Why are the kenpom and net rankings so different?
- Thread starter GoBIG BLUE
- Start date
NET is a bit of a black box.
But I would guess one contributing factor might be that Texas is bolstered by their preseason ranking in Kenpom. NET does not have a preseason contribution from what I can tell.
I'm sure there are other reasons, but it's impossible to know for sure as NET's formula is a mystery.
Probabaly because they use Torvik’s T-ranking this year. I’d assume there are so,we differences in the models against Kenpom.
One of them (Kenpom?) heavily weights last season performance to support rankings early in the season. That factor gradually washes out as the season matures.
It's hard to say because the NCAA isn't completely clear on the calculation of NET.
But FWIW the actual ranking of teams is very close. A 20 spot difference might not be all that significant.
Also I think with all these systems, they generally agree with the teams towards the top. The teams that are capable of winning national titles.
But FWIW the actual ranking of teams is very close. A 20 spot difference might not be all that significant.
Also I think with all these systems, they generally agree with the teams towards the top. The teams that are capable of winning national titles.
KenPompom is only about pace. It's not a ranking. Use kenpompom to bet and go broke. It's fake numbers. He doesn't count an offensive rebound as a possession. That's fine because he's showing pace on paper. Counting an offensive rebound as a possession would make the game look like it was played at a faster pace on paper. But who cares what the paper pace is?
The biggest thing though with not counting offensive rebounds as a possessions is it inflates the PPP numbers. Thus those are fake numbers too. It's just one more thing to the anal ytic types to use to make straw man arguments.
Bottom line, kenpompom removes plays from the game. Removing those plays skews all of his numbers.
Net uses this season only so it will have a lot of variability early. Kenpom is still factoring in preseason expectations at this point so it’s more stable but probably a little less accurate.
Also I believe net has an upper limit on margin of victory, and kenpom doesnt.
Also I believe net has an upper limit on margin of victory, and kenpom doesnt.
Lord lol.
If I make a trip down the court and I make a shot, I've scored 2 points on that possession. 2.00 points per poss.
If I make a trip down the court and I miss a shot, but I rebound the ball and score, I've still scored 2 points on that possession.
In what world would it make sense to count that as 2 possessions? 1 point per possession?
It doesn't. The end result is still the same.
The real world. KenPomPom is the only analysis that doesn’t count an offensive rebound as a possession.
How did you did a rebound if you didn’t possess the ball. How did you put it back in without taking a shot?
Pompom doesn’t count it because all he’s tracking is pace.
That’s to point about 2 points per trip.
OK, that’s right. It’s points per trip. It’s still not points per possession.
His numbers are screwed up across the board by not counting all possessions.
It’s nerd numbers.
You're not really explaining why this is a false way of looking at it tho. The rebound is just a continuation of the current possession. If you missed a shot and rebounded, the ball didn't change hands. It wasn't as if the opponent had a possession. Its still in your possession. Logically it makes perfect sense.
At the end of the day, all that matters is the points scored. That's what wins basketball games.
There's really no logical reason why you wouldn't go this route of accounting for things when the end result is the exact same. Trips? Possessions? It's all semantics. The bottom line is you went down the court and scored 2 points.
FWIW Synergy data DOES count rebounds on offense as separate possessions. This is to my knowledge the only place that does this. The general consensus not just Kenpom but ANY of the systems treats it as one possession.
Directly from his blog:
You might wonder why offensive rebounds are treated as continuing a possesion, rather than starting a new one. I’ve seen two good reasons. First, by including them each team’s possesions can reasonably be assumed to come out equal for each game. Second, getting and preventing offensive rebounds are skills. So if some teams do those skills better than others, it makes sense to attach those skills to a team’s offensive or defensive ability.
Another thing about this. The whole framework from this is based on Dean Oliver's Basketball On Paper book. It wasn't like these current systems just developed this stuff and made it up.
Last edited:
Similar threads
