Stats models and rankings

[slugboy]

A cool QB graph

 

[Lotta Booze]

Good to see us on the right side of the graph this week.

Also, on uga

 

[stinger78]

Good to see us on the right side of the graph this week.

Also, on uga

Can you interpret this graph for me, please?

 

[Lotta Booze]

Can you interpret this graph for me, please?

It graphs the winner and loser of a game in terms of their overall success rate throughout the entire game. Here's how they define success rate.

The winner of the game is the team away from the baseline of 0. The teams on the right, significantly above the baseline of 0 indicate they were much more successful on a down to down basis throughout the entirety of a game. The teams on the left, below the baseline, generally were not as successful on a down to down basis but still came out with win. Could be due to turnovers, red zone lack of success, starting field position, etc. In theory, for the teams on the right if they played that game 10 times the winner would still win a majority of those games. For the teams on the left one would think if they played that game 10 times they would win a minority.

"Did we really get beat that bad?" is the title. If you lost and you're on the right, yes you got beat pretty bad. If you lost and you're on the left, no you didn't get beat that bad at all but had some tough breaks that lost the game.

 

[Lotta Booze]

 

[stinger78]

It graphs the winner and loser of a game in terms of their overall success rate throughout the entire game. Here's how they define success rate.

The winner of the game is the team away from the baseline of 0. The teams on the right, significantly above the baseline of 0 indicate they were much more successful on a down to down basis throughout the entirety of a game. The teams on the left, below the baseline, generally were not as successful on a down to down basis but still came out with win. Could be due to turnovers, red zone lack of success, starting field position, etc. In theory, for the teams on the right if they played that game 10 times the winner would still win a majority of those games. For the teams on the left one would think if they played that game 10 times they would win a minority.

"Did we really get beat that bad?" is the title. If you lost and you're on the right, yes you got beat pretty bad. If you lost and you're on the left, no you didn't get beat that bad at all but had some tough breaks that lost the game.

So, the convergence at -0- is the difference between "right" and "left."

 

[Lotta Booze]

So, the convergence at -0- is the difference between "right" and "left."

Yes...-ish. If i'm understanding your question. My understanding is the losing team is the base of 0 and the winning teams' Net Success rate through the whole game is reflected off of that. So the teams in the middle that are both real close to zero had a very even game, according to "success rate"

 

[slugboy]

So, the convergence at -0- is the difference between "right" and "left."

Yes...-ish. If i'm understanding your question. My understanding is the losing team is the base of 0 and the winning teams' Net Success rate through the whole game is reflected off of that. So the teams in the middle that are both real close to zero had a very even game, according to "success rate"

Usually, the team that wins plays a more efficient overall game—more first downs, more yards per play, etc. That’s why the right side above the axis has much more area than the section below the axis on the left.

To the left of where the curve crosses the axis, that’s “we had more yards and first downs and time of possession and big plays and still lost. How did that happen?”. Often it’s turnovers—we drove to the 2 and fumbled. Sometimes it’s settling for field goals (and sometimes missing). Sometimes it’s punting from the opposing 40 yard line a bunch.

 

[TampaBuzz]

What about ST? We are getting a lot better. Certainly compared to last year where we couldn't even kick EPs.

Anyone know why in the UVa game we had the two short Kick Offs? Ones before and after were TBs.

I was wondering about that. I think Birr was doing the kick-offs this game. He has not been doing that so far this year. I was wondering if an injury or something else to the usual kick-off guy.

 

[yeti92]

I was wondering about that. I think Birr was doing the kick-offs this game. He has not been doing that so far this year. I was wondering if an injury or something else to the usual kick-off guy.

Stewart handles kickoffs. He had the short kicks against UVA and one against UNC as well. No idea why.

 

[roadkill]

Stewart handles kickoffs. He had the short kicks against UVA and one against UNC as well. No idea why.

According to the play-by-play, Birr handled the first three kickoffs in the UVA game. They weren't great. Then we went back to Stewart. I wonder if Stewart has a nagging injury that might be cause for holding him out? Or maybe it's an accuracy vs. distance thing - Stewart has been getting touchbacks pretty consistently.

On this topic regarding stats and rankings, FEI has been updated through the UVA game. Our special teams have gone from dead last in FBS last year up to 45th this year, a bit better than average. That's better than I had hoped for after last year's debacle.

 

[Sarrick]

Shanahan might be the most improved player. From almost getting blocked every punt to a lot of great punts this year

 

[colton]

Usually, the team that wins plays a more efficient overall game—more first downs, more yards per play, etc. That’s why the right side above the axis has much more area than the section below the axis on the left.

To the left of where the curve crosses the axis, that’s “we had more yards and first downs and time of possession and big plays and still lost. How did that happen?”. Often it’s turnovers—we drove to the 2 and fumbled. Sometimes it’s settling for field goals (and sometimes missing). Sometimes it’s punting from the opposing 40 yard line a bunch.

For our Miami game, we were the furthest to the left, therefore lowest, by a considerable margin.

 

[slugboy]

Some rushing stats from Game on Paper, post UVA.
SR= Success Rate (how well we stay ahead of the sticks)
EPA = Expected Points Added (as you get closer to the goalline, you're adding more expected points)

Dontae's stats have moved up a lot.

Player	Carries	Yds	TD	Fum	Yds/rush	EPA/rush	EPA	SR
Jamal Haynes	85	555	6	2	6.53	0.13	10.83	44%
Haynes King	44	445	4	0	10.11	0.48	21.10	57%
Dontae Smith	42	295	3	0	7.02	0.37	15.69	52%
Trey Cooley	41	179	1	2	4.37	-0.26	-10.65	29%
Evan Dickens	9	44	0	0	4.89	-0.54	-4.84	33%
Zach Pyron	4	10	0	0	2.50	-0.22	-0.87	50%

 

[Lotta Booze]

 

[slugboy]

DJU really improved after leaving Clemson, huh?

 

[colton]

This doesn't make any sense! The computer and TV networks told me that Carson Beck was elite and a Heisman Trophy contender! On a serious note, we seriously have something special with Haynes King. As a bonus, he has two years left of eligibility. With a 12-team playoff coming soon, expect national impact in 24 and 25.

 

[Jacket20]

This doesn't make any sense! The computer and TV networks told me that Carson Beck was elite and a Heisman Trophy contender! On a serious note, we seriously have something special with Haynes King. As a bonus, he has two years left of eligibility. With a 12-team playoff coming soon, expect national impact in 24 and 25.

These are pass + rush. Beck doesn't run.

 

[ibeattetris]

BCF Toys - 2023 Game Splits

www.bcftoys.com

for the first time this year, both our offense and defense produced positive value in the same game (might as well shout out special teams as well for being positive).

We also held a team to below 50% of DAY for the first time this season (41.9%).

Our net starting field position was +12.5 and we won the turnover battler.

This was by far the most total team win we have seen since 2018 UNC/Louisville. edit (2021 UNC was also a complete team stomping)

 

[ibeattetris]

Man, looking at some of our past games is wild. The clock rules really have sped up the game. I was surprised to see our 2020 Duke game has positive Defense Value, but then I saw each team had 18 drives. https://www.bcftoys.com/2020-ps-w13#387 . The longest drive we had was 7 plays. The median drive was 4.5.

We haven't crossed 14 drives yet this season.