RealGM Top 100 List #57

[ronnymac2]

My counterargument to your 'stats don't back him' point is that they do - it's just raw stats can be as informative as PER/WS/APM. In this case he is a 25-28ppg scorer on one of the best offenses in the league at his best, and he passes the logic test to me as someone who's helping his team get open shots by attracting defensive pressure

I buy this.

BTW, Dana Barros had a really good season. lol

[ronnymac2]

Vote: Bernard King

Nominate: Vince Carter

Honestly though, I'm glad Marques is going to get in. I can't find my description of him now, but I think he was essentially Shawn Marion with lesser rebounding numbers and defense but legit attacker handles and the playmaking ability that comes with attacking handles (so, below Hill/Pippen playmaking, but solid nonetheless).

Adrian Dantley belongs in this project.

[Doctor MJ]

The thing about pace adjustment is that there's a long history of it screwing up PER and WS stats to me. Like, I know Steve Nash should have much higher offensive win shares in 2005 than Marbury or Billups. I know that Billups dropping from 23.6 PER to 18.8 PER in Denver despite it being clear that he was the same guy, is wrong. I know that in 09 the gap between Kobe and Wade in PER being 24.4 vs 30.4 is wrong. I know that Magic's WS numbers are a complete disaster for the validity of the stat, his best WS season before 87 ranks #237 all time and below seasons by the likes of Hardaway, Schrempf, Arenas, Brand, Billups. His best OWS season period ranks 49th and his best before '87 is 85 ranking 193rd, below Schrempf, Brand, Bailey Howell, Chet Walker, Marbury, Rudy T, Dana Barros. Steve Nash ranks 179th in 2005 and 137th in 2006. Alex English does not have a top 250 OWS season and Kevin Johnson's only one is in 97 (and any stat that says KJ's best offensive season is 97 clearly has something wrong)

I don't understand how you blame these issues on pace adjustment.

Nash & Magic are underrated by PER and WS because the ability to play floor general at genius level is severely underrated by box score stats. It has nothing to do with pace really, and even if it did, using the fact that Role X gets underrated by a stat to postulate that a player in a completely different role might get massively underrated to is wild speculation without any basis.

Also, Magic's PER & WS numbers go up when he starts scoring more. There's nothing really mysterious about that.

That said, pace adjustment is central to Billups' situation. I don't personally see that as particularly damning (as I said, if a guy is being used less per possession than before, that means more of the onus is on his teammates even if he still has the capability to take on the larger usage he had done before), but it's a valid point on your end...

I just see a clear pattern of it pace messing up the numbers as much as it helps. I believe in the case of Monta Ellis and Kiki Vandeweghe, it's a good thing pace is there to downgrade the numbers, and for a guy like Dirk, that he gets his numbers raised by the slow pace. But there's just as many cases with the Magic and Nash's where it clearly messes up the numbers. I believe there is far too many case by case variance to just wrap a pace adjustment blanket over all the players and expect reliable results, some players it's going to help their stats, some none at all

...but then you lose me again. So it's wrong to knock English based on pace, but it's fine to knock Kiki whose scoring as English's teammate peaked at the same level as English? I mean could it be any more clear that you're starting with conclusions elsewhere, and then taking only the statistics that agree with your opinion?

(And again, Magic & Nash's stats are in now way about pace. The far bigger issue is that BBIQ is underrated by box score stats)

As for APM, I like the stat but when I see Scola, Noah, Haslem, Bowen, Redick, Matthews ranking from mediocore to bad, let alone clear stars like Melo, Deron, Amare, Rose, Joe Johnson, Ben Wallace, Rondo, I can accept Melo's number much easier. I have next to 0 doubt about just about everyone on that list, guys like Scola and Bowen are precisely the type that should do well in advanced stats - and I only need a few to confirm that APM can and has been wrong. My general take on APM is when players are way up on the list like a Manu or KG or players are way down like a Jeff Green or Bargnani, they're probably onto something. Everything in between looks far more random to me. And in Melo's case having him rank top 20 in offensive APM and just having most of it wiped out by defensive stats means as much to me as his overall number anyways, I feel like I know exactly what Melo is defensively, you can fit him on a just fine defensive team due to his athleticism, he's not going to be standout, but like most perimeter players as long as they don't completley miss what's happening on defense like George Gervin or Michael Redd it's not going to set you back or prevent you from playing good defense, which the Nuggets have done.

In general with APM, I'm quite reluctant to make strong statements about players unless they are supposed to be stars and I have a lot of data to work with. The reality is that if you aren't the player the team builds around, then your impact is at the mercy of a good amount of things beyond your control. And of course, small sample size is a problem.

The reason imho why there's reason to think APM works as a judge of stars is simply that stars in basketball have such HUGE impact. I mean, LeBron leaves Cleveland, they get 40 games worse in an 82 game season. In baseball, even the best players are only adding about 10 wins in a 162 game season. When impact is that huge, there's every reason to believe that over a multi-year span we can see who truly is and is not having top tier impact. And of course, if you look at the leaders on multi-year lists, they tend to look really damn good.

So when we see who is supposed to be on that level be instead way down with the common people year after year, to me that is pretty definitive about a player's actual imapct.

Now to be clear: This does not necessarily mean that the player in question isn't capable of much greater impact. It's entirely plausible that Karl has simply refused to tailor the Denver offense to Melo as much as he could and that in a new setting, will see some huge impact from Melo. But if Melo was having massive impact in Denver, I think we would have seen some statistical evidence of this after 8 years...

Re: "know what Melo is defensively". Right, I understand your rationale. We've got some proof he's good at offense, and you believe in the right setting, his defensive woes could be neutralized, and I suppose my main thing is again that you should really be careful about doing so much extrapolation about what a player truly is as opposed to he actually did.

I'd also note though that it's not like Karl has a rep for being an entirely inept defensive coach, right? Seattle got to the finals on a team with an elite defense, and in general as you've noted the Denver defenses Melo's played on have been solid if not spectacular. Bottom line, if there was a really stupid easy way to make Melo not be one of the worst sieves in the entire league, I'm sure Karl would have implemented it.

My counterargument to your 'stats don't back him' point is that they do - it's just raw stats can be as informative as PER/WS/APM. In this case he is a 25-28ppg scorer on one of the best offenses in the league at his best, and he passes the logic test to me as someone who's helping his team get open shots by attracting defensive pressure

Well, that's coherent at least. Clearly we just see this stuff differently. When we have 3 extremely different advanced stats all saying the exact same thing ("Melo is not that impressive"), and we've now seen the player leave the team and the results agree with the advanced stats, and the coach has talked about the joy of coaching a team without said player, I don't see how all of this can be ignored.

[Dr Positivity]

I'm fairly certain Magic and Nash's lower than they should numbers are related to pace and not the lack of value of assists, just because we saw Paul put up the 21, 11 statline and had 18 WS 30 PER level. Hard to imagine that 18, 12, 6 or 18 and 11 at .62-.64 TS% wouldn't be enough for numbers nearly as high on a slower team

Re: Kiki - I admit it just doesn't seem like he's that good of a player like a Rashard Lewis type and his big stats have stood out as probably inflated, while I buy English's ability is responsible for his - but I'll admit that may be giving Kiki a short stick

[Doctor MJ]

Alright, my vote for this time (and next time barring some good arguments):

Vote: Paul Arizin

Yeah I said it, and hoping to see some good counterarguments next time around. Here's the thing, Arizin tends to fall from my radar because any RPOY-style project is bound to have Dolph Schayes ranked ahead of him, and Schayes doesn't impress me that much. However, the more I think about Arizin, the more I find him hard to brush aside.

Y'all know what's been said about him: This is a guy with strong shooting ability who is completely comfortable with how modern shooting needs to be done, while at the same time having a nose for the hole, a fantastic reputation as a defender, and strong evidence that teams got a lot better with him as their alpha even in comparison to another star with comparable box score attributes.

Now consider as well Arizin vs Pettit. Pettit being someone most of us agree was a superb player by any era's standards. One of the things about Pettit that I like is that even though his efficiency was weak by modern standards, it showed signs of improving as the league developed, giving me confidence that while he'd never be Barkley's equal on that front, he'd be solid there.

However, despite the fact that Pettit came into the league AFTER Arizin, his efficiency never matched Arizin's peak efficiency as a volume scorer, and whereas Pettit's playoff stats showed the typical falloff for a player facing touch competition, Arizin only got more impressive in the playoffs.

None of this makes me debate who should be ranked higher between Pettit & Arizin, but truly a player with peak that rivals Pettit is tough to pass up down this far in the list.

btw, none of this changes how I feel about Cousy & Schayes. I think Cousy had the least impactful peak of the bunch at the time, and I have major difficulties taking a 6'7" set shooting big man seriously compared to current players. But I don't throw out the baby with the bathwater when it comes to '50s basketball. A guy with huge impact then who has a game that sure seems like it would translate to today is still a big deal.

I end up comparing Arizin with Ginobili, and right now they seem to have quite a bit in common with the exception that Arizin was a big minute alpha. I look at all that, and I'm siding with Arizin at the moment.

[Doctor MJ]

Nominate: Dikembe Mutombo

Outstanding and consistent defender and rebounder.
Solid double digit scoring on strong efficiency.
Great longevity with the ability to decline VERY gracefully. At times it seemed like he never got any worse, he just had to take more breaks.

Lots of guys on my mind right now, but Dikembe is striking me harderst.

[Doctor MJ]

I'm fairly certain Magic and Nash's lower than they should numbers are related to pace and not the lack of value of assists, just because we saw Paul put up the 21, 11 statline and had 18 WS 30 PER level. Hard to imagine that 18, 12, 6 or 18 and 11 at .62-.64 TS% wouldn't be enough for numbers nearly as high on a slower team

Re: Kiki - I admit it just doesn't seem like he's that good of a player like a Rashard Lewis type and his big stats have stood out as probably inflated, while I buy English's ability is responsible for his - but I'll admit that may be giving Kiki a short stick

It's not about assists, it's about floor generalship. While it is true that distributors are systematically undervalued by PER assist-wise, there's much more to a distributor than assists. Box score stats tell us that guys like Brevin Knight and Andre Miller had peaks as passers similar to Nash & Magic, therefore any box score stat is going to massively underrate Nash & Magic. Nash & Magic getting underrated because of their BBIQ-wizardry is no reason to conclude that a mediocre BBIQ guy who is not a distributor could be having the same underrating happening to him.

Re: Kiki. Glad you can admit that, and I'll certainly admit to doing arguably the same thing in place as far as being relatively dismissive of a guy with some big stats. I stand by my concern though with regards to finding so many reasons to dismiss various stats that are independent of each other, and then not standing firm with the same rationale with all players.

I don't believe you're looking to just cherry-pick stats to support your pre-existing opinions, but you do seem to have an approach that is pretty prone to that type of issue in practice. When you find a specific issue with a stat, use that to consider the waters muddied generally, and then use that muddied water to dismiss the stat in a situation that is not only independent but arguably the opposite of the original scenario (which is what you do when you use BBIQ & passer underrating to help a middling-BBIQ & scorer), you've clearly reached territory where if you're not careful your analytical prowess can keep you from ever objectively looking at evidence that disagrees with you.

[therealbig3]

Looks like the vote is over, but just to make sure so that there's no discrepancies:

My count:

Vote:

Unseld-5 (penbeast0, Fencer reregistered, lukekarts, JordansBulls, Laimbeer)

Marques-2 (therealbig3, ElGee)

Manu-1 (drza)

Gasol-1 (Dr Mufasa)

King-1 (ronnymac2)

Arizin-1 (Doctor MJ)



Nominate:

Carter-3 (therealbig3, Dr Mufasa, ronnymac2)

Penny-3 (ElGee, JordansBulls, drza)

Schayes-2 (Fencer reregistered, Laimbeer)

Worthy-1 (lukekarts)

Mutombo-1 (Doctor MJ)

B. Jones-1 (penbeast0)


So Unseld gets in ( ), and either both Carter and Penny make it, or there is no nominee this round.

[therealbig3]

I'd also like to ask exactly why so many people voted for Unseld. I saw pretty much no arguments for him, just a bunch of votes...which just tells me that either A. people are just going off his resume, which just confuses me as to why people continue to insist to do that, when it's been made clear that that's not the intent of the project, or B. people don't really have an argument for him, but feel like he just "has" to make it, because he's frikin Wes Unseld (much like the Cousy vote), and so just vote for him without even trying to justify it.

Clearly, I disagree with Unseld getting voted in now, but come on, make an argument at least, so that maybe it wouldn't be a complete mystery to me as to why he got voted in. A and B above aren't good reasons...at all.

[ElGee]

What the he'll? I didn't even notice the unself votes. Maybe that's bc there was no argument...

[Doctor MJ]

I'd also like to ask exactly why so many people voted for Unseld. I saw pretty much no arguments for him, just a bunch of votes...which just tells me that either A. people are just going off his resume, which just confuses me as to why people continue to insist to do that, when it's been made clear that that's not the intent of the project, or B. people don't really have an argument for him, but feel like he just "has" to make it, because he's frikin Wes Unseld (much like the Cousy vote), and so just vote for him without even trying to justify it.

Clearly, I disagree with Unseld getting voted in now, but come on, make an argument at least, so that maybe it wouldn't be a complete mystery to me as to why he got voted in. A and B above aren't good reasons...at all.

Well I'll say up front that Unseld being nominated etc never really bothered me that much. He certainly never deserved an MVP, but is he the GOAT Bullet (I refused to say Wizard)? Yup I'd say so. Yeah I'd say he went too high, but not so much that it really bothers me.

However, I'm with you that the way he got in bothers me:

1) Very little discussion in general in the thread.
2) Arguments for him not really made.
3) Voted in exactly 10 spots after nominated, i.e. "I guess it's his turn"

I really think anyone voting for a player after not voting for him the previous round should at least be giving some kind of commentary on why they voted as they did.

[ronnymac2]

I swear I thought Johnson was being voted in. lol

Can we just not count people who NEVER give anything reasoning for their picks? I'm sympathetic to people every once and a while not having the time but still wanting to participate, and I'm not expecting a novel for every pick, but a paragraph every once and a while would be nice.

[ElGee]

The thing about pace adjustment is that there's a long history of it screwing up PER and WS stats to me. Like, I know Steve Nash should have much higher offensive win shares in 2005 than Marbury or Billups.

I find PER to be borderline useless. Win Shares is at least a good box score approximation...but it's limited to the box. In this case, you aren't seeing something really relevant to pace, but the box. Nash's 2005 ORtg is 123 and DRtg 111...that's the difference between him and Billups. Nash actually beats Billups in WS/48.

I know that Billups dropping from 23.6 PER to 18.8 PER in Denver despite it being clear that he was the same guy, is wrong.

Box score stats aren't a measure of excellence, they are a reflection of circumstance. Changing teams will change many stats because of that. Just like changing teammates (injuries) will change a players stats. What you are seeing is a reflection of circumstance, not something related to pace in PER or WS.

I know that in 09 the gap between Kobe and Wade in PER being 24.4 vs 30.4 is wrong. I know that Magic's WS numbers are a complete disaster for the validity of the stat, his best WS season before 87 ranks #237 all time and below seasons by the likes of Hardaway, Schrempf, Arenas, Brand, Billups. His best OWS season period ranks 49th and his best before '87 is 85 ranking 193rd, below Schrempf, Brand, Bailey Howell, Chet Walker, Marbury, Rudy T, Dana Barros. Steve Nash ranks 179th in 2005 and 137th in 2006. Alex English does not have a top 250 OWS season and Kevin Johnson's only one is in 97 (and any stat that says KJ's best offensive season is 97 clearly has something wrong)

I just see a clear pattern of it pace messing up the numbers as much as it helps. I believe in the case of Monta Ellis and Kiki Vandeweghe, it's a good thing pace is there to downgrade the numbers, and for a guy like Dirk, that he gets his numbers raised by the slow pace. But there's just as many cases with the Magic and Nash's where it clearly messes up the numbers. I believe there is far too many case by case variance to just wrap a pace adjustment blanket over all the players and expect reliable results, some players it's going to help their stats, some none at all

Again, I don't think pace is the issue. You are talking about the box score. The composite stat doesn't know anything that isn't measured and input into it. Literally. You know Magic is better because assists aren't a very good stat at reflecting playmaking/pressuring defenses and he was one of the GOATs in that regard. Even composite stats might not be able to account for value of rebounding, an area we think Magic was positionally amazing. And they are swinging in the dark on defense. So it's not a pace thing, it's a box score thing. And IN the box score alone, Magic doesn't look like god and David Robinson does. Big deal...the box score is obviously limited.

As for APM, I like the stat but when I see Scola, Noah, Haslem, Bowen, Redick, Matthews ranking from mediocore to bad, let alone clear stars like Melo, Deron, Amare, Rose, Joe Johnson, Ben Wallace, Rondo, I can accept Melo's number much easier. I have next to 0 doubt about just about everyone on that list, guys like Scola and Bowen are precisely the type that should do well in advanced stats - and I only need a few to confirm that APM can and has been wrong. My general take on APM is when players are way up on the list like a Manu or KG or players are way down like a Jeff Green or Bargnani, they're probably onto something. Everything in between looks far more random to me. And in Melo's case having him rank top 20 in offensive APM and just having most of it wiped out by defensive stats means as much to me as his overall number anyways, I feel like I know exactly what Melo is defensively, you can fit him on a just fine defensive team due to his athleticism, he's not going to be standout, but like most perimeter players as long as they don't completley miss what's happening on defense like George Gervin or Michael Redd it's not going to set you back or prevent you from playing good defense, which the Nuggets have done.

Carmelo Anthony has some serious defensive issues. He's actually not a bad one-on-one defender -- he will use his size and dig down in a proper stance. His foot speed isn't the greatest but he's strong and long. However off the ball it can be a disaster, in the way a lazy HS player can be a disaster. He likes to linger near the midcourt at times and not peak up a cutter as they come across the timeline, or sometimes off the ball he just loses his man/won't rotate which leads to an easy score. These are BAD defensive errors when they happen repeatedly, as they do with Anthony.

My counterargument to your 'stats don't back him' point is that they do - it's just raw stats can be as informative as PER/WS/APM. In this case he is a 25-28ppg scorer on one of the best offenses in the league at his best, and he passes the logic test to me as someone who's helping his team get open shots by attracting defensive pressure

There's value in this. But the question is how much value? It's useless to me to reduce analysis at this level to "25-28 ppg scorer on good offense," which might be why you like Alex English. Maybe a better scorer makes them an even better offense. Maybe a more all-around player (22 ppg, more playmaking) makes them even better. It's not as simple as you make it sound to evaluate the impact of that scoring on a good offense...especially when the APM and raw on/off have shown a good offensive team sans Melo.

[Dr Positivity]

It's not about assists, it's about floor generalship. While it is true that distributors are systematically undervalued by PER assist-wise, there's much more to a distributor than assists. Box score stats tell us that guys like Brevin Knight and Andre Miller had peaks as passers similar to Nash & Magic, therefore any box score stat is going to massively underrate Nash & Magic. Nash & Magic getting underrated because of their BBIQ-wizardry is no reason to conclude that a mediocre BBIQ guy who is not a distributor could be having the same underrating happening to him.

The problem there is Miller/Knight and not Magic and Nash to me. An issue with assists is that if you stand on the perimeter, find a guarded 20 foot or 3pt jumpshooter and let him do the work to score (especially with lenient assist stats) it counts the same as a penetration and getting an open dunk. I call it the Jose Calderon assist out of personal experience. But once again, it's pretty clear to me that it's only the fast break PGs who's numbers look wrong. Magic, Nash, Stockton, KJ are low and guys like Paul, Billups, 96 Brandon are too high if anything. This is not a case of 'underrated distributing' because Paul was rewarded as much as anyone in the PG position's history

But this is all pretty irrelevant because this is PER and WS we're talking about, they're made up stats that we can do much better than when talking about players

Re: Kiki. Glad you can admit that, and I'll certainly admit to doing arguably the same thing in place as far as being relatively dismissive of a guy with some big stats. I stand by my concern though with regards to finding so many reasons to dismiss various stats that are independent of each other, and then not standing firm with the same rationale with all players.

I don't believe you're looking to just cherry-pick stats to support your pre-existing opinions, but you do seem to have an approach that is pretty prone to that type of issue in practice. When you find a specific issue with a stat, use that to consider the waters muddied generally, and then use that muddied water to dismiss the stat in a situation that is not only independent but arguably the opposite of the original scenario (which is what you do when you use BBIQ & passer underrating to help a middling-BBIQ & scorer), you've clearly reached territory where if you're not careful your analytical prowess can keep you from ever objectively looking at evidence that disagrees with you.

I really don't believe I'm doing this, I have no horses in this and have made a conscience effort to begin with the evidence first and come up with results afterwards. When we got to 45 on my personal list I took a while to sort it out on my personal list, putting guys like King and Parish as high as top 47 and then later dropping them to the mid 50s. I try to go through 1v1 comparisons for every player on my board and simply don't post most of them once I have an answer I'm satisifed with

As I said in the case of a Carmelo Anthony, the arguments against him come down to PER and WS and APM which have been noted, but my primary basis for picking players here is their skillset on the court and what it means to building a succesful team. To be honest, PER and WS mean practically nothing to me and haven't from the start. As for APM, I've explained my issues with it, all I need to know is that it says Amare, Ben Wallace, Deron, Rondo, Joe Johnson, Rose, Noah, Scola, Haslem, Bowen, Matthews, West and a handful of others are mediocore or bad impact players the last 10 years. I do not trust APM, there are too many variables on the court for +/- stats to mean a ton to me. I might trust Marc Cuban's secret APM machine that he's spent millions on but not the ones we have access to. It doesn't mean nothing, but in this case I feel quite strongly about Carmelo's value by my own logic about offensive scorers who draw attention and help teammates and the clear success the Nuggets have had, being a 50-55 W team with good talent and a 40-45 W team with weak talent and high ORTGs most of the way. I don't love Melo's game but we are at a stage now where the alternative is more limited players like James Worthy and Sam Jones and I think he's generally above those guys in offensive value