Revealing fancystat from Greg F

by Mike (bart), Thursday, January 23, 2025, 07:55 (15 days ago)

https://x.com/greg2126/status/1882140491594211791?t=ZwhlA08sJPvZt2SZ28s3Yg&s=19

The last two ND teams have been head and shoulders about any of the previous 20 seasons from an analytical perspective. Look at the gap between 2023 and the BK playoff teams. Then check out the gap between 2023 and this year. Super-duper interested to see what next year holds. Is it possible we keep improving? Or even maintain altitude?

This is particularly interesting

by BPH, San Diego, Thursday, January 23, 2025, 08:09 (15 days ago) @ Mike (bart)

in light of our recent "ND four-team playoff" discussion. We all agreed that 2024 was the best of the four "playoff" teams, but I don't think anyone considered the possibility that 2023 might wax 2012, 2018 and 2020 as well. And if Carr is the right guy and we manage to replace Golden (huge ifs), the personnel really does suggest that 2025 could continue the upward trajectory.

Think about it, from the team that was within a score of OSU in the fourth quarter (meaning no Morrison, Mills, etc.), what are we losing in terms of draft-eligible talent? Watts might go in the first two rounds. Will anyone else go in the top five? Meanwhile, OSU might have eight players go in the first couple rounds.

'23 was a really efficient team that never felt that way

by HumanRobot @, Cybertron, Thursday, January 23, 2025, 08:30 (14 days ago) @ BPH

I honestly think we would have gone 11-1 with Denbrock and the same squad.

--
I'm a bad take machine!

I think I'd like '23 in most scenarios

by Mike (bart), Thursday, January 23, 2025, 08:17 (14 days ago) @ BPH

It would be hard to see any of those Kelly teams getting to 20 points against that defense.

At the same time, I don't think it would be prohibitively unlikely for those Kelly teams to shut down the offense for long stretches and/or inspire some type of Hartman turnover meltdown

what is F+ again?

by HumanRobot @, Cybertron, Thursday, January 23, 2025, 08:00 (15 days ago) @ Mike (bart)

- No text -

--
I'm a bad take machine!

Denbrock's grade for the season, of course.

by Savage, Around Ye Olde Colonial College, Thursday, January 23, 2025, 08:44 (14 days ago) @ HumanRobot

(I'm completely kidding, just to be clear.)

I think a mashup of FEI (LaFun) and S/P+ (Connelly)

by Mike (bart), Thursday, January 23, 2025, 08:09 (15 days ago) @ HumanRobot

- No text -

S/P+?? Where we finished 8th despite JUST beating 5 & 6.

by Chris @, Raleigh, NC, Thursday, January 23, 2025, 11:10 (14 days ago) @ Mike (bart)

But he's a nice guy.

--
"F--- everyone who isn't us."
#Team128

Taking a quiet little victory lap with FEI this year.

by LaFortune Teller ⌂ @, South Bend, Thursday, January 23, 2025, 16:57 (14 days ago) @ Chris

I don't engineer FEI to make projections, I engineer it to most accurately represent past results. And it does quite well in that regard relative to other computer systems:

But I also produce game projections based on FEI, which I use mostly as a calibration exercise. My hypothesis is that a system that is well-tuned to past results should be well-tuned to the "true" strength of teams, and therefore should do reasonably well forecasting future outcomes.

Well, from Week 14 (when I had phased out all preseason projection data) through the end of the playoff, FEI game projections picked 88 out of 123 game winners outright (71.5%) while the market (per closing lines) only picked 80 correct game winners (65.0%) over the same stretch. FEI projected winners were 11-0 in the playoff (closing line favorites went 10-1), with a mean absolute error of only 6.7 points per game (closing line error was 8.9). This also compared very favorably with other computer systems, most of which (like SP+) are designed to forecast game outcomes by design: https://public.tableau.com/app/profile/andrew.percival/viz/CFBPicker/Standings

Very impressive stuff. Congrats.

by omahadomer, Friday, January 24, 2025, 09:56 (13 days ago) @ LaFortune Teller

- No text -

Very impressive stuff. Congrats.

by omahadomer, Friday, January 24, 2025, 09:47 (13 days ago) @ LaFortune Teller

- No text -

Congrats Brian. That's great.

by Domer99, John Wesley Powell's Expedition Island, Thursday, January 23, 2025, 17:35 (14 days ago) @ LaFortune Teller

- No text -

Yes, really impressive!

by Joe I @, Thursday, January 23, 2025, 17:42 (14 days ago) @ Domer99

- No text -

Does the ND football program ever

by Mike (bart), Thursday, January 23, 2025, 17:00 (14 days ago) @ LaFortune Teller

Ask you to get involved analytically? Would you get time and a half if they did?

Involved, no. But I have had a few conversations

by LaFortune Teller ⌂ @, South Bend, Friday, January 24, 2025, 04:13 (14 days ago) @ Mike (bart)

with folks in the Gug about my stuff. I'm not sure if there's much utility for it to help a team get better, though. There are folks doing far more detailed work in the "coaching decision-making" realm, or scouting tendencies, or with player evaluation for instance. My stuff mostly describes something a team already knows about itself and its opponents.

The most interest I've gotten in my numbers from team personnel has been from special teams folks, actually. But even that has come from a place of curiosity about what I'm measuring and how than an applicability to game prep or anything.

Effin Connelly.

by Domer99, John Wesley Powell's Expedition Island, Thursday, January 23, 2025, 13:17 (14 days ago) @ Chris

- No text -

Correct.

by LaFortune Teller ⌂ @, South Bend, Thursday, January 23, 2025, 10:18 (14 days ago) @ Mike (bart)

Originated when we were both writing at Football Outsiders to be an "official" FO rating, and I've maintained it on my site when that one folded: https://www.bcftoys.com/2024-fplus

powered by my little forum