[RAPM] How do the Rockets compare with other contenders?

Maybe you didn't reed my post. I think stats are a very useful tool when used in conjunction with actually watching the game. Your line of thinking is lazy since you you think you can completely analyze a game or a team with just a stats formula without actually watching the game.

 

What should we watch for that is more important than winning championships? For if that is the ultimate goal, nothing else should take precedence in our evaluation of the players, right? (note, I don't actually believe this, but that is apparently where your logic leads us)

 

Can you give me some examples of you watching the game allowing you to be smarter than RAPM when it comes to judging teams and players?

 

I've said it multiple times. No matter how you try to twist my words my answer stays the same. Stats are a useful tool to be used in conjunction with watching the game.

All you need to know is that WARP has Detroit is the number 2 front court in the NBA. Teams like Atlanta, Dallas and Portland are all ranked in the top 10 and all higher than Indy, Memphis and Houston. That should be all the proof anyone needs about how useless those stats are with out actually watching the game.

 

Ginobli has a RAPM of 3.7. Did you actually watch him play last season. His defense has not just slipped but but he was a liability at times, he can't stay healthy and for every really good game he has, he has three more that are just really bad. By contrast Chandler Parsons is a -0.1, but Parsons was better and more consistent than Ginobli.

I wouldn't have known it if I hadn't seen actually watched the games.

 

I don't think people are disputing that, but rather this:

"At the end of the day the only stat that really counts is wins and losses."

 

My meaning is that that there have always been players that filled up the stat sheet but it doesn't always translate into wins. Winning is why we play the game. Which players have the best stats are a side bar to the real reason we play, to win.

Antoine Walker is a perfect example. He had some monster stats years in Boston when teamed with Paul Pierce but it just never really translated into wins. What's worse is that they were usually a sub .500 team at a time when the East was pathetically weak.

If you just look at the stats you would think Walkler was a great team player with all of the assists. But if you watched him play it was painfully obvious that he was a ball hog and the very definition of fools gold.

 

I'm not going to argue with you over semantics. Tools.... blah blah blah, information blah blah blah.

 

Wins and losses are how you evaluate teams, not individual players. Basketball games are the sum of the contributions of up to 30 individuals, and it's not possible to evaluate a single player's impact based upon the teams' wins/losses alone. This is where stats come in.

 

Almost every advanced stat worth its salt will tell you that Antoine Walker was a negative impact player who actually hurt his team despite his high PPG totals. His stats weren't monster by any stretch of the imagination: he was an average rebounder, poor passer, and a poor defender and he turned the ball over a ton. Even when he was scoring 20+ points per game, he was so inefficient at it that he was hurting his team every time he shot the ball.

Stats will tell you how bad a player Antoine Walker was, but you have to be willing to look past PPG and look at the stats in their totality.

 

If anything, you just brought up a perfect example as to why RAPM is great 'tool,' as you call it, to measure contribution of wins by a player.

NBA started to take +/- in year 2000. Starting that year, which was for accuracy reason and also at Walker's prime year, his RAPM was +1.8 for three straight years. For a PF, and you are perfectly correct on this, he WAS a fools gold. His contribution to wins was only a few games.

People don't need to watch him play for his whole career and STILL able to confirm what type of player Antoine Walker was. Isn't RAPM wonderful!?

 

You do realize that adjusted +/- is not based on individual statistics in a boxscore, but rather on how the team performs with the player on the floor?

Its true that superficially impressive individual stats can mask the negatives of a player. But team record can also be misleading in evaluating players (you mentioned Walker, so I'll go ahead and mention KG). The solution isn't to simply say "screw it, just got to watch the games". That helps, but only to a degree.

We should look for better ways to quantify what's going on, to better capture how players contribute to winning while taking into account context. That's what "advanced" stats are all about.

 

Morey has made it clear on many occasions how much he relies on stats vs. the eyeball test, especially as it relates to defense. If you wish to disagree with him, knock yourself out.

 

So did you watch every Spurs game and every Rockets game last season?

 

Watched the games with both. Hakeem is obviously better for people that don't put too much emphasis on wins and losses for individual player performance. RAPM has its weakness in that it takes a lot of data before it is relevant. This weakness means it has a hard time judging rapid decline. It also has a hard time with young players.

It usually does a good job on the guys that affect wins and losses for good teams.

 

Distinction without a difference.

 

so you basically saying when win/lose theory been proven fool you deter to watch the game. and if watch the game been proven fool where would you go? back to win/lose or what?