Lenovo stats yield interesting results for our Rockets

It's a spin-off, IBM sold it's PC business to Lenovo.

 

So you guys are trying to say that the blowout is skewing the numbers for the first 5 games. Is it realistic to think we will have a blowout once every 5 games? I think so. Hopefully we stay around these numbers. 55 wins would be very realistic.

 

well we are gonna get blown out sometimes too. so yeah we need to wait.

 

This is what I'm talking about - a losing culture. It's the mark of mediocrity -- you expect blowout wins to be rare, and you expect to be blown out every now and then.

This is not a mediocre team, people. There is nothing wrong with those stats. This is NBA basketball, nobody's getting any "unfair" advantages. It's one thing to have tempered expectations of the team, but while we do that we shouldn't underestimate just how good this team can be.

 

Clearly, Hayes and Alston are not the best two players in the league, as these stats would seem to indicate. They aren't even close. They aren't even the best two players on the Rockets. If you were a GM and you had 2 picks to set up a new franchise, you would pick Hayes and Alston?

Alston IS NOT our most valuable offensive player. Alston just had the game of his life against Dallas. That is why his +/- numbers are so high.

Read the previous post about sample size. Unless the Rockets blow out opponents by 31 points a total of 17 TIMES this year, those stats are going to average out. Desperately trying to believe we are the best team in the league when we are 3-2 and have played AVERAGE basketball so far is not very useful.

 

This isn't about fair, unfair, or basketball. This is about statistical methods.

 

interesting but unreliable.

 

I don't think you've been reading (or watching). Alston's +/- numbers are not high merely because of his great game against Dallas (again, why exactly do the numbers in this game not count?). His numbers are high because he's doing a great job of running the offense and setting up plays. Next time you actually watch the game, pay attention to how efficient the offense is when Alston is on the court and when he isn't. These stats aren't about who the best players in the league are (if you're looking for that, go read Hollinger's PER articles). What these stats tell you is how much of a positive impact each player is having on his team. (In case you didn't know this before, Chuck Hayes had one of the highest +/- ratings in the league last season at +10.3 - even much higher than Yao's +7.3).

I see no reason why we can't get 15-17 blowouts of at least 20 points this year. I'd like to hear your reasons for thinking this is unreasonable. This team obviously hasn't played its best basketball consistently yet. The stats may not be comprehensively conclusive, but there's really no reason to dismiss them -- especially since they *do* reflect what we're seeing on the court.

 

You guys arguing about the blowout are just spinning your wheels. If you take it out of the equation, by selecting only road games, we're still easily #1, and the margin is actually larger.

 

What is Lonovo ?

 

Is that because Lonovo is a Chinese company? and so they biased for Rockets ?

 

I'll have to take the opposite side of this argument. Dallas has so far been a paper tiger this year, everyone is beating them. A 4-1 record and a victory against the Jazz or Nok would have made me far more sanguine about our being a top echelon team. Most encouraging to me is that we are 3-2 without Tmac performing at his best and if we can beat some quality opponents like the Spurs, Clips, Lakers and Cavs I'll buy the "we are that good" claim.

 

These Lenovo stats by that chinese company are rigged, because they love the Rockets.

There is no way that Rockets win all the combinations from 5 -man to 1-man, and lose to the Utah Jazz and New Orleans Hornets.

If we had the best 5-man team, we would be in 1st place.

Yao has the hookup on making us look good in meaningless stats by a chinese company.

 

Rafer Alston +/- for 5 games....

UTAH : -4
DAL : +34
NOK : -6
MEM : +7
MIL : +11

If you don't understand the impact of the one of the leagues' three blowouts thus far into the season, it's not statistics or math you're having trouble with...it's basic arithmatic.

 

It's Lenovo. original name is Legend. It's famous in China. Bought ThinkPad(which I'm now using) from IBM(which I'm now working for) .

 

I hesitate to read too much into these stats so early in the season, but I do think it's interesting that Chuck Hayes had the highest single player +/- on the team last year, despite not playing a lot of minutes (this stat is raw total, not minutes adjusted, so it's possible that if he got more minutes his raw +/- could have been much higher):

Single Player +/- for 05/06 season

Last year he benefited from playing against weaker competition. This year, he's play against much better competition, but he's also playing with better teammates.

 

But he was also a significant reason for that blowout. Just as the blowout signifies that the Rockets are probably much better than their 3-2 record, similarly I think it should signify that Rafer is better than his performance in just the other games. You can't throw it out altogether just because it looks like an outlier.

 

Doctor Robert, thanks for the infusion of sanity.


Many of you guys are missing the point of +/- stats. How a team fares when a player or group of players is off the floor is equally important to how the team fares with them on the court. It is the differential between these two that makes the stat.

Therefore, if a stud player does things on court to help the team win (helping to score points or keeping the opp team from scoring), but the team does just fine without him when he's on the bench, he won't have a good +/- stat.

How poorly the team fares without the player is equally as important to how the team fares with him playing. This is a shift from our normal statistical thinking with things such as PER; we wonder how well the guy plays when he plays, period. But the +/- minus stat answers the question how much does the team miss the player when he's gone? Normally we see this kind of analysis when a player is injured, and see how the team does without him (plenty of that talk for us last year, sadly.) The +/- stat analyzes the question on a per-game basis.

If a guy plays well on his team and his backup totally sucks, that will boost his +/- stat. The team performs less well without him because of who is on the floor when he is out. This is just one example- another would be a team that does well because he can score, but does even better when he's on the bench because of chemistry issues, for example.

The complete dominance by the Rockets of this entire statistical category from 1 man to 5 man could show that we truly have an excellent lineup of Yao, Tmac, Hayes, Battier, and Alston that can really score and really defend- and when any of them leave, the dropoff in bench performance is significant, and those 5 players are dearly missed.


In any case, I agree that the sample size is way too small. We should look again in 10 or 20 games. I think we have a decent bench with Snyder and Head.

 

But in the case of these Lenovo stats, Off-court data isn't considered. It's purely look at the team's +/- while the player(s) are on the court.

 

Chuck Hayes became a starter at the Mavericks game.

His plus-minus is going to be higher than the other starters(Yao, Battier, Alston, McGrady), because he doesn't have the Utah game to pull down his plus-minus rating.

He only played 8 minutes in that game. Then again, maybe we would have won that game if we had him as the starter. He does bring along of energy to kick them off with a bang in the first quarter.