Our New Conference - WCC, West Coast Conference

I've never asked him. But as a guess, it is a lot of historical statistics as a starting point. Then weighing in this years games and after a few games, all the rankings are based on this year. Historically, OSU has ended up at #155 (23-24), #214 (22-23), #233 (21-22), #43 (20-21), #87 (19-20), #81 (18-19), #100 (17-18), #264 (16-17), #64 (15-16) and #107 (14-15). Those are the Tinkle years. 

Heading into the Pac-12 Tournament, Oregon State's KenPom rating was 103rd.  Oregon State jumped 60 spots over the next seven games, which is pretty solid, all things considered.  

Which 2 numbers should be closer?

Part of the issue is that KenPom has a preseason poll.  The rankings are based, in part, on what happened in years past rather than what happened just in 2021.

I think the preseason numbers likely are barely a part of the equation. The issue is that team prior to conference tournament was not very good. They were 14-12 (10-10). As someone above pointed out, they had a large jump as they won. And KenPom aren’t like AP poll where ranking is based entirely on votes that week. When you win a game(s), you don’t automatically move up in KenPom — and move up even higher as other teams above you lose. It behaves more like a FG% or something like that. If you’re a 45% shooter for 3/4 of the year and then shoot 50% for last quarter you’re unlikely to be ranked higher than someone who shot 49% for 3/4 of season and then shoots 45% for last quarter. All those early numbers aren’t forgotten in KenPom where they are more in AP poll and completely in a tournament standing.

But the main issue is that people talk about KenPom numbers incorrectly. As written in his explanation of system:

“The first thing you should know about this system is that it is designed to be purely predictive. If you’re looking for a system that rates teams on how “good” their season has been, you’ve come to the wrong place. There are enough systems out there that rank teams based on what is “good” by just about any definition you can think of. So I’d encourage you to google college basketball ratings or even try the opinion polls for something that is more your style.”

Right on, thanks.  It makes sense if a regular season loss in November means as much as a tourny win.

I still feel like these rating systems that rank you vis a vis who you have played and what their results are tend to slant against the west coast simply because there aren't as many mid and lower level programs for everyone to build their resume against.  RPI, FPI, Elo, NET, etc all seem to churn out the similar results every year.  Have a .500 year in the Pac12 and you are in the cellar.  Have a .500 year in the SEC, and you're a bubble team.

Even when the Pac12 had a good year in general, we were still scrambling to get more than 4 teams in the tourny.  And it always seemed like the conference was fighting an uphill battle with seeding on top of that, for the same reason.  I always had to laugh when the narrative would be about the Pac12 "overperforming" in the post season so much of the time.  Well yeah, because we keep getting underrated.

I'm just an idiot, so what do I know.  But if a rating system is continually turning out the same results, year after year, at some point it stops being a coincidence.  I'll never understand why seemingly nothing was ever said or done about it by anyone with real clout in the Pac12 sphere.