed metric is there are FOUR other players on the court when the player is ON THE COURT, and FIVE other players on the court when that player is OFF THE COURT. Actually, check that, it's actually more than that. If you count the other team's players, it's NINE other players on the court when the player is ON THE COURT, and TEN other player on the court when that player is OFF THE COURT.

There is so much variability regarding players and events on the court that the on/off metric is almost entirely useless.

A gave the example of 2-player sports. If a doubles tennis Player A plays 10 games (or minutes, or whatever) alongside Player B, and then another 10 games alongside Player C, and all of those games were played against Players D & E, you may be able to measure the impact of Player B versus Player C based on the success they had playing with Player A. In this context, the impact statistic may hold some degree of weight.

You don't know that what happens when Player X is ON THE COURT is due whatsoever to the presence of Player X, because there are NINE OTHER PLAYERS who also impact that statistic. You also don't know that what happens when Player X is OFF THE COURT is due whatsoever to the absence of Player X, because there are TEN OTHER PLAYERS who impact that statistic.

I stand by my comment. It is a useless metric.

On/off and plus/minus are separate and distinct statistics.

On/off compares how a player's team performs when he's on court vs. off court.

Plus/minus looks at how a player's team performs vs the opponent only when he's on court.

I am aware of the difference.

My analysis applies to both metrics. There are 10 other players plus other random variables and noise that simply cannot make the metric useful whatsoever. This applies to both metrics.

[THKNKG]

Garnett is a top 15 player at best.

Kobe is very clearly a top 10 player.

Nash is the 3rd best PG of all time.

Nowitzki is the second best PF of all time.

Wilt ranks higher on the GOAT list than anyone except Russell and Jordan.

This is a solid list, because I disagree with 2 points, agree with 2 points, and am undecided on 1.

Which are which?

Disagree with the first two, agree with 3/4, undecided on the last one

[GeneralManager]

It's useless just like field goal percentage outside of 3 feet right? Both are single stats that, if used in isolated circumstances to prove a point, are useless.

However, you're talking purely about raw +/-. You're ignoring all of the various other stats in the +/- family (that make efforts to deal with the idea of multicolinearity), not to mention that raw plus minus even has value. Statistics are ways for us to see tangible realities in numerical expression. They are always context dependent, because the idea of statistics is that they are intrinsically "other"-dependent.

But what do I know, keep giving 3+ ft fg% stats while giving a reductionist definition of plus minus.

The other guy is the only one who has to be consistent, right?

Really?

You believe that +/- is as equally intrinsically valuable as FG%???

One measures ONE SINGULAR PLAYER'S ability to shoot. Sure, some take better shots than others, but if a player takes a bad shot isn't that at least in part the fault of the shot-taker? Bottom line, while not a tell-all statistic, this is a measurement regarding ONE SINGULAR PLAYER.

The other measures an amorphous ephemeral stat that is FULLY DEPENDENT ON 10 PLAYERS SIMULTANEOUSLY.

Um, excuse me, the stat that is 10-player dependent is so much less reliable as to be nearly irrelevant.

Forgive me for being argumentative, I try to keep things lively and entertaining. Nevertheless, I strongly believe I am correct.

[Goudelock]

Really?

You believe that +/- is as equally intrinsically valuable as FG%???

One measures ONE SINGULAR PLAYER'S ability to shoot. Sure, some take better shots than others, but if a player takes a bad shot isn't that at least in part the fault of the shot-taker? Bottom line, while not a tell-all statistic, this is a measurement regarding ONE SINGULAR PLAYER.

The other measures an amorphous ephemeral stat that is FULLY DEPENDENT ON 10 PLAYERS SIMULTANEOUSLY.

Um, excuse me, the one that is 10-player dependent is so much less reliable as to be nearly irrelevant.

A counterargument to this is that FG% measures accuracy, but doesn't provide context to what lead up to each individual shot. What kind of shot was being taken and when? How contested was it? What were the circumstances of the shot? Was it early in the shot clock, or was it a desperation shot? Was the shot open, or was it contested because the player's teammates were not spaced properly. Who was guarding the player? Is the best defensive player on the other team guarding the player as he shoots, or is it some scrub? What distance is it being taken at, and why? Is it off a cut, a layup off the dribble, a shot off the catch? Were these things affected by the skills of other players on the floor?

[rebirthoftheM]

Bill Russell is not a top 15 player of all time

This has to be the ultimate unpopular opinion. What's your reasoning for holding such an unpopular opinion?

I put a very strong premium on offense and I tend to not rate non-wilt players from that era.

[GeneralManager]

Didn't you make an entire account to troll LeBron on a basketball discussion board? That's real dedication.

Wrong.

But I knew my statement, while entirely true, would be unpopular.

I bring balance here to an otherwise over-millenial'd board.

Pointing out the man cannot shoot outside 3 feet is not trolling.

One
As I've pointed out previously, Lebron collectively for nearly his whole career ('05-present) has been well above average from 3-10 feet, including the single-best efficiency that I've yet found from that range. I know, I know....you're going to say you're not saying "bad" compared to the average player, but "bad" compared to top 10 players. But actually, that isn't true either.

Compared to top 10(ish) players, he more or less holds his own in this range (for reference, Lebron collectively in the rs----from '05-present, so I can't be accused of cherry-picking years----was 43.6% from 3-10 ft)........
You'd posted some numbers for Michael Jordan (I think it was like a 3 or 5 year sample??, and I didn't question the source) as 50% from 3-10 feet......so he's clearly better than LEbron.
I don't have any shooting stats for Larry Bird, but I'm going to go out on a limb and say he's better, likely in a similar range as Jordan from that range during his prime. Ditto Kareem; with that sky-hook, I'm just going to assume he's significantly better in this range (probably the GOAT, actually).
Kobe (I don't have him in my top 10, but some do) from '01-'13 was 45.2% from this range....so he's just a slightly better.
I have no numbers, but I'll go out on a limb and suggest Magic is maybe a little better too (probably in similar range as Kobe); though this is unproven. EDIT: Kevin Garnett '01-'12 was 44.6% from this range (so, slightly better).

And that's basically it for top 10(ish) players who are clearly better from this range. Tim Duncan from '01-'13 was 44.0% from 3-10 ft. I suspect it's unlikely that Hakeem was fantastically better from this range, considering that outside of his rookie season (53.8% FG's), he was <53% FG's every year of his career (51.4% from '86-'96).....so unless his finishing % from less than 3 ft is actually really pedestrian, it's reasonably safe to assume that he wasn't any higher than maybe 45% from 3-10 ft (was 37.9% from this range in his final two seasons, for whatever that's worth).

Shaquille O'Neal was 42.5% ('01-'05) from 3-10 ft; so Lebron appears at least equal if not marginally better that Shaq in this range. Based on watching film, total FG% and assumed finishing % at the rim, etc, it's a reasonable assumption that Wilt's conversion rate in the 3-10 ft range was no better.

And based on watching some film, his total FG% and assumed finishing %'s at the rim, it's a VERY easy assumption that Bill Russell was significantly worse from this range.

So actually still not bad, even compared to top 10 company; just middle of the pack here.


Two
I've asked previously (and not received a reply) if you hold various aspects of other top 10(ish) players to similar standards, or if this is simply a double-standard you hold to Lebron alone.
Example: Magic was a mediocre to poor defender. And not mediocre to poor compared to his top 10 competition; mediocre to poor relative to a league-average guard. But this is apparently totally cool, yes?


Three
For the sake of argument, let's just assume the hyperbolic "cannot shoot" statement were true.....I'd previously asked (and again not received an answer) why it matters when he nonetheless manages to be an all-time great level scorer?

For example, let's make a statistical comparison of him to Larry Bird (whom I think we can all agree was an amazing shooter from all ranges) as a scorer (using per 100 possession numbers btw).....
Larry Bird '86-'88 (best 3-year span)
34.4 pts @ 60.0% TS, 3.8 tov in rs
29.5 pts @ 57.6% TS. 3.3 tov in playoffs

Lebron James '12-'14
37.8 pts @ 63.3% TS, 4.6 tov in rs
37.4 pts @ 60.4% TS, 4.3 tov in playoffs

Larry Bird '82-'88 (solidly in prime, 7 years)
32.2 pts @ 58.0% TS, 3.8 tov in rs
29.2 pts @ 56.2% TS, 3.6 tov in playoffs

Lebron James '06-present (solidly in prime, 12 years)
37.7 pts @ 59.4% TS, 4.7 tov in rs
36.4 pts @ 56.8% TS, 4.6 tov in playoffs


I mean, this "inability" to shoot would really only effect his game as a scorer......and yet he's clearly managed to be a better scorer [and for a longer period of time] than Larry Bird (a fantastic shooter). So I ask again (not that I really expect a genuine and non-convoluted response), why does it matter?

Thanks for the well thought out comment. I'm going to sleep soon. I will get back to you with a full response.

My short response is that there are other posters here (perhaps not yourself) that make it seem like shooting outside 3 feet is wholly irrelevant to a player's evaluation. I'm sorry, shooting outside 3 feet has all sorts of reality based impacts, not the least of which is the ball not going through the hoop.

Yeah, shooting outside 3 feet is quite relevant, especially in the demanding context of all-time greatness.

I did say why it matters in a previous comment comparing Magic with LeBron. Perhaps you didn't catch it, that will be a part of my response.

I will get back to you on a point-by-point basis.

[Footnote: I wish we had shot location data on Bird and Magic. Someday we will, and that data will likely benefit my argument.]

[bondom34]

No, I am correct. You are wrong. The definition of the stat is included already in my analysis.

On/off or +/- is a metric that looks at how teams perform with a certain player on the court, how they perform with a certain player off the court, and calculates the overall impact that player has on team success.

And my answer to this flawed metric is there are FOUR other players on the court when the player is ON THE COURT, and FIVE other players on the court when that player is OFF THE COURT. Actually, check that, it's actually more than that. If you count the other team's players, it's NINE other players on the court when the player is ON THE COURT, and TEN other player on the court when that player is OFF THE COURT.

There is so much variability regarding players and events on the court that the on/off metric is almost entirely useless.

A gave the example of 2-player sports. If a doubles tennis Player A plays 10 games (or minutes, or whatever) alongside Player B, and then another 10 games alongside Player C, and all of those games were played against Players D & E, you may be able to measure the impact of Player B versus Player C based on the success they had playing with Player A. In this context, the impact statistic may hold some degree of weight.

You don't know that what happens when Player X is ON THE COURT is due whatsoever to the presence of Player X, because there are NINE OTHER PLAYERS who also impact that statistic. You also don't know that what happens when Player X is OFF THE COURT is due whatsoever to the absence of Player X, because there are TEN OTHER PLAYERS who impact that statistic.

I stand by my comment. It is a useless metric.

On/off and plus/minus are separate and distinct statistics.

On/off compares how a player's team performs when he's on court vs. off court.

Plus/minus looks at how a player's team performs vs the opponent only when he's on court.

I am aware of the difference.

My analysis applies to both metrics. There are 10 other players plus other random variables and noise that simply cannot make the metric useful whatsoever. This applies to both metrics.

Also, on/off and plus/minus don't calculate anything, at least more than a stat like FG%.

As well, we can get some idea of a player's impact using these metrics. If you're dismissing them entirely, you're greatly undervaluing them.

[Prez]

I think James Harden is pretty overrated, and I think given the personnel and system the Rockets employ, they would still be pretty damn solid if you replaced him with any solid starting PG.

Which leads to my second opinion, I think Westbrook is clearly MVP and I see no argument for Harden being better.

I think Kyrie Irving gets a massive boost for simply playing alongside arguably the best player ever, who is not only that good but also stylistically the perfect fit to compensate for all of Kyrie's weaknesses as a floor general/playmaker, leader, team defender, etc. I think if Kyrie had to be the best player on a team like other upper level PGs (Wall, Westbrook, etc) he'd be totally irrelevant, and all the clutch moments he has wouldn't be used as arguments for him because they wouldn't even be in position to have those clutch playoff moments.

And prime for prime I'd take LeBron over Jordan.

[GeneralManager]

On/off and plus/minus are separate and distinct statistics.

On/off compares how a player's team performs when he's on court vs. off court.

Plus/minus looks at how a player's team performs vs the opponent only when he's on court.

I am aware of the difference.

My analysis applies to both metrics. There are 10 other players plus other random variables and noise that simply cannot make the metric useful whatsoever. This applies to both metrics.

Also, on/off and plus/minus don't calculate anything, at least more than a stat like FG%.

As well, we can get some idea of a player's impact using these metrics. If you're dismissing them entirely, you're greatly undervaluing them.

Okay, perhaps careless usage of the word "calculate." Are we really going to play semantics? Come on bondom. If I use the word "evaluate" does it change the essential point? I'm pretty careful with words, but doesn't semantic debates waste valuable space and redirect the debate to useless things?

"Some idea?" "Some idea?" There are people on this board who use "IMPACT" statistics for their MOST IMPORTANT METRIC of players. It is befuddling and egregious. How about this...the way people use this stat on this board distorts their player rankings far more than I undervalue the metric.

NINE OTHER PLAYERS ARE ON THE COURT! Other relevant distortions: bench players, game strategy, coaching decisions, streaky shooting, momentum and "runs" that take minutes or entire quarters. Any serious analysis would literally have to evaluate EACH OF THE NINE PLAYERS individually, their relationship to their teammates, and their relationship to the opponent's matchups. There are so many other variables you could discuss the "why" on this stat for hours and hours and hours.

Anyone that tells me they can measure one player's impact by measuring a total of nine or ten players at a time are not persuading me of anything.

Sorry, I ain't buying these impact stats at all whatsoever.

[bondom34]

I am aware of the difference.

My analysis applies to both metrics. There are 10 other players plus other random variables and noise that simply cannot make the metric useful whatsoever. This applies to both metrics.

Also, on/off and plus/minus don't calculate anything, at least more than a stat like FG%.

As well, we can get some idea of a player's impact using these metrics. If you're dismissing them entirely, you're greatly undervaluing them.

Okay, perhaps careless usage of the word "calculate." Are we really going to play semantics? Come on bondom. If I use the word "evaluate" does it change the essential point? I'm pretty careful with words, but doesn't semantic debates waste valuable space and redirect the debate to useless things?

"Some idea?" "Some idea?" There are people on this board who use "IMPACT" statistics for their MOST IMPORTANT METRIC of players. It is befuddling and egregious. How about this...the way people use this stat on this board distorts their player rankings far more than I undervalue the metric.

NINE OTHER PLAYERS ARE ON THE COURT! Other relevant distortions: bench players, game strategy, coaching decisions, streaky shooting, momentum and "runs" that take minutes or entire quarters.

Anyone that tells me they can measure one player's impact by measuring a total of nine or ten players at a time are not persuading me of anything.

Nobody would claim these as the be all end all in their raw forms. Adjusted plus/minus can be much more useful, but raw numbers help.

[bondom34]

I think James Harden is pretty overrated, and I think given the personnel and system the Rockets employ, they would still be pretty damn solid if you replaced him with any solid starting PG.

Which leads to my second opinion, I think Westbrook is clearly MVP and I see no argument for Harden being better.

I think Kyrie Irving gets a massive boost for simply playing alongside arguably the best player ever, who is not only that good but also stylistically the perfect fit to compensate for all of Kyrie's weaknesses as a floor general/playmaker, leader, team defender, etc. I think if Kyrie had to be the best player on a team like other upper level PGs (Wall, Westbrook, etc) he'd be totally irrelevant, and all the clutch moments he has wouldn't be used as arguments for him because they wouldn't even be in position to have those clutch playoff moments.

And prime for prime I'd take LeBron over Jordan.

Actually to spin off of point 1 here (which btw agree with), I feel TS is vastly overvalued as a measure of offense as a whole.

[GeneralManager]

By the way, I love all you guys. Love the debate.

[GeneralManager]

And prime for prime I'd take LeBron over Jordan.

It's apparent you think shooting outside 3 feet is wholly irrelevant. Case in point, trex.

You should push back on these people more than pushing back on me...the overrating of LeBron is astonishing.

[Prez]

And prime for prime I'd take LeBron over Jordan.

It's apparent you think shooting outside 3 feet is wholly irrelevant. Case in point, trex.

You should push back on these people more than pushing back on me...the overrating of LeBron is astonishing.

It's actually not apparent at all, considering I never said that. If you're gonna use my post to prove a point somewhere else, maybe don't strawman.