Comments on: SPM’s Most Improved Players

By: Kevin

Kevin — Sat, 16 May 2009 17:31:09 +0000

I agree with the oft-cited criticism of too many players winning this award because of increased minutes. But I also don't get why so many young players win the award. Players are supposed to make significant jumps between the ages of 20 to 25.

That's what Granger did. Durant and Devin Harris, too. Those were the top three vote-getters.

Turkoglu was a great choice last season. At 28, he had his best season and not because he hadn't reached his potential beforehand. It was because Stan Van Gundy put him in a new role, he accepted it and flourished in it.

The worst non-selection I can think of was Jason Williams in 2002-03. Hubie Brown became coach of the Grizzlies. Together, they turned around Williams' career. He went from the guy who threw the ball out of bounds to one of the best decision makers in the game -- 8.3 assists and 2.2 turnovers.

Gilbert Arenas won the award, going from 20 to 21 years old and going from playing 25 minutes a game to 35.

By: Jason J

Jason J — Fri, 15 May 2009 13:39:56 +0000

My point exactly (tho it contradicts my "pick" of Durant), and it's true of almost all the guys who were ever in the running this year. Before the injury, Devon Harris was out in front, but had he really improved dramatically or is this what happens when he gets the chance to be the #1 option on a team for the last couple of years? Granger has said himself that he moved his focus from defense to offense (said he patterned himself off Artest in the past). Does his becoming the focal point of the offense and taking it a bit easy on defense account for his improved play? Would anybody have noticed Rondo's increased confidence and aggressiveness if the Celtics had been a healthy team this year and Garnett had given his usual dominant performance night in and night out? Situation is critical and something I don't think we can untangle from performance. Maybe more than other categories this vote needs to have a lot of context factored in and less emphasis on empiric evidence. Or maybe not.

By: Neil Paine

Neil Paine — Fri, 15 May 2009 08:13:02 +0000

I guess that sort of depends on whether you thought Durant actually deserved the ROY last year... But what do we mean by "improved" anyway? Here we're talking improved stats, but seeing how the coaching/position change instantly made Durant's numbers so much better, it's possible he could have posted these numbers last year as well, had he been properly coached. Which in turn raises the question: how much did he really improve, and how much did his situation improve?

By: Dave

Dave — Fri, 15 May 2009 07:02:02 +0000

My question is about Durant for all you people who would give this award to him: how much better do we expect the ROY to get?

By: Neil Paine

Neil Paine — Thu, 14 May 2009 22:43:38 +0000

Villanueva actually had never been a positive player by SPM until this season; his career scores are -0.95 in '06, -1.89 in '07, -2.85 in '08, & 2.41 in '09. I think this year it's been about a big leap in scoring, assists, free throw attempts, and blocks (per minute) all in the same season, coupled with Milwaukee going from the absolute worst defense in the league to the 15th-best. Rondo's statistical development has been amazing, especially considering that his jumper is still shaky. Everybody's saying it, but it's true -- if (when?) he adds that aspect to his game, he creeps up on Deron Williams and Tony Parker as maybe the 2nd-best overall PG in the league behind Chris Paul.

By: Jason J

Jason J — Thu, 14 May 2009 21:03:53 +0000

I'm always a little unsure how to look at this award - is it for players who added skills or already skilled players who got more meaningful minutes for one reason or another?

With Charlie V. do you think his increased SPM is a result of him getting better or him getting more opportunities in the wake of team-wide injuries? Bogut and Redd missed significant time, and he picked up the slack very well, but he had some nice runs as far back as Toronto.

Durant we could see actually improved his game. His decision-making, confidence, and aggressiveness going to the basket were miles ahead of his rookie year. He would have been my pick too, though his big leap did coincide with him moving from guard to foward...

Rondo deserves some mention I think - not only did his numbers bump up, but his overall responsibilities increased, and he became the premier defensive point guard in the league - the All-D Team popularity contest that put Chris Paul ahead of him and Delonte, Dahntay, and Courtney Lee notwithstanding.

By: Guy F

Guy F — Thu, 14 May 2009 18:09:02 +0000

This is slightly off topic, but would it help your formula at all to use a variable like (%of games started) in your regression? That way you could take account of who a player tends to play against, whether it's the other teams first or second unit, and then reward or penalize their SPM score appropriately.

Also a stat like (%of win shares) might be helpful in taking account of who you're playing with.