RealGM

Soooo, anyone else remember the TrueHoop mini-debacle over Kevin Durant's poor plus/minus numbers last year?

I asked Winston if he'd advise his team to accept if the Mavericks were (in some alternate universe) offered Durant for free. "I'd say probably not," he replied. "I would not sign the guy. It's simply not inevitable that he'll make mid-career strides. Some guys do. But many don't, and he'd have to improve a lot to help a team."

Surprise, surprise, guess who's leading his team in plus/minus stats this year (scrubs excluded)?

http://basketballvalue.com/teamplayers. ... 0&team=OKC
http://www.82games.com/0910/0910OKC.HTM

So what does Winston have to say about all this?

http://waynewinston.com/wordpress/

Our numbers indicate that KD is an all star caliber player this year. Could this be because Henry Abbott in his gutsy TRUE HOOP column pointed out KD’s shortcomings (failing to play the pick and roll correctly and shooting when doubled and tripled teamed)? Maybe KD corrected these flaws in his game and this led to his improvement.

Reading +/- numbers like they're black and white isn't really a good idea.

LMAO at that original quote.

+/- is a joke

This is where academics like Winston fail.

Putting blind faith in a model is a recipe for disaster. I run into this all the time as an engineer -- you should constantly be questioning anything that is empirical; picking holes; looking for applicability limits, etc.

You don't make such pyschotic line in the sand statements such as "I wouldn't sign KD" based off on 1 model.

He made himself look like a tool. FYI Dallas no longer employs him as his contract recently ran out.

The reality behind the KD crappy A+/- numbers last year is this --- he got injured for a 6 game stretch last year and the team managed to somehow go 3-3. The +/- of everyone BUT Durant improved a lot during those 6 games, b/c overall that 6 game sample was way better than the rest of their season.

But we see this happen all the time - the first time the star player goes down often a team finds a way to hang around for a few games without him as players get a chance at a bigger role. The fact that they went 3-3 without Durant is not significant enough to say they're better off without him. Yet that's exactly the logic he's using when he sticks to APM so blindly.

The other thing you're going to find is the +/- data needs massive sample sizes ---- You already see websites using rolling 2 year averages for the outputs. I could see the sample sizes only getting bigger to help dampen the variance and make the results come out looking more "right" based on intuition.

based on stats, zach randolph is an equal player to bosh...but in reality, he's not....not even close....

stats only tell part of the story....

I'm a guy that actually likes detailed stats and think they have value. But some academics definitely get tunnel vision in their rush to apply mathematical rigour. I compare plus/minus to IQ tests - they don't actually measure IQ, they measure how well you write IQ tests! Similarly, it's indisputable that +/- measures points for/against when player X is on the court. However, it doesn't tell you WHY, and that's the crucial ingredient missing when jumping to conclusions about whether the player X is actually good or not.

Anybody who's studied economics will have encountered the same thing. Economists love to reduce everything to neat little, elegant mathematical theorems, especially in game theory and macro. It makes for good journal papers because they get to use axioms and be logically robust. They conveniently ignore the fact that people act illogically a lot of the time! So you end up with these perfectly sound "thought experiment" theories that don't reflect reality.

Indiana Jones wrote:based on stats, zach randolph is an equal player to bosh...but in reality, he's not....not even close....

stats only tell part of the story....

well..if you look at eFG%, FG%, defensive stats (opponent PER/OFF or ON court numbers..whatever they use for that stuff), rebounding rates..etc. I'm sure you can find a ton of differences that would separate them.

Stats never lie..it's just the dumbasses who misuse them.

Volcano wrote:Reading +/- numbers like they're black and white isn't really a good idea.

Exactly, it's idiotic to use just one single measure in basketball.

D-Wins-RingsIMO wrote:The reality behind the KD crappy A+/- numbers last year is this --- he got injured for a 6 game stretch last year and the team managed to somehow go 3-3. The +/- of everyone BUT Durant improved a lot during those 6 games, b/c overall that 6 game sample was way better than the rest of their season.

It wasn't the 6 games. The rest of the year was also bad for him for whatever reason. I don't see why anyone would care, but his +/- wasn't screwed for the year by that 6 game stretch.

you can sway stats to your opinion

a player can simply get a lay up
or miss 2-3times and score and get fouled

one gets 2 points
the other gets 2 rebounds and 3 pts

I still think Durant is a flawed, yet talented player, so there is definitely something behind his poor +/-.

He is exactly how I feel about Bosh sometimes. Extremely talented scorer, but doesn't always bring the intangibles that help you win games (ie. Defence). Players can run around, get points & rebounds all game, but not play within the system & help the team win. It comes down to B-Ball IQ.

Unfortunately, it's the role players on teams many times have the highest B-Ball IQ. They have to understand how to play well within the system, or they won't get any playing time. Talented scorers on the other hand are given a free pass at times because they know they will be on the floor regardless & there is not as much consequence when they don't do "the little things".

It takes special players like Kobe that either learned the game the proper way growing up, or make themselves accountable because they want to be the best all around player they can be.

This is why I am glad that we have been cautious & patient with DeRozans development.

A big part of the argument about Durant is that he doesn't make his teammates better, that the team is better when he's not on the court. Last night, he had half his team's points. He played a great game statistically. But only Westbrook barely made it to double digits in scoring and no other Thunder player had more than seven points!

Basically, Boston said, okay KD, you go get yours, we'll make sure no one else scores, and we'll kick your butt.

They did.

dagger wrote:A big part of the argument about Durant is that he doesn't make his teammates better, that the team is better when he's not on the court. Last night, he had half his team's points. He played a great game statistically. But only Westbrook barely made it to double digits in scoring and no other Thunder player had more than seven points!

Basically, Boston said, okay KD, you go get yours, we'll make sure no one else scores, and we'll kick your butt.

They did.

This can be said for a lot of players. The key is for the other guys to be good enough that Boston or whatever team can't shut them down completely.

There are some plays stats can't quantify, for everything else there's 82games.com

Wow this guy is a joke and should be stripped of any credibility he might have.

plus minus is a laughable stat... imagine if you got amir jonhson and bargs playing the D...

MEDIC wrote:I still think Durant is a flawed, yet talented player, so there is definitely something behind his poor +/-.

He is exactly how I feel about Bosh sometimes. Extremely talented scorer, but doesn't always bring the intangibles that help you win games (ie. Defence). Players can run around, get points & rebounds all game, but not play within the system & help the team win. It comes down to B-Ball IQ.

Unfortunately, it's the role players on teams many times have the highest B-Ball IQ. They have to understand how to play well within the system, or they won't get any playing time. Talented scorers on the other hand are given a free pass at times because they know they will be on the floor regardless & there is not as much consequence when they don't do "the little things".

It takes special players like Kobe that either learned the game the proper way growing up, or make themselves accountable because they want to be the best all around player they can be.

This is why I am glad that we have been cautious & patient with DeRozans development.

Kevin Durant has weaknesses, not "flaws". 26 year old allstars who score a ton of points but can't lead their team to a winning season are flawed. Kevin Durant
is a 21 year old beanpole and is already doing things that once highly touted prospects will never accomplish in the NBA. Just last year, and the year before, his team had a horrible record. This year some are projecting them to steal the 8th spot in the West.

Kevin Durant IMO is Lebron James minus the global hype (and the Godly athleticism). He's all basketball. He's already responded to those +/- based attacks by leading his young team to a 10-9 record. Doubt him if you want, but you better be wearin clean socks with all the time your foots gonna be in your mouth.

+/- is fascinating since it catches a lot of intangibles, but it's really not an exact science as there's a ton of noise. I usually find people use it mainly when the numbers agree with their preconceived notion and discard when the numbers don't match the (expected) results. I still find the results interesting, especially for the lineups, but for individual players I don't know if it's worth much.

First to give it some meaningful significance, the numbers gotta be at least twice the standard error (using the good old 95% normal distribution rule of thumb thingy, although my memory in stats 1 is hazy) And even when the numbers meet that rule, the year-to-year variation is a bit much.

According to basketballvalue.com Durant's two year +/- was -7.4 with a standard error of 3 so it was significant...yet he turned it around to +9.6 this season. Although with a SE of 11 it's a still bit iffy.

As for the Raps, Bosh has been the best player for the past few seasons, but last year, Parker was clearly the best with a +9 rating (4.5 SE, hey it's statistically significant!) However the year before though, Parker was -4.48 (5.3 SE)

See what I mean by "noise"?

+/- is and always has been a flawed stat for assessment of generalized individual play. It CAN answer questions like :did the team play better tonight with X on the floor? But it doesn't prove causation, and especially on teams like OKC, where it's built around one significant young star, you will often see the guy have a poor +/- despite obviously being the best player on their team because other teams, as a rule, check in and out their best guys when OKC does Durant. I recall Chris Paul while playing insane good in his first or second year, and being head and shoulders above anyone the Hornets had, being a -. It wasn't a reflection of Paul so much as a reflection of how much teams who played the Hornets recognized him as all they had, and basically went to their benches a lot more when he sat down.

Ari_Emanuel wrote:Kevin Durant has weaknesses, not "flaws". 26 year old allstars who score a ton of points but can't lead their team to a winning season are flawed. Kevin Durant
is a 21 year old beanpole and is already doing things that once highly touted prospects will never accomplish in the NBA. Just last year, and the year before, his team had a horrible record. This year some are projecting them to steal the 8th spot in the West.

Kevin Durant IMO is Lebron James minus the global hype (and the Godly athleticism). He's all basketball. He's already responded to those +/- based attacks by leading his young team to a 10-9 record. Doubt him if you want, but you better be wearin clean socks with all the time your foots gonna be in your mouth.

I wasn't trying to say thet KD couldn't improve/ change. I was just basing my opinion on what I have seen so far.

It just drives me nuts when people fall in love with a player that is all flash & no substance. Last season, some people were annointing him one of the best players in the league. Yeah, he could score, but he had a long ways to go.