APBRmetrics

[Statman]

[Flint]

Honestly, pointless discussion. Iverson had a horrible series. Small sample size.

I think he is an average player, and overall below average for his career. Some of you guys think that's ridiculous. That's fine. You guys can go with the conventional wisdom. Mike, you want to say scoring is what wins basketball games, I am not going to bother to change your mind.

AI is a polarizing figure. I dont hate him because of his behaviour. I hate his game, that's it. You just can't look at his numbers and say those are elite numbers. You just can't. Steve Nash rebounds better than he does for chrissakes.

And vis a vis the +/- analysis of Efg, I'm sorry Statman, but I think its just absolute nonsense, never mind that it disregards how little Iverson does on a basketball court other than shoot. How you can say that Iverson raises his team's Efg by shooting 2% less than the team average. I just can't figure it out. What use is that statistic when a guy playes 42 minutes per game?

Put him in the Footwear Hall of Fame in Beaverton, but not the one in Springfield please...

[gabefarkas]

[gabefarkas]

[bchaikin]

Iverson's coming off his worst season

06-07? maybe his 03-04 season - in 06-07 he scored the same pts/g playing the same min/g as he did in 03-04, but shot 6% better overall, with more ast/g and less to/g...

[Flint]

Thanks BChaikin for backing me up there.

FWIW, I was an all league high school basketball player. I was second in the league in rebounding, first in offensive rebounding, so yes i do have a personal stake in thinking it's important. Yes we sent a player to the NBA, yes there were a few D-1 players also. I was also the captain of my college's club team, though very far from the best player there. Yes I am glued to the Mav-GS game. If I didnt love basketball why would I be here?

I love watching games, and have seen tons and tons of Iverson games in my life. I never thought he was very good. I always noticed him missing tons of shots and trying to force play and turning the ball over. When I watched that great Philly team, I always thought Mckie was better than Iverson. I thought Lynch, Hill, and Mutombo were the reason they were good, and I was right. Iverson was a bit above average that year I think. There were a lot of players on that team who had better seasons than he did. I thought that then, before I read the WOW, I think that now.

I may be mistaken but I think a great scorer isn't someone who just scores a lot of points, its someone who scores a lot of points fairly efficiently, more efficiently than his teammates preferably.

Charles Barkley just said Tracy Mcgrady has had a great series, that if they lose it wont be his fault. That is the kind of idiotic thing people generally say about scorers. The guy is 51-134 so far this series. 38%. Thats not good. He hasn't helped his team there. He has helped with his rebounding, but overall I think its fair to say his performance has been below average. He had 26 and 10 tonight, but it was still a bad game. He was 8-23 with five turnovers. Do you disagree? I mean is it too much to ask a great scorer to complete more than 40% of his shots before we talk about how great he has played?

Raw scoring is very overvalued in the NBA. There is very little question of that. Berri shows that it in a number of ways that are quite independent of his controversial player metric. No one I know has disputed that finding, despite everything that has been said about Wins Produced.

GM's clearly pay NBA players for how much they score, with relatively little regard for what percentage they shoot and everything else they do on a basketball court. That is dumb. Just like focusing on batting average rather than OBP was dumb.

I appreciate the value of creating an open shot, but it still doesn't make any sense to me to say Iverson increases his teams efg despite shooting well below it. What you seem to be implying is that if you replaced Iverson wiith a player who took as many shots while shooting at the team average, the teams Efg will still decline. That's just strange. Honestly, that just seems illogical.

[Ben F.]

I had excused myself from this argument because I don't think we will convince anyone who is staunchly in one corner - but I do think the discussion with Flint is more of a case of miscommunication than attempting to persuade.

I would restate the whole "inefficient scorers are valuable" argument by citing the skill curves chapter in Dean Oliver's "Basketball on Paper." The idea behind skill curves is that players cannot limitlessly increase possessions used without a decrease in efficiency. This makes perfect sense from a basketball standpoint. Imagine the player who is not an offensive threat at all - he gets all of his points from put backs. This player can shoot a very high percentage because all of his shots are right around the basket directly off a teammate's miss. Say he shoots 90% on these shots. That's quite an efficient player. He doesn't turn it over because he's only touching it off a board and he's got plenty of good looks.

But considering this player, we must also note that the player cannot get these opportunities at will - they are dependent on his teammates missing shots and him being in position. Therefore his 90% shooting is limited to this situation. Say his coach then decides: "I have a 90% shooter on my team! Why not have him take all my shots?" This doesn't make any sense. Clearly the player's 90% shooting is dictated by the context of his shots. If the teammates were to give him the ball every time down the floor and refuse to take any shots, he would NOT shoot 90%. If we can agree on this we're essentially on the same page.

Now to the skill curves. The idea is then that each player has a reachable efficiency at a certain usage level. In other words, say Mr. Offensive Rebounding 90% Shooter is asked to take on more of a role in the offense. Now he's only asked to score if he has a mismatch with a much smaller player on him. His usage will increase a bit. He may have a nice baby jump hook so that he can shoot fairly efficiently in this situation, but it's still not put backs and dunks. His shots become a bit harder. Perhaps the smaller player has quick hands and can get a steal now and then. With the coach asking him to take on a larger role in the offense, maintaining his extreme efficiency is harder. How hard it is to maintain this efficiency is clearly dependent on the skill of the player. If he has an array of post moves, solid hands and a knowledge of how to take advantage of his size then his efficiency might not decrease at all with his increased role in the offense. But if the coach then asks him to do more and more - posting up larger guys who can defend him, playing through double teams, etc. - his efficiency is bound to go down. The rate at which efficiency decreases as usage increases is what defines the skill curve.

All this seems fairly straightforward - so how does it impact our discussion of Iverson and McGrady? I'll let Dean Oliver take it away (with the caveat that instead of McGrady we include Stackhouse, who was clearly the McGrady of his day ): "Iverson and Stackhouse do frustrate a lot of people with how much they control the ball...This is a problem for coaches, a lot of whom have been raised with the thought that more equal distribution of the ball is better, especially when their stars aren't very efficient...With stars using such a high percentage of a team's possessions, they allow other player to use fewer possessions, raising their offensive ratings. For example, if Iverson used 35 percent of the team's possessions, that leaves about 65 percent for the other four guys...This helps a player like Eric Snow...improve his rating several points from where he'd be if he had to use 20+ percent of Philly's possessions...Remember...how much better Philly's offense has been with Iverson than without him. They have lots of guys who do reasonably well if they use 15 to 20 percent of the possessions, but not if they use more than 20 percent."

This doesn't even approach the issue of double teams and passing and such - it just looks at player's roles. I feel like a lot of the time we rely on math explanations, where straight basketball examples are a lot easier to understand. So let's go back to our example of the offensive rebounding big man. Sure he can shoot 90% if he only shoots off of offensive rebounds, but that means all the other shots will be taken by his teammates. The key here is that someone has to use the possessions, otherwise it's a shot clock violation. That means staying in his role forces his teammates to use a large share of the possessions. If he has teammates who can score at relatively high rates every time down the floor, then that's great. But if his teammates are just jump shooters that are being defended tightly, then he has to step out of that role and start to take tougher shots that will decrease his shooting but still be better than a teammate shooting.

Ideally, though, he'd love to shoot 90%. So if we were building a team around this player, we'd stock it with players who could be relatively efficient with large numbers of possessions. The key word in that sentence is "relatively" since as we've noted above it takes a certain amount of skill to maintain efficiency at high levels of usage.

As Dean wrote, someone who can maintain average efficiency while using a huge number of his teams possessions allows his teammates to stay within their desired, high efficiency roles. I think his chapter covers this a bit better than I can, but I hope this all makes sense.

The trick is always determining what a player can and cannot do, and how to optimize the offense. It could very well be that when Iverson shoots 9/23 with 5 TOs that he is well below the level of efficiency that will help his team - that it would have been better if on a number of those shots he would have let Linas Kleiza have more of the offense. But the general idea is that when teamed with players who are very good within their roles and bad out of them (say a Chuck Hayes) it is usually better to have average or below average efficiency from Iverson than have whatever horrendous thing results from Chuck Hayes taking double teamed contested jumpers.

[Statman]

[Statman]

[Mike G]

[Harold Almonte]

Flint said:

[Flint]

Mike - That was a joke about the rebounding. I obviously dont think reboundng is important because I did a lot of it in high school. I think its important because it's one of the most valuable ways to contribute to winning a basketball game.

Ben - Nice post.

Statman.- I am not demeaning anyone. I understand how savvy people are around here. Mike G did however say:

"Putting the ball in the basket is how basketball games are won. Iverson is an alltime elite player in this most-important area."

I like the WOW partly as I have said before because it confirms previously held opinions, on players like Iverson, Allan Houston and Eddy Curry.

Berri has written a number of posts on Iverson, who has been one of his prime benefactors. I dont have time to respond right now, so I will give you back his responses, which are better than mine anyway.

This one is a response to the argument that because the Sixers are better with him on the floor he must be incredibly valuable.

http://dberri.wordpress.com/2006/07/25/an-allen-iverson-comment/

This one is an analysis of the Sixers teams that went to the Finals.

How Did Philly Get to the Finals in 2001
http://dberri.wordpress.com/2006/07/24/how-did-philadelphia-get-to-the-finals-in-2001/

It's not a Holy Grail, its not a Silver Bullet. I understand there are a lot of talented people here who have come up with wonderful ways of looking at Basketball. I am going to order Basketball on Paper later. All I am saying is that when I read those posts, they made a lot of sense to me.

[gabefarkas]

[Ben F.]

[jeffpotts77]