The Shrine

Written by the current most famous Michael Lewis, but one day i will surpass him.

http://www.nytimes.com/2009/02/15/magaz ... wanted=all

I read this last weekend. It's a fascinating read. (Also, I changed a typo in your title; Shame -> Shane.)

By the author of Moneyball, huh? Would explain the leaning towards stats and the insight into the "important" parts of the game. One of the best sports articles I've ever read

Kupek wrote:I read this last weekend. It's a fascinating read. (Also, I changed a typo in your title; Shame -> Shane.)

That's not a typo; that's just Flip being friendly. I think we all remember the "Tony Homo" phase...

Flip, it wouldn't kill you to lay off all the insults, you know.

I read something similar to this like an year ago and I don't see it being anything meaningful beyond a guy who can defend pretty well and be sort of okay at offense can be a useful player. You don't need fancy statistics to tell that Bruce Bowen or Ben Wallace isn't playing for their offensive skills.

At one point it mentioned like Kobe had 30 and Shane had 0 and that was supposed to be good. Well if someone else was guarding him when he have 40 or 50? Maybe some guy could've scored 10 and hold Kobe to 40? Who is to say that because you got a guy not scoring that just means the LA defense ends up not having to expand much energy and was able to defend the 4 other guys better? I saw the Laker vs Cavs game and all the Laker did was double LeBron and guard the 3 point shooters. There was some guy who was wide open with nobody on him that caught the ball in the key and he didn't even shoot, probably because that guy was good at defense but not so much offense. Well his lack of offense aptitude means Lakers can double on LeBron which means they're able to play better defense knowing 1 of the guy out of the 5 can't score so there was no point guarding him.

I don't see how you can possibly say because Kobe only shot 40% while scoring 35 while Shane scored 0 that is definitely better than him shooting 45% for 40 while his counterpart scored 10. I'm not saying which is better but just that you can't possibly know which is the outcome you want. Shane Battier is known as a good defender and that obviously has its value but I find this article seems to only exist to validate the correctness of some arbitrary metric kind of like how John Holinger in ESPN always has to write about how his PER or playoff predictor is totally correct even though it doesn't seem to ever work.

Kupek wrote:(Also, I changed a typo in your title; Shame -> Shane.)

Thanks, it was a typo, of course.

Don wrote:At one point it mentioned like Kobe had 30 and Shane had 0 and that was supposed to be good. Well if someone else was guarding him when he have 40 or 50? Maybe some guy could've scored 10 and hold Kobe to 40? Who is to say that because you got a guy not scoring that just means the LA defense ends up not having to expand much energy and was able to defend the 4 other guys better?

I think one of the more telling reasons as to why this kind of support player in valuable is due to Shane's teams WINNING. Granted, there could be a multitude of other factors in play, but it hard to ignore that his teams have miraculously had great seasons... This brings good defense to a new level that a lot of people would not think about, thats what was interesting to me.

Also, the part where Shane saw that his team would recover a loose ball, and he positioned himself to the most efficient spot on the court to drain a 3 to give his team the lead goes to show that not only is he smart, but a good shooter when he needs to be.

Basketball is a game where diversity in players works really well. You cant just have 5 Shaq's out there and expect to win. Therefore i think a role guy like Shane is amazingly useful as opposed to a decent defensive player who will give you 10 points a game. The Rockets have scorers, two pretty good ones, that cant all be double teamed.

Houston hasn't gotten out of round 1 in the Yao Ming era. I don't know anything about Memphis but I don't recall people ever attributing their turn around to Battier. Battier gets about $6 million from Houston, which supposedly got him because of what he does. $6 million is not a particularly high amount in NBA. Ben Wallace gets more than $10 million an year and he can't play offense. Maybe he was overpaid but pretty sure he was still getting quite a bit more than $6 million before.

I realize you can't just rewind time and find out what happens if you had X instead of Y but to me that article seems to only exist to try to validate why some metric nobody has heard of which ranks this guy highly must be good despite the fact that nobody can easily attribute any star-like qualities to him, and certainly nobody is willing to pay him like a star or even $10 million for what he does. At a rough glance at his stat sheet and accomplishment it looks like he's a below average scorer (doiesn't have a good FG%, only 40% or so) and a very good defender. There's no indication what he does would be better than a very good scorer who is a below average defender for example, though to be fair it's usually better to improve on defense than offense since the latter is probably more limited by talent while the former can be made up by hard work.

So you say he might be a hard worker with a good basketball IQ. That's certainly useful but every team has a player like that and only one team can win the championship.

Don wrote:I read something similar to this like an year ago and I don't see it being anything meaningful beyond a guy who can defend pretty well and be sort of okay at offense can be a useful player. You don't need fancy statistics to tell that Bruce Bowen or Ben Wallace isn't playing for their offensive skills.

At one point it mentioned like Kobe had 30 and Shane had 0 and that was supposed to be good. Well if someone else was guarding him when he have 40 or 50? Maybe some guy could've scored 10 and hold Kobe to 40? Who is to say that because you got a guy not scoring that just means the LA defense ends up not having to expand much energy and was able to defend the 4 other guys better? I saw the Laker vs Cavs game and all the Laker did was double LeBron and guard the 3 point shooters. There was some guy who was wide open with nobody on him that caught the ball in the key and he didn't even shoot, probably because that guy was good at defense but not so much offense. Well his lack of offense aptitude means Lakers can double on LeBron which means they're able to play better defense knowing 1 of the guy out of the 5 can't score so there was no point guarding him.

I don't see how you can possibly say because Kobe only shot 40% while scoring 35 while Shane scored 0 that is definitely better than him shooting 45% for 40 while his counterpart scored 10. I'm not saying which is better but just that you can't possibly know which is the outcome you want. Shane Battier is known as a good defender and that obviously has its value but I find this article seems to only exist to validate the correctness of some arbitrary metric kind of like how John Holinger in ESPN always has to write about how his PER or playoff predictor is totally correct even though it doesn't seem to ever work.

There were a few underlying points behind the article:

1) Basketball just doesn't have the statistics to properly measure their players. They basically only have flashy stats or ones that are easy to measure. So it's tough to gauge how a guy like Battier really affects the game without developing some of your own stats as the Rockets' new guy did

2) Trying to use current stats to value a player like Battier is silly. The 30 vs 0 comparison was just to make a point: everyone in that article was talking about how great he was yet Kobe was outscoring him by 30. It just flat out didn't matter what that difference once because of how he had affected the game. Ironically, it's the use of these stats which have no correlation to his actual value that have made him a marginal player in everyone's eyes except for the Rockets

So, it wasn't holding Kobe to 30 that was the big deal. It was holding Kobe to 30% shooting that was the big deal. Sure he got his points but only at the expense of many attempts. That affects the entire team. Wouldn't it be better for Kobe to get 20 points and take 20 less shots while his other team takes those 20 shots and hits 50%? He simply put Kobe off of his game and it hurt the Lakers.

And another getting 10 points with Kobe getting 40 is not a good comparison. For Battier to get 10 points on 40% shooting (a pretty solid FGP), he would need to hit 5 of 12 shots. Remember, Battier just doesn't shoot at all. So he's also not costing his team the misses that come with trying to get the 10 points. Let the rest of his team, who likely can shoot better than him, try for a higher percentage. That's what it's really all about, turning as many of your teams possessions into points that you can.

Can you find anyone else who can throw the big-time players off of their game that consisently AND hit 40% from the field consistently? Probably not. For argument's sake, let's say you can. Now that you've brought in someone who's got the stats that contracts are based on, how much is he going to cost you? There's a luxury tax to worry about. So a guy like Battier should be paid a bit more if there were stats to show his true worth, but to try and put the two together (high scoring and shut-down defense) would take up a huge part of your salary cap.

And yes, there is the whole thing that because he doesn't shoot much that he's easier to defend. But that doesn't mean that defenses can give him open shots. If he starts seeing those he's gonna start shooting more. They have to guard him still. And part of that article was talking about how his positioning on the floor even on the offensive side opens up lanes or opportunities for others. Again, something stats don't measure but can mean a lot.

I love the idea of the +/- rating in basketball. It shows just how that player affects the game, just like in hockey. Considering how much impact one player can have in a basketball game - other than the goalie, pitcher, or QB, any basketball position can affect a game more than any other position in any other major team sport - this should be a huge stat. It also allows for a large variety of skills - both offensive and defensive - to come into play when determining a players' worth. This should be one of the major stats IMO. It then forces you to analyze why. Sure, sometimes it could be because you play with Pau and Kobe most of the time and you're likely to have a +\- that's favourable, but like in hockey, you're a team out there and what you bring has a big effect on how well your team does (as mentioned above, single players have more effect in basketball than any other sport).

But basketball more than any other sport is very much in its infancy when it comes to stats. They're still trying to get over the flashy stuff and down to the important stuff. It takes time (look how long it took for the WHIP and OPS stats to become huge in baseball) but now that there's a couple of generations with basketball being a very huge sport (it was the Johnson-Bird era that took basketball from hockey-level in popularity to baseball-level in the pro leagues) you're starting to see some of that come in.

And the Rockets still lost that game. Rockets still hasn't won a series in the playoffs yet even though they got this guy who is supposedly a 'no stat all Star'. They also have 2 guys who definitely have the stats to be an all-star (Yao Ming and Tracy McGrady). Winning in itself does not validate if your metrics are correct but it's rather bogus to say on paper this team should be totally awesome and then they never even got out of round 1, let alone using that as a justification for why your stats are good.

James Posey gets paid around what Shane Battier does. Some people think he was instrumental to Celtics won last year but right now he looks like he's overpaid on the Hornets. I'm guessing Hornets thought Posey must have some intangible qualities that don't show up on stats when they signed him and now it looks like these intangible qualities are more like nonexistent qualities. I sure remember a lot of the experts talk about how you can't measure Posey's effect on the Celtics by just stats (he didn't score much and just played D) and I don't see them issuing an apology saying it looks like he really didn't have much beyond his stats say.

While basketball's basic stats aren't exactly the most complete, I don't see why we're supposed to believe in bogus stats people made up that often fails to have any bearing on reality. In baseball WHIP is probably better than ERA and it even make sense. If you need to explain why player X is good even though nobody in the world think he's good (or at least any better than any other guy making $6 million in NBA) that to me is just coming up with an excuse why your stats predicted this way.

John Hollinger has a column in ESPN every time whenever his formula predicts a different team would be the one that wouldn't make the playoffs in the Western Conference out of the 9 in contention. So far I've seen Spurs, Suns, Jazz, Mavs, and probably Rockets and Hornets showed up there, and everytime there is a totally logical statistical explanation as to why this is the team that wouldn't make playoffs because the formula says so. Except only one of those 6 teams can be the team that didn't make the playoffs, and if you look at the trend the team that won't make the playoffs is almost always the team that is currently in 9th place.

He's also been saying no team in West has a shot at dethroning the Lakers, until the Spurs crept up to #2 and then suddenly his formula predicts the Spurs has a chance at beating the Lakers. And this guy is one of the more well respected guys in the realm of new statistics.

Now stuff they keep track in 82Games.com (I think that's what is called) like scoring during clutch time is actually pretty useful. But then those guys are more interested in just compiling like every stats that might help instead of trying to prove they got this magic formula that tells them everything they need to know about the universe.

Here's Shane Battier's stats from 82games.com

http://www.82games.com/0809/08HOU7.HTM#bypos

And compared to rest of team:

http://www.82games.com/0809/0809HOU.HTM

And of course most of these makes even more references to obscure ratings that I have no idea what it means, but one thing you can notice is that the guy Shane Battier is guarding scored more than he did. That is to say although he is a good defender he is not a good scorer either. I think the site uses the Roland Rating to determine the overall value of a person and I have no idea what it even means, but he's certainly not very high on that (Yao Ming is +10.2 vs Shane Battier at +1.1).

Don wrote:And the Rockets still lost that game. Rockets still hasn't won a series in the playoffs yet even though they got this guy who is supposedly a 'no stat all Star'. They also have 2 guys who definitely have the stats to be an all-star (Yao Ming and Tracy McGrady). Winning in itself does not validate if your metrics are correct but it's rather bogus to say on paper this team should be totally awesome and then they never even got out of round 1, let alone using that as a justification for why your stats are good.

James Posey gets paid around what Shane Battier does. Some people think he was instrumental to Celtics won last year but right now he looks like he's overpaid on the Hornets. I'm guessing Hornets thought Posey must have some intangible qualities that don't show up on stats when they signed him and now it looks like these intangible qualities are more like nonexistent qualities. I sure remember a lot of the experts talk about how you can't measure Posey's effect on the Celtics by just stats (he didn't score much and just played D) and I don't see them issuing an apology saying it looks like he really didn't have much beyond his stats say.

While basketball's basic stats aren't exactly the most complete, I don't see why we're supposed to believe in bogus stats people made up that often fails to have any bearing on reality. In baseball WHIP is probably better than ERA and it even make sense. If you need to explain why player X is good even though nobody in the world think he's good (or at least any better than any other guy making $6 million in NBA) that to me is just coming up with an excuse why your stats predicted this way.

John Hollinger has a column in ESPN every time whenever his formula predicts a different team would be the one that wouldn't make the playoffs in the Western Conference out of the 9 in contention. So far I've seen Spurs, Suns, Jazz, Mavs, and probably Rockets and Hornets showed up there, and everytime there is a totally logical statistical explanation as to why this is the team that wouldn't make playoffs because the formula says so. Except only one of those 6 teams can be the team that didn't make the playoffs, and if you look at the trend the team that won't make the playoffs is almost always the team that is currently in 9th place.

He's also been saying no team in West has a shot at dethroning the Lakers, until the Spurs crept up to #2 and then suddenly his formula predicts the Spurs has a chance at beating the Lakers. And this guy is one of the more well respected guys in the realm of new statistics.

Now stuff they keep track in 82Games.com (I think that's what is called) like scoring during clutch time is actually pretty useful. But then those guys are more interested in just compiling like every stats that might help instead of trying to prove they got this magic formula that tells them everything they need to know about the universe.

They lost the game on a low-percentage shot that the best player in the world on arguably the best team made at the buzzer with the Rockets' best player on the bench. The fact that they were even in a position to win - they had the lead until that final shot - was in no small part due to the defensive job Battier did on Kobe. You can't argue that. At the end of the day, in any sport, talent can overcome a lot.

No one at any point said the Rockets were good. This discussion has never been about how good/bad the Rockets are. Winning/losing team-wise is a different animal altogether. There's a reason the Yankees don't win the world series every year.

Not sure why Posey was paid what he was paid. I'd have to look into it more. It could simply be someone thought he was worth it. Maybe he's paid going rate for what he brings and Battier is underpaid? Not saying that's the case, just bringing up another possibility. Also, don't they play different positions?

+\- is not a bogus stat. Ask anyone who knows anything about hockey. It's not a conclusive stat but it's a decent measure and one that would work well in basketball. Besides, the Rockets' guy was trying to explain why Battier is better than he appears to be not trying to justify anything. Remember, he was considered a fringe player until he got there and now he's become a pretty important part of the team. He's trying to show you why not to explain a big contract he gave up.

And I agree that they should keep more timely stats like scoring in the clutch, scoring in the last 5 minutes of the 4th quarter when your team is down by 10 or less, etc. The idea is you keep trying these stats, scrutinize the fuck out of them, and keep the ones that makes sense. That's how baseball got the clusterfuck of stats it has now that everyone seems to use in some way (like the Hold).

There are like at least 3 kinds of +/- stats in basketball. If you read the article they picked one where he was +6 that nobody has ever heard of. In fact I think it's something only one team claim that it works. That to me is bogus until proven. I looked it up and Roland rating is basically a +/- stats relative to your team, so Yao being a +10 means his presence is way more important than Battier at +1.1.

Have you ever seen the Cavs play? If they got someone like Ben Wallace on, who cannot score unless he's directly under the basket and also can't shoot free throws , then whoever was supposed to be guarding him will be doubling on LeBron because they know Ben Wallace isn't going to score. Now Ben Wallace is a great defender but you can't just ignore the fact that Cavs play basically 4 on 5 on the offense when he's on. If LeBron can't score through 3 guys like some kind of manga superhero the Cavs would be really bad at scoring. Lakers aren't particularly strong on defense and Battier scored 0 in the first 3 quarters or so. That means whoever was supposed to guard him probably didn't have to work very hard and can be doing something else.

Any team can beat any other team in NBA on a good night. Jazz beat the Celtics yesterday and I think they're still missing some of their good players so where's the story on that? Sure Battier hold Kobe to a low % and that's a plus, but if you look at say PER (a reasonably good measurement of a player's offensive prowess), Shane Battier is far below the average on that, which means his offensive contribution is much less than an average NBA player of his position, so that's got to be a minus too, especially since Rockets really can't score that well in the 4th quarter (Yao Ming and Tracy McGrady gets tired easily having to do so much of the offensive since everyone else on the team can't do it).

In one of the final games between Spurs and Pistons, Robert Horry scored like 0 points in 25 minutes and then made a 3 to win the game at the end. It is possible if you have any normal player in his position in his first 25 minutes then you'd never needed the last second 3 to win the game, or you might not, but you can't just say the fact that a guy did nothing on offense isn't a liability. Taking that particular game in question, it is almost unimaginable any player would've done worst on offense than Battier in his position and you can't just that has no effect because he plays good defense.

Again, I wasn't talking about the wins-loss by a team, more what we should be using as a measure to determine a player's worth to that team. I'm arguing that there's more to a player's worth than the standard stats that are used right now in basketball and that there are other stats that need to be brought in.

The version of the +\- used by the Rockets mirrors that of hockey: how many more points (goals) did your team score vs the other team during the minutes you were on the floor (ice). Not sure what the other +\- ratings are but that's the one that mirrors hockey and the one that I was referring to. I'll have to see what other factors the other +\- stats take into consideration, but I'm not sure why you'd want to tweak something like this. I'm guessing that the Roland rating takes into account a whole whack of stats on a pre-determining weighting, come up with a median, then say this person is a + or - based on that median.

The hockey +\- referred to in this article is a simple yet relatively complex measure of a single player's effect on the team's success. It takes into account both offense and defense but is not a conclusive measure of either. In the same way you'd expect LeBron or Kobe to have a high +\- as they score so damned much, a very good defensive player (in hockey, the defensemen usually have the highest +\- as they're on the ice the most) could also have a great +\-. Sure, it could also be a function of who you play with, but you're still bringing your contribution to that team and at this level, you need to contribute something to the team, it's just too tight on the pro level.

That was the argument made by the Rockets' guy on Battier. Sure he didn't score, but his team did better - most games are decided by 10 points or less, so signficantly better - when he was on the floor than when he didn't. This would be an argument against him being an offensive liability. Even if he doesn't score, if his team is shooting a higher percentage when he's on the floor or is just scoring more, he's obviously helping.

So the question then becomes: if your team scores 6 more points during your time on the floor than they let in, are you more valuable because you scored those points yourself or because you helped prevent those points from being scored or helped the scorer score the points? Battier is a very extreme example, I'm talking more generally. Right now, there is no generally accepted measure of seeing if someone like Battier could possible be as valuable as someone like Kobe because the stats are more one-sided. A stat like that could go a long way to helping measure the worth of a player who may otherwise be overlooked.

This doesn't automatically mean that it'll help any defensive player. It will help someone like Battier but it could hurt someone like Posey. If Posey has a rep for being a good defensive player but has a -ve +\-, then that perception could be argued against using a stat like that.

Roland is some kind of +/- stats that compares how much your team score when you're on versus when you're not. I think that article says they used some generic +/- and found he was only +1, and then use some other +/- where he's +6. It's pretty clear if you tweak the variables enough you can obviously have any kind of +/- results. At a rough glance most of the +/- measures have Battier as a positive, but not necessarily very high number. He's obviously not very good at offense but very good at defense. That may mean he's better than average or not overall.

I'm not saying I don't believe in stats but that article to me reads like 'Shane Battier is awesome because I found this new stats that says he's totally awesome.' Besides alluding to some stat no one has heard of (which may be right, but it sure isn't proven yet), most of it is just anecdotal like 'yeah look at how he played on this play, if it's some other guy no way he can do it!' But this doesn't factor things like if it's some other guy maybe he wouldn't scored 0 all this time too. By most common stat metrics I'm aware of, Shane Battier is a pretty average guy. I know every NBA team and stat geek has their own super secret stats they're using but until someone can prove they're really that much better than say just looking at points, FGA, rebounds, etc., I'm not going to believe something that seems to outright defy conventional wisdom.

So far as the on/off argument goes, in NBA you're more likely to see significant swings in talents compared to hockey. I'm pretty sure your offensive powerhouses aren't going to be on during a penalty killing time. It is entirely possible someone has a good +/- because he's often playing when the other team's second unit is up. I'm pretty sure Ginobli has a very high +/-, because he's basically a starter that comes off the bench. Besides, the normal +/- I can find puts Battier at the +1 range, which is hardly a difference maker. Like I said Yao is +10 and that's what common sense would tell you too, so no need for fancy stats there to see that Yao is the difference maker on Rockets.

Replay wrote:

Kupek wrote:I read this last weekend. It's a fascinating read. (Also, I changed a typo in your title; Shame -> Shane.)

That's not a typo; that's just Flip being friendly. I think we all remember the "Tony Homo" phase...

Flip, it wouldn't kill you to lay off all the insults, you know.

Pfft, anybody that's seen It's a Low Down Dirty Shame, would get Shame and Shane confused.