The Bratwurst - Milwaukee Bucks Blog

(IPM is available here. This article references the file “IPM-DCS Combined Rankings”)During the past several years I have developed a player ranking system that attempts to measure and compare the relative value of NBA players. This system , which I call IPM (Impacts Per Minute) attempts to equate the value of each different statistical category with each other and come up with a single number that can be used to compare one player to another. The system (and results) is similar to John Hollinger’s PER. Recently I found this article at www.82games.com and have been able to use it to add defensive rankings to my system, making IPM a somewhat unique product.

By taking the defensive rankings in the DCS article, adjusting them for team factors, and converting them to IPM scores I was able to come up with an “overall” ranking (IPM) and a “defensive” ranking. By combining the two (80% IPM and 20% Defense) I arrive at a single number that I feel gives a good representation of how a player performed vs. his peers.

With a few exceptions I feel that the rankings give a great picture of player performance.

One problem is that because I am giving per-minute numbers, the fewer minutes a player plays then the less accurate they become. It’s not very realistic to assume that because a guy played 8 min a game at a high level then he would become a superstar at 35 mpg. So those players’ rankings have to be taken with a grain of salt. There are a few cases where a players’ teammates might be affecting the defensive rankings a little as well , guards who play with dominant big men seem to have their rankings boosted, for example. But overall I feel that IPM gives a great starting point when comparing one player to another.

The full rankings, along with more detailed descriptions of how the calculations were made are available here. A typical superstar ranking is 1.0 IPM, All-Star is around .85, and solid starters are above .75.

Applying IPM to analyze some of the major additions that some teams have made:

Charlotte - added Jason Richardson:

When I first heard about this move I was skeptical, but now I think that Michael Jordan may have seriously improved the team here. Richardson’s .7758 combined IPM only ranked #32 against other guards but J-Rich had to come back from a broken hand, which probably slowed him somewhat. However, Richardson is a pretty adept defensive player , something that Golden State actually had in abundance last year, which is surprising given how bad their team defense was , and he should fit in well as a replacement for the undersized Brevin Knight and limited Derek Anderson.

Most importantly, though, the addition of Richardson and resigning of Gerald Wallace pushes Adam Morrison deep down the bench. Morrison was one of the worst players in all of basketball last year , a low percentage shooter, poor rebounder and passer, and lousy defender. It’s no coincidence that the Bobcats’ strong finish in 06-07 coincided with Morrison sitting out the last 10 games. Richardson is easily a 30% improvement over Morrison and with J-Rich in and Morrison out then Wallace can concentrate on his natural small forward position.

I feel that Charlotte is one player away from potentially making a huge leap in the Eastern Conference and that player is Anderson Varejao. Power forward is a question mark for Charlotte, and while Sean May had a great season last year (.8953 IPM) it came in only 35 games. Charlotte is currently about $8 million under the salary cap (depending on Matt Carroll’s contract). They have a great opportunity to offer Varejao (with a defensive IPM of .9331, he was the 8^th best defensive forward last season) a contract along the lines of 3 yrs/$27 million that Cleveland, who is right against the luxury tax line, would be hard pressed to match.

A starting lineup of Okafor, Varejao, Wallace, Richardson and Felton would be a balanced, talented squad ready to make noise in the Eastern Conference right now.

<Orlando , added Rashard Lewis, lost Darko Milicic:

It’s obvious that Lewis is better than Milicic (.8537 IPM vs. .7577), but the problem is that high scoring, below-average defenders are easier to find than average centers. The combination of Ariza and Turkoglu is only a little worse than Lewis but replacing Milicic with Adonal Foyle is a huge step down. Lewis simply isn’t worth max money , he was only the 26^th best forward in the game last year, and while he’s not old he’s not really going to continue improving much. I guess that the logic was that since Howard plays center almost exclusively now that Milicic was never going to be fully utilized as a power forward for the Magic so letting him walk makes some sense. But I feel that pursuing a trade for Corey Magette instead of Lewis could have resulted in an equal player for a much smaller commitment, and could still have left more flexibility to add a power forward.

Boston , added Kevin Garnett and Ray Allen:

At this point in his career, Ray Allen is probably somewhat overrated. While still putting up All-Star numbers offensively (.8780 IPM) his defense is poor enough that his offense has to be on fire to make up the difference. Kevin Garnett, however, is another story , he’s still at the top of his game, ranking as the #2 player overall (1.102 IPM) and #1 defender (.9515 defensive IPM) in the entire NBA. Boston gave up a pretty nice player in Al Jefferson to get Garnett but the difference is that while Garnett is about 20% better than Jefferson; his defensive prowess can also carry over into making defense easier for his teammates. And his big-name teammates need the help. Allen was one of the worst defenders on Seattle last year and Paul Pierce was quietly atrocious at the defensive end as well. Kendick Perkins and Rajon Rondo had very good defensive IPMs last year, though, leading me to believe that the presence of Garnett will create a lineup with the talent to be an exceptional defensive unit next year. Whether or not Doc Rivers’ coaching style allows that to happen, though, will be another story (as Golden State showed last season , it’s possible to have a very bad defensive team with very good defensive players).

Bottom line is that Boston will have such a thin bench that they will need an exceptionally injury-free season, but if they get it then this team will have a very high ceiling.

New York , added Zach Randolph:

I just can’t figure out how this is supposed to work. Randolph and Eddy Curry have such similar games , demand a lot of touches, good at getting easy shots, shoot at a high percentage, not very quick, don’t block any shots or get any steals, slow getting back on defense. They are both the sort of players that make the game very difficult for their teammates as they both need other players to work hard to get them the ball where they want it and also don’t offer any help on defense.

Randolph’s offensive IPM ranked him as the #8 forward but his defense was so bad that in knocks his overall ranking down 10 spots (to .9276). But here’s the thing about Randolph’s supposed “bust out season” last year , it really wasn’t any better than he has been in the past.

Sure, he set a career high in scoring at 23.6 per game but while his shooting percentage (46.7%) was an improvement over the previous two years , when he was hampered by knee problems , it was below his 50% shooting from 2002-04. His rebounding also simply returned to pre-injury levels. He does deserve some credit for returning from his knee surgery so well, but the real difference for Randolph this year was that he got tons of shots. As the only offensive option on his team he touched the ball on nearly every possession, getting 3 more shots and 3 more free throws per game than he ever had before. That explains the increase in his scoring right there. And that is an issue when you consider that by going to the Knicks there simply aren’t going to be as many touches or shots available to him since Curry needs his, and Marbury and Crawford take plenty of their own.

Defensively, the Knicks are going to have the most pathetic frontcourt in the league, as both Curry (.6065 defensively) and Randolph (.6519) are quite subpar. They also already have the worst defensive backcourt as Marbury (.6038) and Crawford (.6218) rate as two of the worst defenders in the entire league.

There are other problems with Randolph, namely the character and entourage issues. From legal issues to embarrassing behavior around Randolph with Portland, his entourage , which calls themselves the “Hoop Family” , has been an even bigger problem. Apparently he is a good guy from a tough upbringing who is surrounded by some bad people, but he’s supposed to perform in the media capital of the world now , and they are already wondering where he is.

Oh, and his presence takes minutes away from David Lee, who was the Knicks best player by quite a bit last season.

I just don’t think that Randolph on the Knicks solves more problems than it creates.

And that sums up how I feel about the major player additions to the Eastern Conference. Feel free to look at my data and draw your own conclusions.