[Bill Simmons] Nothing Has Altered the NBA Like Mike D'Antoni

I do disagree with this blog post, as well-written as it is. I can't stand Simmons as much as the blogger but in this case BS is correct, albeit overdramatic and snotty.

The post makes two separate points:

First, that Nash was not a D'Antoni creation. This is a strawman, because Simmons doesn't make any claims this Knicks fan does not: that is, he went from being a good player in Dallas into a two-time MVP because the Suns play[ed] at such a frenetic pace, inflating Nash's assist total and making him look like a better distributor than he is. The post rips Simmons for raising this obvious point that the poster himself acknowledges, in an attempt to defend the number-crunching accountants who double as ESPN commentators and bloggers. Stat geeks were in part to "blame" for Nash's MVP ascendancy, but moreso was the media hype behind Phoenix. Advanced statistical analysis done at the time might not have favored Nash, but the casual fan/sportscaster looked at his assist total and thought, "Jerry West!"

Second, that D'Antoni is an elite coach. True, that the Suns never won a title isn't to say that D'Antoni was a complete failure. But the fact is that, barring luck (and some system coaches do get lucky, like the '04 Pistons; Larry Brown's half-court snoozefest of an offense is not in vogue anymore for good reason), they could not have won a title with D'Antoni's system. It is unintelligent basketball that happens to maximize the talent of a sorry Knicks team, but its a style truly-elite coaches can outsmart. Same goes for other gimmick coaches, like Jim O'Brien (remember when people thought he was a genius?)

 

I've always enjoyed Simmons and used to read his column regularly, but lately his style is getting pretty stale. This article in particular is really bad -- his mailbags are usually funny -- the football picks and factoids are almost always annoying.

 

If the post is a little all over the place, it's because Simmons is. He starts out by talking about how Chris Duhon has become some awesome amalgamation of several HoF point guards, when if he knew anything about per-minute or pace-adjusted stats, he'd know that Duhon is playing (13.1 PER) at about the same level you'd expect by looking at his career arc.

He was at 11 PER last year, he's a year older (and better), and his increase in per-minute production is what you'd expect from someone who has seen their minutes go way up (about 17 more minutes per game). With guys like Oliver Miller and Michael Sweetney as the exceptions, the overwhelming majority of players see their per-minute numbers go up as their per game minutes go up. It's a case of being more comfortable on the court, having a defined role, etc. Everything you'd expect.

Simmons and I have been at this for about the same amount of time, me starting back in 1997 with OnHoops, him at that Boston city site. Years ago, I took it upon myself to question what I knew and work hard to learn about things I would have dismissed in years previous. Simmons never put in that work. Guys like Henry Abbott or Chad Ford, dislike them all you want, but they took the time to learn about different ways of looking at the game, when they couldn't tell you what the hell points per 100 possessions meant three years ago.

Simmons decided, years ago, that he didn't have to learn anything new about the game that teaches us something new every damn day. And columns like this are a result.

But it won't matter. Every university gets a new freshman class every year, ready to go apebonkers over the fact that he can compare the Colts to an episode of Entourage.

 

"Stat geeks were in part to "blame" for Nash's MVP ascendancy, but moreso was the media hype behind Phoenix. Advanced statistical analysis done at the time might not have favored Nash, but the casual fan/sportscaster looked at his assist total and thought, "Jerry West!""

Those aren't the "stat geeks" Simmons is sneering at. He's associating John Hollinger and Dean Oliver with the sort of drooling nonsense Jack McCallum and Marc Stein came through with that year. But Simmons can mitigate McCallum and Stein's influence. He can't do that with Hollinger and Oliver, because they work harder than he does.

Stat geeks had nothing to with Nash's MVP. They were beside themselves when he won in 2006, and less so in 2005. A lot of the votes in that APBR thread, that's not a good cross section of what they were thinking. There are a lot of lurkers, blogger types, on that site, and they were there back then. Hell, Hollinger didn't even have Nash ranked as the best player on his team back in 2004-05.

All the "stat geeks" have done is give Simmons an intelligent way of getting around the asterisk that he frets over. "Stat geeks" don't need that asterisk, because we've worked harder and understand context, while utilizing the statistical measures given to us in order to make sense of it all.

Then again, if he understood this, he wouldn't have any column fodder, because none of this would be surprising to him. But because he stays ignorant, he's allowed to be surprised, and then make rash and idiotic declarations that people who don't pay attention and are reading at work get to hold as gospel.

 

The changes to the rules and points of emphasis is what put this style of play into the limelight. Well, until the playoffs where the officials forget where they put their whistles and it's an entirely different game... SSOL goes to crap.

 

Thanks for the insight. I had no idea stat-heads thought Nash was overrated.

I am not someone who is versed in statistics, and find it hard to relate to NBA commentary based off of it. That said, when I was in middle school (late nineties/early '00s) I enjoyed reading Dean Oliver's site: it had the perspective of someone who both knew math/logic and obsessively watched basketball. He even talked about meeting Bill James as a kid and discussing how to use the Pythagorean theorem to determine which teams were better/worse than their record. I assume some NBA team's scouting dept. snatched him up.

I am not a fan of Hollinger's articles, though I can't criticize him personally--his formula has made him a lot of $. I think the PER is a gimmick, though, because such a catch-all formula has an inherent degree of subjectivity. Who determines how each stat gets weighted--why 2/3rds for assists, e.g.? Why is it that players incapable of playing with team's first units are ranked so highly? Why is passive play, a la Jose Calderon, to be valued so highly, simply because he doesn't turn the ball over? (Some of the best PGs are turnover machines). And this is on top of the obvious problem that basketball is not very quantifiable.

But however wrong Hollinger &co. are, at least they try, as you say. You nailed the fact that Simmons never, ever changes his philosophy of analyzing basketball, and merely glosses over the billion times he's wrong.

 

Its around the corner....You know whats coming


NBA All Star Weekend.

with your boy B-Simmz at the scene reporting all the events.

 

Exactly. You have a real good offensive line it can make any mediocre back look like a hall of famer. Simmons obviously has no idea what the hell he's talking about

 

Passive play being rewarded, or smart play? I'm going with the latter, and also consider that some of the most turnover-prone PGs who are some of the greatest (Nash, Magic) have PERs much, much higher than JC's. JC is around 19/20, and he's at his peak right now. Nash at his peak was constantly at 23 and 24, while Magic even got up to 27 one year.

So these guys aren't being rewarded up to the turnover-prone(ish) types like Nash and Magic. Not even close.

And, let's face it: turning the ball over is the absolute worst thing you can do in any given offensive possession. Obviously worse than a positive contribution (a made shot, an assist, a made FT), and way worse than missing a shot. It should be rewarded.

As to the weight of the stats, Hollinger has gone into great detail in his books (which are still around, at Borders and the like) about why each statistic is given the weight that it has in the PER formula. It's not a gimmick. And in his books, it's a relatively minor character. It's just ESPN that has hyped it beyond his other work, findings, and analysis.

It doesn't help that people like Simmons constantly refer to it as a "better than" system, as if PER is a way to rank players. Hollinger has never, ever, ever, ever, ever, ever, ever, ever, ever, ever, ever claimed that it's anywhere near an end-all stat. It's one tool among many.

And, as it happens, far and away the best catch-all stat we have for quick reference regarding basketball players.

 

I fully understand Simmons's other "what if"s, except this one above. Could anyone explain the background story about it? it happened long ago for me to catch, tia.

 

Kelly, thanks for the insight. I would think, new waves of froshies aside, that Simmons would run out his pop culture cache in the not-too-distant future. I know from teaching frosh that one can only be hip enough for so long. (But I thought this of Jim Rome too, and his formula keeps moving along; it may be different enough from Simmons though, in that Rome isn't nearly as pop culturey as he is self-referential and cultish.)

I want to ask you about Chad Ford, not to derail the topic... He strikes so many of us as being confused about some basketball basics, but your take is that he's learning and working at it. I'll take your word, since I haven't read anything by him in probably a year. I just could never subscribe to the Insider thing. So, (A), you think he's improving? and (B) do you know to what extent he was the "brains" behind making himself an exclusive insider commodity?

 

I just don't see why PER is useful at all. Consider the example in your post:

We know that Magic Johnson (one of the NBA's greatest players) is better than Steve Nash (superstar), who is better than Jose Calderon (average NBA starter). If the efficiency index told us differently, we would know it were flawed. Instead, it tells us what we already know. It's lose-lose.

Let me explore in more detail. What tells us who the more productive player is, apart from the stat, are our pre-theoretical intuitions we cultivate from watching the games. If the statistic lines up with said intuitions, we learn nothing, and if the eyes and the stat work at cross-purposes, we necessarily question the latter. This is because there is no reason the stat should be preferred over our intuitions; the stat is a thought experiment, not a physical entity with intrinsic/independent value.

As a Celtics fan, I can say, for instance, that there is no way Leon Powe is one of the NBA's most efficient reserves. (He's good but, damn....with the number Hollinger assigned to him he'd be third-team all-NBA good). I cannot change my opinion because a stat says differently, because the stat does not confirm what I see. The stat may make me watch more closely, but as an analytical tool it seems awfully redundant. Anyway that's my two philosophical cents.

 

First of all, Jose Calderon is not an "average NBA starter," at least on the offensive end. He has shot very efficiently, racks up a lot of assists, and keeps his turnovers low. He's a good player, and PER tells you that.

You're choosing to make PER "bad." If a metric tells the truth, then it is a good metric. If not, then it's flawed and not very useful. But for you it's damned if it does, damned if it doesn't.

No, no, no. "Intuitions" mean nothing. Anyone can think or feel anything they want when "watching" a game. I hate how people say "I watched the game" as if that's actual support for any assertion they make about basketball. I watch games too, and pay attention to a lot of the little things. Even then, I find it amazing how many things I miss (for example I may not know that a certain role player is really hot, or has gotten 7 rebounds in the first half already, until the announcers say so). The human mind is not a perfect recorder of information. I don't presume to have a perfect memory of everything that happened in a game. Other guys, who may be casually "watching" the game while chilling with their friends, or getting pissed off about a player they hate, or being total homer fanboys, also don't see everything perfectly.

Now, what you're saying is that the intuitions of flawed people, who bring prejudices and preconceived notions when they watch a game, should be used as the "truth?" That doesn't even begin to make sense. It's coming to a conclusion without any real evidence.

Stats are not some made-up crap with no connection to reality. They're a record of what actually happen. Now this isn't to say that there aren't some flaws in how stats are recorded, or that various stats aren't limited in what they represent, or that people can't use stats in misleading ways. However stats are for the most part unbiased, quantifiable measurements. That's better than some guy saying "I think Kobe is the best in the league!"

How does one determine that Kobe is, in fact, the best player in the league? Can someone use their "intuition" to explain how or why he's better than Lebron this season? Many people will say that Kobe is the best without backing it up. Stats can show that Lebron is performing at a higher level right now, and explain why.

There is no denying that Leon Powe played at a very high level in his limited minutes last season. He scored a lot for his minutes, shot very efficiently, got to the line a lot, and grabbed a high percentage of available rebounds when he was on the court. That all actually happened, and it was all reflected in PER.

Does that mean he was really that good? A deeper understanding of PER and other statistics would suggest not. Powe's jump in PER from his rookie to sophomore year was extremely high. Hollinger understands that sometimes players just have a "fluke" year and come back down to earth afterwards. And that is exactly what happened, as Powe's third season is good but not at star-level efficiency like his second season.

 

This isn't the case at all. Whatever year it was that the suspensions happened, Phoenix was up 3-2 on San Antonio, with a home game coming up. Without the flukey suspension issue, they have a very good shot of winning the title that year. Another year, they made the WCF and lost with one of their star players (Joe Johnson) hurt.

They've had some bad luck in their best years. They didn't need good luck to win a title - just a lack of bad luck. They've more than shown their ability to compete with other elite teams in the playoffs.

 

Stats suck. PER sucks. I hate this new Era of basketball that revolves around stats. Just watch the freaking game to see who is good or not.

 

foodworld is right.

PER isn't terrible but if used as a catch-all stat allows people to say "STFU look at Trevor Ariza's 45th ranked PER of 18.46, Carmello Anthony has the 56th ranked PER of 17.68. Ariza> Melo!!"

Is Ariza more efficient than Melo? Maybe, but does that mean he's a better player? At some point we've crossed over from using stats as a tool for understanding basketball to taking it as the gospel. Sometimes "intuition" is a better evaluator of the game. If you've been around basketball your whole life and understand its nuances sometimes even subconsciously you see things that have yet to be put in statistical form. There are so many variables in the game of basketball that it is next to impossible for someone to back everything through simple arithmetic. If it were that simple Morey's moneyball GM style would have us ripping through the league by now.

 

Stats are what happens when somebody watches the game and pays attention. If I were to evaluate a basketball game and had all the time and incentive to do the best job, I would sit down with a tape for hours, recording all the stats and taking notes on every play. Fortunately for many fans, the NBA's statisticians already do a lot of the work here. People like Hollinger and the guys at 82games.com do some more work for us.

I would NOT chill with my friends and knock back a couple beers while hating on the other team, then presume to have a perfect understanding of everything that happened.

foodworld was way off the mark. In other fields of study, people will actually put in some effort when analyzing things. They look at the evidence before coming to a conclusion, and try to measure things in quantifiable terms. But when it comes to sports, sports fans and even commentators and columnists think they can make up their own conclusions without any support. Yes, that was what foodworld actually advocated.

They haven't done the work; they likely just watched the game the same as everybody else. Yet they feel that they are above defending their arguments, and shout down the people who do put in more effort. Maybe it's because sports is just entertainment, or because many of the people involved in sports aren't interested in "geeky" numbers. But there is definately a lazy and anti-intellectual streak running through sports fandom. And while sports is just entertainment, the owners are spending millions of dollars and everyone from the players to the fans are emotionally invested in the results. These people would all benefit if coaches and general managers made more rational decisions supported by the evidence available to them.

 

Darn lack of edit function.

Why do people continually bring this up? Hollinger himself doesn't use it as the end all, be all. He openly states that PER doesn't do anything to analyze defense except for rebounds, blocks, and steals. I don't think I've seen many people saying PER should be everything, yet I do see a lot of people saying that people take it as gospel when they shouldn't. It's looks like the only way they can argue against the pro-stats crowd is to distort their position.

Carmelo may be getting his 20+ ppg, but that number is inflated by Denver's fast pace. He also does a lot to hurt his team while on the floor, between all the missed shots and turnovers.

All PER is saying is that, at this point in the season, per-minute and adjusting for pace, Trevor Ariza has been more efficient than Melo. Does that mean he can handle the scoring load like Melo can, or that he will be better next season or even by the end of this season? No. Ariza got off to a very hot start and his PER has been dipping as the season goes by. He's probably fluking. This is Melo's lowest PER in years. He'll probably surpass Ariza before this season is over.

Morey doesn't have the freedom or resources to do everything he wants, nor are all his rival GMs absolutely stupid enough to let him swindle them any which way. He has a salary cap to deal with, and inherited contracts. Morey also can't do anything about Yao or Tracy's injuries, or Artest coming in and sucking like he hasn't in years (still okay since we basically got him for nothing, and his contract ends after this season).

What Morey has done is make a series of smart, responsible decisions that have steadily increased the talent level on this team. It wasn't long ago that the Rockets were T-Mac and Yao surrounded by a bunch of old scrubs. Now we're one of the deepest and most talented teams in the league. It's up to guys like T-Mac and Artest to stay healthy and stop throwing bricks.

 

Intuition tells us that the notion of a round Earth is ridiculous.

 

Common sense tells you that a combination of watching games AND using stats gives you the best perspective. Just doing one or the other does not make one markedly more knowledgeable.

How is everyone arguing so much about this?