Fun with interesting statistics

by Tom Martin on Nov 18, 2009 3:06 PM CST in Whimsical Posts

Thank the lord for Roland Beech and 82games.com. Otherwise, I really don't know what I would do with my free time.

I went treasure hunting for some obscure Rockets statistics and found the following three to be particularly compelling.

1. Hayes/Scola tandem has much higher effective field goal percentage than Landry/Scola

Really? We actually produce better on offense with the Chuckwagon instead of Carl Landry? That's odd. Perhaps it's due to usage, but you'd still think that Luis Landry, accompanied by Brooks, Battier, and Ariza, would have a higher eFG than .364.

2. The second-most used unit is significantly worse than all the others.

Maybe we shouldn't play our worst unit so often. Clearly, our defense struggles when Brooks, Hayes, and Ariza are all off of the floor. Also: collectively, Rockets are better defensively with Dave Andersen on floor than Chuck Hayes. Weird...

3. Shane Battier has the worst plus/minus on the team, as well as the third-worst Def rating (opponents points per possession while on the floor).

Well, Battier does play a good chunk of minutes and guards the opponent's best player. But wow - this is pretty stunning. Then again, +/- is a pretty flawed statistic in many, many ways.

There - your Wednesday fun post is finished.

Tweet Comment 12 comments | 0 recs |

Do you like this story?

Comments

On point one

would chucks offensive rebounding play a major roll in this? He’s more likely to get an offensive board as most centers drift off him slightly to double or help so he’ll get us an extra possession and therefore we are more likely to score

"We obviously didn’t know the scouting report that well on Andersen. We knew he’s a shooter but we didn’t see him play at that level." Phil Jackson on underestimating the aussie

by DREAM34 on Nov 18, 2009 3:18 PM CST reply actions

More likely

Hayes is strictly shooting in close w/a high percentage-when they’re not getting rejected :)
Landry is shooting a lot of jumpers w/a correspondingly lower %.

by Tisbee on Nov 18, 2009 7:06 PM CST up reply actions

ouch

I’d like to thank you for crushing all of my dreams and preconceived notions about the strength(s) of my favorite team. Thanks, Tom.

by grungedave on Nov 18, 2009 3:33 PM CST reply actions

I thought that the start to the season would have already done that,

Mr. I-Think-The-Rockets-Will-Be-Lucky-To-Win-34-Games.

ZING!

by Only_A_Lad on Nov 18, 2009 5:55 PM CST up reply actions

read closely

I said they would be great in November, but the energy we see now may not be there in January and February – and that’s when losses will mount.

by grungedave on Nov 18, 2009 6:13 PM CST up reply actions

oush!

by kmbrly429 on Nov 19, 2009 11:36 AM CST up reply actions

oops i meant ouch!

by kmbrly429 on Nov 19, 2009 11:36 AM CST up reply actions

too much noise in 11 games to truly evaluate based on +/-. You really want to look at a season of data before drawing much of a conclusion.

Fun post though.

i cry for nic

by Cablinasian on Nov 18, 2009 4:14 PM CST reply actions

exactly

at this point in the season, too little has happened. There’s just no way to properly adjust the +/- stats yet.

by Only_A_Lad on Nov 18, 2009 5:54 PM CST up reply actions

I was going to come in and play everyone's favorite song

“It’s A Small Sample Size After All”

but I see I don’t need to do that.

Go on, try to beat them. Try to score on them. Stay in front of those guards. Try to tap out an offensive rebound on that Lilliputian front line. They're so awful, you're so good ... go on. Beat ‘em. - Kelly Dwyer on The Rockets

by Xiane on Nov 18, 2009 10:32 PM CST up reply actions

On point two

The #2 lineup happens to have all 4 non-starting rotation players in them (Lowry, Budinger, Landry, Andersen). I imagine that all of the lineups with the 4 bench players in them are used mostly to give the starters rest, and not because we would prefer to have that lineup in the game. So let’s compare the three lineups above that have the entire bench in the game: lineups #2, #5, and #9.

The difference between these three is the starter that fills in the 5th spot: Battier, Brooks, and Ariza, respectively. So the lesson I would take from this is that Brooks + bench has a much higher success rate than Battier + bench since the win% for lineup 5 is 100 and for lineup 2 is 28.5.

Why? Well, Lowry, Budinger, Landry, and Andersen are much more sprinty than our starters, so adding Ariza or Brooks allows them to really fly down the court, while Shane might slow them down a bit.

So why play lineup #2 the 2nd most minutes? My guess is happenstance: the 33.7 minutes played by #2 is two hours less than the starting lineup, but only 18 minutes more than the 9th most used lineup.

So we’re playing our “worst” lineup 2nd most often, but not so much more than every other lineup that it can’t be explained by foul trouble or a couple of games where a few minutes passed before a dead ball allowed a planned substitution.

by Moondebah on Nov 18, 2009 4:14 PM CST reply actions 1 recs

Interesting

regarding your 2nd point, the only difference from our best unit (+18) and worst unit (-21) is one substitution. Simply replacing Battier with Lowry is all we’d have to do.

With that said, small sample size. Battier’s +/- will look a lot better when T-Mac comes back.

by goingforthecorner on Nov 19, 2009 3:51 PM CST reply actions