In the summer lull of basketball news here are some stats that we could use to further speculate as to what our players are capable of. RAPM is regularized adjusted plus minus. In short, it's suppose to be the more accurate version of adjusted plus minus. Link: http://stats-for-the-nba.appspot.com/ It could be interesting to compare these numbers to see what improvements our players made from 2011/2012 and see how our new editions might look on the squad. For instance if you compare Dwight and Asik... Dwight's Def xRAPM is a full point higher than Asik's while playing 177 more minutes (and some/most of those minutes injured)... That's pretty impressive and bodes well for the future. Also here's a little definition of the stat, just for clarification: http://www.nbastuffer.com/component.../term,Regularized Adjusted Plus Minus (RAPM)/ The main website also provides their aging curve: http://stats-for-the-nba.appspot.com/
Check out Kendrick Perkins. I know RAPM is flawed, like all stats, but a lot of these pass the "eyeball test". The OKC roster, in particular.
Yea, while not all players will have accurate RAPMs, it is at the very least interesting to see a completely objective statistic relying only on plus/minus and not box score. And it definitely passes the eyeball test.
Imagine if you could some how fuse James Hardens Off xRAPM with Asik's Def xRAPM. Would be amazing / Lebron-like.
Looking through some of these older ones from the 90's, wow. Hakeem was ranked so high solely on defense. T-Mac had a 7.6 Offensive rating in that 02-03 season, incredible. Kirilenko was really, really solid with the numbers.
Keep in mind, though, that those 90s RAPMs are rough estimates based on an analysis of the boxscores, not the play-by-plays (which aren't available for those seasons). I'd take those ratings with a grain of salt.
To put RAPM calculations into more layman friendly terms, the process takes APM for a season and adjusts it using the weighted APM from the 3 previous seasons (or less if 3 seasons data is not available). It is an attempt to minimize offensive and defensive scores that are heavily influenced by other sources (read that as team mates). An example would be a player who typically is surrounded by team mates who heavily influence his APM. Like if a player only played with the four strongest defensive players on the Pacers roster. That player would show a very good defensive APM, but he might just be a lousy defender surrounded by really good defenders. Please note also that RAPM is just a better estimate of performance than APM, and should always be assumed to be an approximation of the correct value.
Off-topic, but check out 2005-2006's leader... purely defense! http://stats-for-the-nba.appspot.com/ratings/2006.html
<blockquote class="twitter-tweet"><p>Just posted: Houston Rockets projection by xRAPM- from 55 to 62(!) wins, depending on Dwight. <a href="http://t.co/NoIYkcBkN3">http://t.co/NoIYkcBkN3</a></p>— Kevin (@NBAcouchside) <a href="https://twitter.com/NBAcouchside/statuses/367309465608921088">August 13, 2013</a></blockquote> <script async src="//platform.twitter.com/widgets.js" charset="utf-8"></script> ^This guy is projecting records for next season using xRAPM, and according to his calculations the Rockets will win 55 games with last year's version of Dwight and 62 if he gets back to his Orlando form.
Pretty cool write up. Naturally you shouldn't base everything off of one stat, all stats have flaws, but it still shows some of our possibilities. What's especially interesting is what he says in his update at the bottom... When he increased Asik's minutes to account for them playing together for about 10 minutes a game, we go up to a 59 win team with old Dwight and a 65 win team with Superman Dwight.. Pretty crazy.
This is certainly not the only metric that projects the Rockets winning ~60 or more games next season. And most of them factor in only minimal or no improvement from the Rockets least experienced players. If 2 players out of Lin, Bev, Jones, and D-Mo are significantly improved (and all Lin has to do is maintain his post all-star game numbers) the Rockets could pick up 3-5 wins just from those players added contributions. However all predictions are predicated upon the incredible health of the Rockets and Howard last year. As with every team in the NBA, the Rockets are only a couple of tendons away from being average next season.
http://bbs.clutchfans.net/showpost.php?p=8169503&postcount=3 I made projection with about the same results based on RAPM: A lot hinges on Dwight's performance and minutes play by Asik and Dwight. Summary from my post two weeks ago: Same performance as last season just by adding Dwight of Lakers: MOV (Margin of Victory/game) +6.0 Wins projection: 55 wins. Considering that young players will improve, based on Influence of Player's Age by Engelmann. MOV: +7.2 Wins: 59 wins (Which is what OP's estimated) Dwight of 2010-11 with improved roster: MOV: +9.7 Wins: 65 wins
This stat doesn't pass the smell test, according to this stat, David Robinson was the best player in the NBA every year from 1990-1996.