Stats models and rankings

stinger78

stinger78

Helluva Engineer
Messages
4,925
SP+ Stats 1970-2023

Overall:

SP+ Overall 1970-2023.PNG


Offense:
SP+ Offense 1970-2023.PNG


Defense:
SP+ Defense 1970-2023.PNG
 

GT33

Helluva Engineer
Messages
2,275
Wow, what a surprise. GT plays better when we run the ball.... a lot.

Bell, Burns, Mays, Choice, Dwyer, JT, JfN, Mills, Marshall, Hollings, Haynes, etc.
 
BurdellJacket

BurdellJacket

Ramblin' Wreck
Messages
514
Location
Atlanta
slugboy said:
I should add @ibeattetris and @GetYourBuzzOn here, to add in more info.

A lot of the models have “priors”. In the case of FEI, it’s the last 5 seasons., so it’s one of Johnson’s worst seasons followed by 4 of mostly Collins. We started that model this year with an F- average to dig out of.

The overall consensus for us (averaging out all these models), is #78. That doesn’t really mean anything, especially 4 games in, but if you have 9 models, hey why not average them?

There are the Massey Ratings. We’re #70 on the Power Ratings and #67 on the regular ones.

Some models tell more about how they work. FEI and Beta Rank and Sagarin and SP+ will tell how they’re calculated. and what the inputs are. Others are harder to sus out.

FEI and SP+ were popular among a lot of people because they were easy to find, and explained how they worked. SP+ is now in the ESPN paywall, so you either pay for ESPN+ or you turn to another model.

And, as a complete aside, SMU might be the best team we added in expansion. Even before becoming a P5 team.

There’s TeamRankings—we’re #58. I have no idea how they build the model.

There’s Doktor Entropy. He does weekly predictions, and picked Wake by 8. Not sure how his model works, but he’s been selling it for ages.

There’s ESPN’s FPI. It seems to think we’ll have between 5 and 6 wins this year. Ranks us 49th. It’s the one you know about, because it’s ESPN.

There’s Kelley Ford (KFord Ratings). He’s on Twitter and will explain his model. It holds up very well in Pick’em contests. We’re up to #57.

The last two are Brian Fremeau’s FEI (at bcftoys.com) and Bill Connelly’s SP+ (over at ESPN, paywalled)

Outside of those, there are the graphs and info at gameonpaper.com and cfb-graphs.com, plus other info at https://collegefootballdata.com/.

Unless we beat Bowling Green 222-0 (or even if we do) Miami will still be favored in game 6. However, we’re 2-2, if we beat Bowling Green, we have a winning record for the first time this season.
Click to expand...
 
stinger78

stinger78

Helluva Engineer
Messages
4,925
roadkill said:
Here’s an interesting ranking he doesn’t use: https://www.colleyrankings.com/rank.html

In this system, where .5 is average, GT now sits at .54, good for 67th. His ranking system was designed to converge on a valid national champ at season’s end, or to be useful in comparing playoff-eligible teams. He has a link to a paper explaining his mathematics in great detail. It's kinda like the transitive property on steroids.

Colley’s is unlike most of the others in that it just uses current season wins and losses, and thus starts with nothing. This makes it less useful early in the season (Then again, what poll is?). At the same time, it claims that as an asset since it doesn’t use preseason or historical data as a bias, which has been a frequent critique of the current system where last year’s champion always gets the benefit of the doubt, and can prevent teams that don’t start highly ranked from ever getting into the playoff, regardless of record. In the modern game, given the transfer portal and frequent coaching turnover, those biases are less useful than 10-20 years ago.
Click to expand...
I like Colley for just that reason: wins and losses. Everyone else tries to model wins and losses using some type of proxy. However, as I’ve said many times, the factors used as indicators for the proxy (both formative and reflective) are many and confounding and full of statistical error.

We all know the “transitive property” doesn’t work well in athletics, but if you generate a large enough sample you can minimize that error. Using first and second level W/L is a way to do that. Still, like any algorithm, there will be some head scratchers along the way.
 
roadkill

roadkill

Helluva Engineer
Messages
1,921
stinger78 said:
I like Colley for just that reason: wins and losses. Everyone else tries to model wins and losses using some type of proxy. However, as I’ve said many times, the factors used as indicators for the proxy (both formative and reflective) are many and confounding and full of statistical error.

We all know the “transitive property” doesn’t work well in athletics, but if you generate a large enough sample you can minimize that error. Using first and second level W/L is a way to do that. Still, like any algorithm, there will be some head scratchers along the way.
Click to expand...
Since you mentioned it, here are some interesting end-of-year results from Colley's:

Our final rank landed at 51st, or slightly above average. About where you would expect with our 7-6 record. Our final SoS ranked 41st.

Bowling Green landed at 72nd - very slightly below average for an FBS school, and interestingly, 3 spots ahead of UCF. We ought to have won that game, but it is not in the same league as losing to say, The Citadel, or Furman, which are FCS.

Colley's also allows you to play the "what-if" game to see hypotheticals. I did an optimistic take, but you can also review where we'd be if we for example lost the Miami game.
  • Had we beaten Bowling Green, we'd move up to 37th, which is near where our overall talent level is ranked.
  • Had we beaten both Bowling Green and BC, we'd land at 28th, within spitting distance of being ranked at the end of the season.
 
stinger78

stinger78

Helluva Engineer
Messages
4,925
I copied this over from the Conference Realignment thread as it seemed more appropriate here:

So, I looked at the strength of schedule list for CFB 2023. Here’s what I found for SECheat teams who form The Greatest Football Conference In The World. I display number of SECheat teams every 5 slots:
1-5: 0
6-10: 2
Top 10 hardest schedules: 2

11-15: 2
16-20: 0
Top 20 hardest schedules: 4/20

21-25: 2
26-30: 1
Top 30 hardest schedules: 7/30

31-35: 1
36-40: 2
Top 40 hardest schedules: 10/40

41-45: 1
46-50: 2
Top 50 hardest schedules: 13/50

At 5 conferences, perfect balance would be 2/10, 4/20, 6/30, 8/40, 10/50. The SECheat, for all it’s purported difficulty, is in perfect balance with the four other conferences through the top 20 most difficult schedules, and pulls ahead only by 1 team in the Top 30 programs. It pulls further ahead of average between #41-50.

This would seem to indicate that, if they are a collection of the best, they are not playing each other enough to matter much at the highest level. Better than average, yes, but not dominantly so.
 
slugboy

slugboy

Moderator
Staff member
Messages
11,725
BurdellJacket said:
Amazing! Should move up to a top 10 offense next year. Plenty of room for year 2 improvement with almost all the O back next year.
Click to expand...
Moving from 27 to 15 is in some ways harder than moving from 116 to 27.

Ole Miss, Ohio State, Oregon, UGA, Michigan, Oklahoma, Washington, LSU, and a bunch of factories are in that range. There are some absolutely loaded teams in the 16-25 range, too. As in, lots of players getting drafted kind of teams.

Liberty, Kansas, and UNC are there, too. It’s possible to get there, but a lot of things have to fall the right way for that to happen
 
Top