Lenovo stats yield interesting results for our Rockets

I felt Hayes was the answer to Boozer in that game. Can't wait to play them again and watch him and Boozer tangle. Still don't understand why JVG kept him on the bench that game.

 

The 1996 Bulls (best single season record in NBA history) that won 72 games had 22 blowouts of 20 or more points, so we aren't going to get 15-17 of them. Besides, we would need 31 point blowouts, not 17. The 1996 Bulls only had two of those. BUT... apparently that isn't even relevant for the Lenovo stats like I thought.

So, that explains why eliminating the Dallas game from the stats makes them higher. During that game, the bench was very productive so the differential between the best and worst players on our team was lower than the away games.

Seeing the Rockets topping out these Lenovo stats really only means that the productivity difference between the Rocket's best and worst is greater than any other team in the league. But once again.... even that doesn't really mean anything because of the sample size like you mentioned.

My biggest question is, if you look at the regular season stats from last year, you see mostly playoff teams at the top. Does that means that good teams have the highest production differentials between their rotation players? Does it mean that in addition to the internal team differentials, that the Lenovo stats are also weighted for overall team performance?

I would think that the best teams would want the lowest production drop-offs between their regular rotation players, and therefore score low on the +/- stats.

 

What it basically means is a team has got a unit that they can put out there that pretty much dominates every other team. The playoff teams and championsip team should be at the top. After all, when you get down to crunch time, only 7-8 guys play max.

 

I dont know what the Leveno or whatever is, but anything that puts TMAC at his rightful position at SG and keeps Battier at SF...Im all for!

 

Actually... it can't be just the difference becasuse I just checked the Knicks from last year and almost everyone on the team is negative. If it were purely the difference, then about half the team would be positive and half the team would be negative. And on the flip side, on playoff teams, almost everyone is positive.

Their stat explanation doesn't tell us everything.

 

Other factors would seem to be at play here also, so its not a purely internal team measurement. Substitutions are likely to be made by both teams, so the bench and starters on a given team are not likely to be playing against exactly the same players on the other team.

 

Much taller, much slower, his size just right. Otherwise, we will have 2 slow players on court, just like last year.

 

Whoops!

 

Wait a minute- from the site:
"The Lenovo Stat is a plus/minus statistic that looks at the point differential when players are both in and out of the game, to see how the team performs with various combinations."

 

Lakers should get a negative score with Kobe.

 

if the rockets are number 1 after tomorrow night, we're for real.

 

That may be what they say, but the +/- statistic they are citing is only describing On Court data. Perhaps they mean that the Lenovo stat indicates point differential for different player lineups, as opposed to the entire team as a whole. But for each 5-man unit, the stat is only describing On Court point differential.

 

Look at the best 3 - man combo, rockets are the top 10 of the top 10, nice!

 

Could you tell me where you're getting the info on how they calculate the statistic? I've googled and can't find the formula. I want to know more about this statistic but it's just flat out weird how the Rockets completely dominate this stat right now. From 5-man down to 1 man, Rocks totally dominate here. The sample size is small, but a LOT of total games have been played right now, and of all possible combinations of all players on all teams in all games so far this season, Rocks dominate every one? Weird.

For me, a good explanation is weak bench play.

 

By definition, +/- refers to point differential when a player(s) in on the court. "+" refers to points scored by your team, "-" refers to points scored by the opposition.

They don't explicitly spell this out on NBA.com, but I know that's what the stat is showing because other sources of +/- data are consistent with that. for example, look at 82games 5-man units. The "Off" and "Def" in the first table correspond to "+" and "-", respectively. The numbers are very slightly different (for our starting unit Lenovo has us at +52 while 82games has us at +49), probably due to minor differences in how they parsed the play by play.

Right now, our starting lineup has just played very well relative to the opposition. We've pretty much outscored the other team in every 1st and 3rd quarter, because our starting unit has played so well. JVG sticks with that starting lineup for about the first 7 to 8 minutes of the first and third quarters, and then usually comes back to it at some point in the second/fourth quarters. All the 1/2/3/4 player combinations have a similarly high +/- because they are benefitting from the stellar play of the 5-man unit as a whole.

If you look at our best 5-man units, you can see that our starting lineup has far and away been the best. None of our other lineups that have played significant minutes have a strong +/-. The Lenovo stat page makes it seem like we're dominating all the categories, but really that's more a testament to how well our starting lineup is playing (relative to the opposition) than anything else.

In general, the more you try isolate this information to individual players, the less reliable the +/- data is. So early in the season, I don't think the 1/2-player combination data is particularly relevant.

 

Interesting stuff, the Rockets don't really have any kind of significantly negative unit right now that played any minutes together. Seems like the coach is sticking with guys who got him results on the court (maybe risking wearing them down) and making substitutions quickly if the unit on the floor isn't producing.

Hopefully, some of the bench player will emerge and afford the starters some rest.

So far, I've liked how Head and Mutombo played, but they really need at least 1 or 2 more guys to step it up in order for the starters to rest without the team losing momentum.


One more thing, remember how Van Gundy was searching for that 5th starter all preseason to fit in with Alston, Yao, McGrady, Battier? I think we found him. The task now is to search for guys 6-9 to give 1-5 some relief.

 

I still think Van Gundy is itching to put a perimeter thread into that PF slot in the starting lineup ... but Battier is too good on the perimeter defensively to play him at PF and Novak just isn't NBA-ready. It's just killing him to play a second year player over Juwan Howard, I bet. If Hayes slumps for a few games, I don't think JVG will hesitate to replace him with Howard.

 

I think you are absolutely right there. Chuck probably knows this too... as long as he plays well, he will start. If Juwan starts out performing him with his minutes on the court, JVG will undoubtably put Chuck on the bench.