APBRmetrics

Neil Paine wrote:

Mike G wrote:a team that goes from -5 PPG to -2 PPG, with one player substitution, has found a player with positive impact. How could it be otherwise?

That would only be true if the player being subbed out was precisely average. In J.E.'s example, LeBron/Wade/Dwight/CP3 are all +5 players, so even if the 5th guy was average they should be winning by 20 pts/100 poss. If they're losing by 5 pts/100 instead, the 5th guy is, by definition, a -25 player. You sub me in and we only lose by 2 pts/100, I suppose you could say that was a positive substitution, but I'm still a -22 player. Even if you compare me to NBA replacement level (something like -3.35) instead of average, it's quite inaccurate to say I am a player with "positive impact" in any context-neutral sense.

Dang. I should have asked for a "non-absurd" example.
Assuming the league didn't punish a team by stripping away all their players but 4 superstars and limiting their remaining roster to guys who have never played competitively -- and that we are talking about replacement-level and better players filling the lineup of a team -- then your 5th-best player is not a -25 or a -22 guy.

Why would that 5th player have to be precisely average? If you can get a #5 guy who makes the team 3 points better, that's a positive impact.
Lots of teams do not have 5 average-and-better players. Not sure any team has 5 such starters.
Every team, of necessity, gives minutes to below-avg guys.

Marreese Speights is pretty terrible for a guy that averages that many minutes on a (almost?) contender. Really low basketball IQ, once tried to grab a ball that would've went out of bounds off an opponent, then fumbled that ball; also very bad shot selection, and lots of loose ball fouls (likes to push opponents in the back alot when rebounding)

They're at least close to contending, it seems. Gamble on KMart and/or Chris Andersen? (these two should really pay me money for all the advertising I do)

What is up with Ty Lawson? He has looked average, at best, in the games I've seen. From just watching the games he seemed less aggressive which made me assume that his overall FGA and FTA numbers were down, but over the entire season they're almost the same; it's just that his %s are waaay down across the board. He's shooting 38/23/54. Effects of the contract extension?
(That's something I might look into with RAPM)
Denver playing a slower pace (9th vs 2nd last year) could explain some of the FG%/3% drop, but with him shooting 54% from FT also, he just seems to be in a major shooting slump.
They had lots of away games, and Denver really plays alot better at home; maybe that'll help with %s and pace a bit.
There's really no reason for them running less than last season, now that they have Iguodala, is there?

Ty Lawson's sudden inability to shoot has me questioning many things.

Any thoughts on the Thunder?

8-3 is nice but it has been a fairly easy schedule to date.

Lots of lineups better than the starting lineup. I found that recent champs average the 4th best lineup amongst high minute lineups in the league on net rating. For lineups over 50 minutes, the Thunder's best is in 18th place currently. The starting lineup with 2.5 times the minutes is half as good and in 21st place. http://bkref.com/tiny/cu2ua

NYK and LAK are in the top 4 curently. Miami's best is 8th right now.