Stats models and rankings

roadkill

roadkill

Helluva Engineer
Messages
1,821
slugboy said:
I’ll answer the second question first—they used yards per play. The defensive stat is yards per play against. This year we were #118 in yards per play against at 6.2 yards/play. The year before we were at 5.4 (#63). Not only were we atrocious last year, but we plummeted from the year before.
(Three players graduating—two linebackers and one EDGE—were probably responsible for a lot of that)

Yards per play and points per play probably tell you a lot about our offense and defense, except for one thing—it’s not adjusted for strength of schedule. Ole Miss, for example, was one of the top 10 offenses, or just outside of that. If one team is playing teams like Ole Miss and Louisville while another one is playing UMass, it’s not a fair comparison.

Still, it’s not good.

=======

For being turnover-prone, I’d just go to EPA. There are others, but that one is easy and available. Turnovers end up being negative expected points added. A pick-six can be anywhere from -4 points to -14, depending on where on the field it happens (if it happens in the red zone at the 1 yard line, it’s about -14 points, if it happens from your 1 yard line, it’s about -4 points).

bestballstats.com

Expected Points Added: A Full Explanation

Expected Points Added (EPA) is a commonly used advanced statistic in football. In short, this stat measures how well a team performs compared to their expectation on a play-by-play basis. EPA and E…
bestballstats.com bestballstats.com

There’s a stat called “Bad Play Rate”, but I don’t see a place to find that. Ridder and the Falcons would have been awful on that stat. In fact, Ridder was #24 out of 32 QBs with an individual EPA of 28, while Sam Howell—who people thought played awful this year—was #19 at 49.7. Dak Prescott was #1 and Mahomes was #3.
Click to expand...
Thanks again for the insight. I see now that the Pick6 tweet stated yards/play - I should have read it more closely. Although, oddly, their rank is 5 places higher than the stat you provided. Hmmm.

Agree that yards/play can tell a lot about both offense and defense but without the context of SoS. So if you had to pick the best single metric for each, for ranking purposes, wouldn't it be something that adjusted for SoS? This would allow for a truer comparison among teams within the ACC, for example.
 

JacketOff

Helluva Engineer
Messages
2,953
roadkill said:
You keep repeating this as if you are arguing with someone who doesn't think our defense was bad last year. I'm not sure which poster you think that is.
Click to expand...
He just wants you to know that we had the absolute most atrocious defense that’s ever sniffed a football field last year, and there were zero positives that ever happened on that side of the ball. Even the slightest hint of credit to the defense should be met with shame. And if you think otherwise then you’re an idiot that’s never watched football before.

A bit Dunning-Krueger ish tbh
 

Root4GT

Helluva Engineer
Messages
3,032
roadkill said:
You keep repeating this as if you are arguing with someone who doesn't think our defense was bad last year. I'm not sure which poster you think that is.
Click to expand...
Being #62 or #75 is close to average in FBS football. Anyone saying that is where we were is saying we were not all that bad a defense. So yes people are saying we were just below average on defense when it's clear we were very very bad on defense and nowhere close to being an average defense.

I also find it humorous that several people are plugging using SOS as a reason we were not so bad on defense when the 2 best teams we played were SEC schools yet many of the same fonts constantly say the SEC is way overrated and their SOS is BS because they were so overrated. Oh the irony!

Either SOS is not real useful or the SEC was actually a very strong conference. I contend SOS is only marginally useful and the SEC was strong at the top 4-5 teams. Meh at best below the top 4-5 teams.
 
roadkill

roadkill

Helluva Engineer
Messages
1,821
Root4GT said:
Being #62 or #75 is close to average in FBS football. Anyone saying that is where we were is saying we were not all that bad a defense. So yes people are saying we were just below average on defense when it's clear we were very very bad on defense and nowhere close to being an average defense.

I also find it humorous that several people are plugging using SOS as a reason we were not so bad on defense when the 2 best teams we played were SEC schools yet many of the same fonts constantly say the SEC is way overrated and their SOS is BS because they were so overrated. Oh the irony!

Either SOS is not real useful or the SEC was actually a very strong conference. I contend SOS is only marginally useful and the SEC was strong at the top 4-5 teams. Meh at best below the top 4-5 teams.
Click to expand...
I'm glad you, like me, can find some humor in some of the posts on this board. :)

I think SoS is a factor. How much of a factor is a debate for some of the advanced stat folks. But I do think we defended two of the most potent offenses in the SEC and did rather poorly at it.

It's been shown that our run defense was abysmally bad (bottom quartile) by most any metric. I can't see how anyone would disagree with that. Conversely, as @slugboy showed, our pass defense was above average ("passable", in his words). Also, as I pointed out after the Miami game, we gave a textbook demonstration of bend-don't-break.

Again as @slugboy noted, we were very inconsistent. We had some games where the defense showed up. Others had us throwing things in frustration. It's the latter examples that tend to stick in our minds. It's human nature to remember failures more vividly than successes, and losses more than wins.

I think with our new coaching staff, hoping for a 20-rank improvement isn't too much to ask. Whether that gets us to average or to below average (up from "bad") is less important to me than seeing a significant improvement that translates into winning an extra game or two.
 

GetYourBuzzOn

Georgia Tech Fan
Messages
77
Bill Connelly's first offseason returning production ratings can be found here: https://archive.is/wQDTy

As always, he gives good insight into his process and how he has evolved it to quantify the transfer portal.

GT is 46th in the country at 66%. That should boost our power rating by 3 points or so, which is pretty significant!
 

Root4GT

Helluva Engineer
Messages
3,032
roadkill said:
I'm glad you, like me, can find some humor in some of the posts on this board. :)

I think SoS is a factor. How much of a factor is a debate for some of the advanced stat folks. But I do think we defended two of the most potent offenses in the SEC and did rather poorly at it.

It's been shown that our run defense was abysmally bad (bottom quartile) by most any metric. I can't see how anyone would disagree with that. Conversely, as @slugboy showed, our pass defense was above average ("passable", in his words). Also, as I pointed out after the Miami game, we gave a textbook demonstration of bend-don't-break.

Again as @slugboy noted, we were very inconsistent. We had some games where the defense showed up. Others had us throwing things in frustration. It's the latter examples that tend to stick in our minds. It's human nature to remember failures more vividly than successes, and losses more than wins.

I think with our new coaching staff, hoping for a 20-rank improvement isn't too much to ask. Whether that gets us to average or to below average (up from "bad") is less important to me than seeing a significant improvement that translates into winning an extra game or two.
Click to expand...
Good post. My only rejoined is our run defense was so bad teams didn’t need to pass against us very often. The UVA game was really the only FBS game where we had a big enough lead that the opponent was forced to pass to have a chance.

Team’s leads, trailing margins often impact the passing yards in a game. If you have a big lead teams are forced to pass. If you trail big you are forced to pass.

The 3rd down conversions rate by our opponents was very bad. I would like to have stats on distance needed for the first down and whether the 1st down was achieved via a run or pass and the percent success on running and passing.

Still you post was very good.
 

Root4GT

Helluva Engineer
Messages
3,032
57jacket said:
Confused Looking For GIF by Looney Tunes
We've beaten this to death. Isn't there another topic to discuss? LOL
Click to expand...
Please come up with one :ROFLMAO:
 
forensicbuzz

forensicbuzz

21st Century Throwback Dad
Messages
8,833
Location
North Shore, Chicago
Root4GT said:
Being #62 or #75 is close to average in FBS football. Anyone saying that is where we were is saying we were not all that bad a defense. So yes people are saying we were just below average on defense when it's clear we were very very bad on defense and nowhere close to being an average defense.

I also find it humorous that several people are plugging using SOS as a reason we were not so bad on defense when the 2 best teams we played were SEC schools yet many of the same fonts constantly say the SEC is way overrated and their SOS is BS because they were so overrated. Oh the irony!

Either SOS is not real useful or the SEC was actually a very strong conference. I contend SOS is only marginally useful and the SEC was strong at the top 4-5 teams. Meh at best below the top 4-5 teams.
Click to expand...
It seems that there is an invalid assumption being made by some fonts. Where a team defense is ranked relative to the other teams is not an indication of whether the defense is good, bad, so-so, or sucks. The rankings are a relative scale of all the teams. If the #1 ranked team isn’t very good, then they just means all the teams below them are worse. A team can be ranked in the statistical middle and still be a horrible defense. Rank location is not an indicator of quality.
 

Root4GT

Helluva Engineer
Messages
3,032
forensicbuzz said:
It seems that there is an invalid assumption being made by some fonts. Where a team defense is ranked relative to the other teams is not an indication of whether the defense is good, bad, so-so, or sucks. The rankings are a relative scale of all the teams. If the #1 ranked team isn’t very good, then they just means all the teams below them are worse. A team can be ranked in the statistical middle and still be a horrible defense. Rank location is not an indicator of quality.
Click to expand...
That’s not what many on this board say. But the point is our defense was very bad in the areas that matter. Anyone who has ever watched college football could easily see that. Trying to defend our defense in 2023 is silly at best.

We have turned a new page with the. coaches so hoping for a major step forward this year. Looks like we got a solid staff based on their past stops. Exciting turn of events.
 
ibeattetris

ibeattetris

Helluva Engineer
Messages
3,604
Man, I just found out today that the football outsiders site is completely dead. I'll always fondly remember checking every week for the updated F+ rankings.
 
slugboy

slugboy

Moderator
Staff member
Messages
11,490
ibeattetris said:
Man, I just found out today that the football outsiders site is completely dead. I'll always fondly remember checking every week for the updated F+ rankings.
Click to expand...
It got sold. The founder left. The acquiring company stopped paying people last year. Updates stopped in the spring, I think. By September 2023, the site was shut down.

Aaron Schatz (founder) is over at a sports book kinda site
 
Top