Warriors +- Report Card

Post #1 » by **floppymoose** » Thu Jan 3, 2008 6:27 pm

The Warriors are 40% of the way through the regular season, so let's take a look at the +- report card.

In first place by a wide margin, your team MVP, Baron Davis at +19.8 (!!!)
In second is Action Stephen Jackson at a gaudy +11.5
In third place is my most improved candidate, Al Harrington at +9.5
Next is Monta Ellis at +6.6, and I expect him to finish higher still after his slow start.
Then the last of the players with significant minutes and a positive +- is Andris Biedrins at +2.8. He's not doing as well as I'd hoped but he has a positive influence on a winning team so he's still doing pretty well.

The remaining players with significant minutes are all in negative territory:
Barnes -5.7
Pietrus -6.9
Azubuike -11.9

Looking at who is subbing for who, the short take is:
Baron >>> everyone
Stephen+Al >>> Kelenna+Matt+Trainwreck

Sid the Squid · Post #2 » by **Sid the Squid** » Thu Jan 3, 2008 6:47 pm

All the warriors have been pretty consistant with their +/- performace as the year winds along, except for Matt...His numbers are like a friggen rollercoaster...He's easily our most inconsistant player...When Matt is on on any certain night, he's really good..That could last through 3 or 4 games in a row..Then he nose dives and plays through a horrible stretch of ball.

Post #3 » by **FNQ** » Thu Jan 3, 2008 7:10 pm

+/- is useless... its so random sometimes for one game or so... useless stat, SPG/BPG and blind eye judgement is much better.

St.Nick · Post #4 » by **St.Nick** » Thu Jan 3, 2008 7:41 pm

What is blind eye judgment?....

And I agree with LF about Barnes. The guy plays pretty badly most of the time but can surprise with a big play or even a big night every so often.

I think he's a good guy to have for $2M a year or so, but he is not a championship level role player for 82+games a year.

turk3d · Post #5 » by **turk3d** » Thu Jan 3, 2008 7:44 pm

St.Nick wrote:What is blind eye judgment?....

And I agree with LF about Barnes. The guy plays pretty badly most of the time but can surprise with a big play or even a big night every so often.

I think he's a good guy to have for $2M a year or so, but he is not a championship level role player for 82+games a year.

Actually, he's a great guy to have on your bench who can fill in if a starter goes down (which he demonstrated last year). Never should be a starter (imo) under non-emergency circumstances (are you listening Nellie? No, of course not).

Souvlaki · Post #6 » by **Souvlaki** » Thu Jan 3, 2008 9:26 pm

Wow, you mean there's finally a stat out there than can tell me that the players who play the most minutes on a team contribute the most to either winning or losing? Amazing!

+/- is about as useless as that stupid historical win percentage thing they did last year. +/- from last night tells us that Jackson was the worst player on the floor. Every single night that stat can be shown to be garbage.

MightyReds2020 · Post #7 » by **MightyReds2020** » Thu Jan 3, 2008 9:46 pm

+/- isn't totally useless. It just has to be used within context. For instance, Baron's pretty much tells the story, that he's our most important player. Buke's is a little skewed consider the major minutes he logged when SJax was suspended and the Warriors couldn't find a win.

Post #8 » by **floppymoose** » Thu Jan 3, 2008 9:50 pm

No one is talking about one night's worth of data here. Souvs, you don't appear to understand what this is about.

-bob- · Post #9 » by **-bob-** » Fri Jan 4, 2008 2:40 am

how about the PER stat:

Baron Davis PER: 20.95
Andris Biedrins PER: 19.48
Al Harrington PER: 17.09
Monta Ellis PER: 17.00
Stephen Jackson PER: 15.17
Patrick O'Bryant PER: 14.67
Kelenna Azubuike PER: 13.73
Austin Croshere PER: 13.38
Matt Barnes PER: 11.87
Mickael Pietrus PER: 11.66
Brandan Wright PER: 11.61
DJ Mbenga PER 6.83
Troy Hudson PER: 5.41
Marco Belinelli PER: 1.25

acidicality · Post #10 » by **acidicality** » Fri Jan 4, 2008 4:11 am

yea the +/- thing sometimes is meaningless...but especially in close games i think it does mean something.

Post #11 » by **Sleepy51** » Fri Jan 4, 2008 4:26 am

acidicality wrote:yea the +/- thing sometimes is meaningless...but especially in close games i think it does mean something.

That's not the point at all.

It means nothing about a player in any single game. It requires additional data from contrasting circumstances in proportion to the frequency of those circumstances. That's why it gets more meaningful over more games. Abberrations get dilutted by their recurring frequency. If you look at a single game whether blowout or nailbiter, you could still be watching a freak performance. The score couldn't be more irrelevant.

Post #12 » by **FNQ** » Fri Jan 4, 2008 4:48 am

Patterns are the key... inconsistent players (re: Barnes) have inconsistent +/-s, consistently

Some players, (Baron / Jax) obviously make our team better and usually grade out in the +/-... other players (Pietrus/Buke) do not make our team better and that has been consistently shown as well...

The hard point is deciding where the line in the sand is.... you can make a case for Biedrins being the line, or Monta (the average line)... but if you go by strict +/- #s, you wont learn anything...

Basically you have to figure the Warriors are overall in the +, because of our W/L record... so a +3 per game player on the Warriors is a lot less impressive than a +3 player on the TWolves...

mait · Post #13 » by **mait** » Fri Jan 4, 2008 8:22 am

wov this PER stats shows Patrick O'Bryant as our best player off the bech being 6th in this category after our starters. Maybe we made the decision to cut him way too early.

Post #14 » by **floppymoose** » Fri Jan 4, 2008 11:09 am

Souvs says:

you mean there's finally a stat out there than can tell me that the players who play the most minutes on a team contribute the most to either winning or losing? Amazing!

This comment really sums up the misunderstanding. It is understandable that someone would think the players who play the most on a bad team must be too blame, but it is not always true.

Two seasons ago the Warriors were a bad team. But they had 4 players who had such good +-, that the Warriors won the minutes these players were on the court: Richardson, Davis, Biedrins, Ellis. Two of those 4 clearly were among those playing the most minutes. The other two were young and improving, and could reasonably be expected to play major minutes in the future.

By Souvs reasoning, the Warriors suck so Baron and Richardson suck, and any team where they play big minutes is going to keep on sucking. But if you believe the +- data has meaning, you could take those four players and add in some players who fit better, and suddenly the team would be a winning team.

If only we could run some kind of experiment to test this theory.

oh wait we did.