Ratings & Models postmortem

slugboy

Moderator
Staff member
Messages
13,978
Vegas liked Tech by about 3 1/2. Most of the other stats predictors liked Colorado by a point or two.

Even though it was a 7 point win, most of the post game metrics said it was a thorough win by Tech--about a "98% post game win expectancy".

Sagarin predicted a win. FEI, SP+, ESPN FPI, and most others predicted a narrow Colorado win.

I expect we will be outside the top 25 polls this week, but "getting votes". I think FSU is at least in the top 20, and probably in the top 15.

 

bke1984

Helluva Engineer
Messages
4,422
I think FSU is at least in the top 20, and probably in the top 15.
Funny how that works. If we had played and beat Alabama that way yesterday we’d be ranked like 21st. But FSU does it and they will be something like 13th just based on brand name alone - likely one spot ahead of wherever they drop Alabama to.
 

slugboy

Moderator
Staff member
Messages
13,978
I’ll put the five factors back out here (FootballStudyHall.com)

=================================
Explosiveness, efficiency, field position, finishing drives, and turnovers are the five factors to winning football games.

  • If you win the explosiveness battle (using PPP), you win 86 percent of the time.
  • If you win the efficency battle (using Success Rate), you win 83 percent of the time.
  • If you win the drive-finishing battle (using points per trip inside the 40), you win 75 percent of the time.
  • If you win the field position battle (using average starting field position), you win 72 percent of the time.
  • If you win the turnover battle (using turnover margin), you win 73 percent of the time.
=================================


We lost the turnover battle badly-> 0-3.

We had 7 explosive plays to Colorado's 3.

We beat them on success rate by a lot.

Our defense did well taking away the big plays, but we below average at keeping them behind the chains

They won the field position battle, mostly because of turnovers.

We won scoring within the 40 battle--our defense did very well there

Us winning despite being -3 on turnovers means we had to solidly beat them in all other aspects of the game, and we did

Concept from Robert Binion (@robert_binion). Data from GameOnPaper.com by Akshay Easwaran (@akeaswaran) and Saiem Gilani (@saiemgilani). Cell colors reflect the percentile of a team's performance against all single-game FBS vs FBS performances in that stat in 2024.
OverallGT Colorado
EPA/Play0.23 84th %ile0.03 50th %ile
Success Rate57% 97th %ile44% 65th %ile
Yards/Play6.93 77th %ile5.27 33rd %ile
EPA/Dropback0.21 71st %ile-0.05 40th %ile
EPA/Rush0.24 83rd %ile0.12 69th %ile
Yards/Dropback6.86 63rd %ile5.83 45th %ile
Explosive Play Rate10% 71st %ile5% 17th %ile
3rd Down Success Rate53% 76th %ile36% 31st %ile
Red Zone Success Rate50% 73rd %ile45% 59th %ile
Def Run Stuff Rate10% 0th %ile9% 0th %ile
Havoc Rate5% 11th %ile6% 15th %il

DrivesGT Colorado
Total1010
Avg Starting Field PositionOwn 25Own 41
Plays/Drive8.608.08
Yards/Drive56.8740.92
Available Yards %76%70%
DefensiveGT Colorado
Scrimmage Plays5967
Stop Rate32%24%
Havoc Plays Created3 (5%)4 (6%)
  Passing1 (3%)1 (5%)
  Rushing2 (7%)3 (7%)
TFLs Generated34
  Passing11
  Rushing23
Sacks Generated1 (3%)1 (5%)
Passes Defensed00
Interceptions00
Fumbles Forced00
 

RonJohn

Helluva Engineer
Messages
5,648
How is the Yards/Drive calculated? It says each team had 10 drives, and GT-56.87, CO-40.92 Yards/Drive. However GT had 463 yards, not 568.
 

RonJohn

Helluva Engineer
Messages
5,648
How about this model?



FPI is a joke. It is proprietary and their algorithms are not public. As far as I know not even a high level explanation of their calculations is public. Everything I see tells me that it is geared towards ranking teams high that the SEC wants to be ranked high. I don't doubt that they "adjust" the calculations week to week.

Alabama lost to the team that is currently 40th in the FPI (was 65th before) and is still in the top 15. Texas lost to Ohio State, FPI says less chance of making playoffs that Ohio State, FPI says less chance of making NC game than Ohio State, but is ranked higher than Ohio State and has a better chance of winning the NC than Ohio State. It makes no sense. Especially, how can Texas be calculated to have a lower chance to get to the NC game, but a higher chance of winning the NC game.
 

roadkill

Helluva Engineer
Messages
2,769
How is the Yards/Drive calculated? It says each team had 10 drives, and GT-56.87, CO-40.92 Yards/Drive. However GT had 463 yards, not 568.
Yeah, Yards/Drive doesn't line up with both teams' total yards and total drives, and you can't make the numbers work by tossing out drives either. Something's amiss here. Also, there are other apparent errors in the raw data that call into question the overall integrity of the advanced stats for this game. The ESPN stats don't include King's interception, and the Game on Paper advanced stats that @slugboy reproduced here (which use ESPN data) show only 3 TFLs, yet ESPN counts 6. I thought perhaps ESPN was overcounting TFLs when multiple players participated in the tackle, but that doesn't line up either.
 

TechPhi97

Helluva Engineer
Messages
2,029
Location
Davidson, NC
I’ll put the five factors back out here (FootballStudyHall.com)

=================================
Explosiveness, efficiency, field position, finishing drives, and turnovers are the five factors to winning football games.

  • If you win the explosiveness battle (using PPP), you win 86 percent of the time.
  • If you win the efficency battle (using Success Rate), you win 83 percent of the time.
  • If you win the drive-finishing battle (using points per trip inside the 40), you win 75 percent of the time.
  • If you win the field position battle (using average starting field position), you win 72 percent of the time.
  • If you win the turnover battle (using turnover margin), you win 73 percent of the time.
=================================


We lost the turnover battle badly-> 0-3.

We had 7 explosive plays to Colorado's 3.

We beat them on success rate by a lot.

Our defense did well taking away the big plays, but we below average at keeping them behind the chains

They won the field position battle, mostly because of turnovers.

We won scoring within the 40 battle--our defense did very well there

Us winning despite being -3 on turnovers means we had to solidly beat them in all other aspects of the game, and we did

Concept from Robert Binion (@robert_binion). Data from GameOnPaper.com by Akshay Easwaran (@akeaswaran) and Saiem Gilani (@saiemgilani). Cell colors reflect the percentile of a team's performance against all single-game FBS vs FBS performances in that stat in 2024.
OverallGTColorado
EPA/Play0.23 84th %ile0.03 50th %ile
Success Rate57% 97th %ile44% 65th %ile
Yards/Play6.93 77th %ile5.27 33rd %ile
EPA/Dropback0.21 71st %ile-0.05 40th %ile
EPA/Rush0.24 83rd %ile0.12 69th %ile
Yards/Dropback6.86 63rd %ile5.83 45th %ile
Explosive Play Rate10% 71st %ile5% 17th %ile
3rd Down Success Rate53% 76th %ile36% 31st %ile
Red Zone Success Rate50% 73rd %ile45% 59th %ile
Def Run Stuff Rate10% 0th %ile9% 0th %ile
Havoc Rate5% 11th %ile6% 15th %il

DrivesGTColorado
Total1010
Avg Starting Field PositionOwn 25Own 41
Plays/Drive8.608.08
Yards/Drive56.8740.92
Available Yards %76%70%
DefensiveGTColorado
Scrimmage Plays5967
Stop Rate32%24%
Havoc Plays Created3 (5%)4 (6%)
  Passing1 (3%)1 (5%)
  Rushing2 (7%)3 (7%)
TFLs Generated34
  Passing11
  Rushing23
Sacks Generated1 (3%)1 (5%)
Passes Defensed00
Interceptions00
Fumbles Forced00
Dude, this is great. I love it.
 

slugboy

Moderator
Staff member
Messages
13,978
Yeah, Yards/Drive doesn't line up with both teams' total yards and total drives, and you can't make the numbers work by tossing out drives either. Something's amiss here. Also, there are other apparent errors in the raw data that call into question the overall integrity of the advanced stats for this game. The ESPN stats don't include King's interception, and the Game on Paper advanced stats that @slugboy reproduced here (which use ESPN data) show only 3 TFLs, yet ESPN counts 6. I thought perhaps ESPN was overcounting TFLs when multiple players participated in the tackle, but that doesn't line up either.
I’m not sure what’s wrong with the game on paper data. It’s clearly missing the INT in the overall stats , but it has it in the drives.

Sometimes, penalties will throw off some of the numbers

The numbers should line up with the ESPN data—it’s supposed to be the same data. Possibly, ESPN’s feed was messed up at one point in time
 

AUFC

Helluva Engineer
Messages
4,542
Location
Atlanta
The metrics seem to reflect my eye test during the game. I thought we were composed and in control most of the game, early turnovers aside.

One thing that has gone a little unnoticed is the quiet confidence of Key/Faulkner on that final drive of the first half. Given the time and score, it really felt like they told HK: “it’s 10-10, we are clearly the better team despite some dumb mistakes. Run clock, don’t do anything stupid, get 3 points if you can, but as long as we are at least tied, we are winning the second half.” Very good drive and then great kick by Aiden to cap it off.
 

gtrower

Helluva Engineer
Messages
3,921
Vegas liked Tech by about 3 1/2. Most of the other stats predictors liked Colorado by a point or two.

Even though it was a 7 point win, most of the post game metrics said it was a thorough win by Tech--about a "98% post game win expectancy".

Sagarin predicted a win. FEI, SP+, ESPN FPI, and most others predicted a narrow Colorado win.

I expect we will be outside the top 25 polls this week, but "getting votes". I think FSU is at least in the top 20, and probably in the top 15.


It’s an interesting dilemma for pollsters. Last year we beat #10 on a neutral site after a 7-6 season. After a 2-0 start we appeared at #23 and unranked IIRC.

FSU beats #8 at home after a 2-10 season. Do they dare keep Bama above them? If not how far do they dare drop Bama? Because they’re sure as hell not gonna be unranked. So that means fsu probably ends up top 20? Top 15? Which is ****ing ludicrous. They shouldn’t get the benefit of the doubt after last season. Rather neither team should be ranked until they prove it a few more weeks.
 

Heisman's Ghost

Helluva Engineer
Messages
6,106
Location
Albany Georgia
Vegas liked Tech by about 3 1/2. Most of the other stats predictors liked Colorado by a point or two.

Even though it was a 7 point win, most of the post game metrics said it was a thorough win by Tech--about a "98% post game win expectancy".

Sagarin predicted a win. FEI, SP+, ESPN FPI, and most others predicted a narrow Colorado win.

I expect we will be outside the top 25 polls this week, but "getting votes". I think FSU is at least in the top 20, and probably in the top 15.

Beating Coach Prime and Colorado just doesn't move the needle anymore. IIWII
 

TechPhi97

Helluva Engineer
Messages
2,029
Location
Davidson, NC
It’s an interesting dilemma for pollsters. Last year we beat #10 on a neutral site after a 7-6 season. After a 2-0 start we appeared at #23 and unranked IIRC.

FSU beats #8 at home after a 2-10 season. Do they dare keep Bama above them? If not how far do they dare drop Bama? Because they’re sure as hell not gonna be unranked. So that means fsu probably ends up top 20? Top 15? Which is ****ing ludicrous. They shouldn’t get the benefit of the doubt after last season. Rather neither team should be ranked until they prove it a few more weeks.
You’re acting like people take an unbiased and rational approach to ranking college football teams! :LOL:
 

Root4GT

Helluva Engineer
Messages
6,025

GaTech4ever

Helluva Engineer
Messages
1,784
Especially, how can Texas be calculated to have a lower chance to get to the NC game, but a higher chance of winning the NC game.
Because Ohio State couldn’t do anything against Texas’ defense, the game wasn’t played on a neutral field, and that’s probably the worst game Arch will play all season.

But Texas did lose, therefore the chances they make the NC game are less. But if both teams win out, Texas would likely be favorites if they played on a neutral field in January.
 
Top