Back at it again, nerds. Enjoy:
Instead of retyping the glossary, I'm just going to point to CFBD's Glossary.
Any feedback on the format, the data displayed, etc is welcomed and encouraged!
Forums:
DISCLAIMER: Forum topics may not have been written or edited by The Key Play staff.
Comments
Money downs seems to be an area of strength for us. That said, I'm surprised we only had 2 opportunities in this game
Yes... I'm a big proponent of going for it on 4th & 1-2. If you convert 50% of the time, it's just about always the right move once you pass your own 45.
Found it interesting that since he actually respects the Syracuse program, Norvell didn't run a chicken shit 2 point play to open the game Saturday.
Umm, that's a stretch unless he personally told you his respective feelings on the two programs. I think it's more likely he felt they wouldn't need it since the 'Cuse has been playing so horrifically bad in conference.
Disagree- if going for 2 is the right call there- you go for 2- against VT or Bama or anyone.
Granted I wasn't part of the initial discussion on this one, but there are a dozen different reasons why they may have done it against VT and not against Syracuse that have nothing to do with how much he respects each program. That's all I was trying to say.
Edit: I upvoted the above to counter the downvote since it's an opinion. Otherwise I would have left it alone.
You still haven't let this go?
I just said I found it interesting. The argument was made on here that he did it because it was the "right" call. Well? if its the right call to go for 2, you do it every opening series. Like Oregon used to do. In my opinion- nothing more- Norvell didn't and does not take VT seriously. Which is fine. We need to get to the point where we can compete with them.
Counter: Norvell did it against VT because he does take us seriously, and he thought that if he didn't put the game away early, we could win it. But he does not take Syracuse seriously, so he didn't think it was important to do then
Maybe-but given recent history, that does not seem plausible. Cuse has been much better than VT the past 4 years
ironically enough, Syracuse and VT both have 17 wins in the last 4 seasons. If you want to go back 5 seasons, Syracuse has 2 more wins than VT.
Syracuse has a history of looking good in their first 6-7 games and then looks like garbage in the 2nd half of the season.
Not ironic- I am just a moron. It "seems" they have been much better. I was dead wrong. Should have looked it up.
First the Kenny Pickett thing, now this --- is DC finally not just doubling down repeatedly on being wrong about facts by moving the goalposts?
i am here for it
Wrong opinions are just that- wrong opinions. Eye of the beholder/subjective. Too many people on here get wrapped up in that, IMO. I can't recall ever saying something like VT was 10-2 in 2022 and doubling down on it or VT has won its last 10 against Syracuse in football, etc. When I do, I typically correct myself. But I do appreciate your post.
Cool- it took you 6 days to move away from "its the right call".
Sorry for being so snarky, I couldn't resist.
For what it's worth, I do think it's the right call. What I don't care about is what it says about Norvell (or any other coach for that matter). I don't care if other coaches respect us or not. I care about getting wins and improving.
Semi agree. I think its a good idea from your own 45 until your opponents 25. Then depending on how good your power running game is it might be a good idea again inside the 10. From our perspective I think inside the 25 taking the points is the better decision.
Cool stats I'd like to point out:
The inability to pick up first downs and run clock in Q4 has cost us several games the last 2 years. Really glad to see the team stepping it up in that regard. That's exactly how we start turning those close losses into wins.
Bar, I want to ask you about points per opportunity. Collegefootballdata's advance box score has us at 3.83 points per opportunity. They excluded Lane's 75 yard TD from their data on scoring opportunities, essentially saying "we scored from outside the 40, so it wasn't a scoring opportunity."
Conceptually, what should points per opportunity be measuring? Points scored/opportunity or points scored on drives inside the 40/drives inside the 40? It seems to me that if it's just points scored, and not points scored from inside the 40, that each and every drive should be considered an "opportunity."
I should have done the same... let me look back at how I calculated everything.
UPDATE #2: I did not (and have not) been ignoring scores from outside the 40. Part of this (not gonna lie) is a mistake by me. BUT, the upside is that it does a good job of illustrating our break-but-only-sortof-bend defense. Seeing Pitt averaging 14 points/Opp was jarring, but it showed that teams aren't getting to the 40, but are scoring because they are breaking plays that go for >40 yards
Points scored from inside your opponent's 40 yard line. The idea is that the 40 yard line is a better indicator of success than the 20 yard line (the traditional redzone)
That is points per drive. I'll do some digging to determine why your opponent's 40 yard line is considered a 'scoring opportunity', but being at the 45 or 50 is not. I'm not sure if it's arbitrary, or if it's backed by something statistical (eg; EPA).
Update #1: from page 158 of Football Study Hall by Bill Connelly
He has some simple tables in there that illustrate the point, but basically points per 40 yard opportunity and points per 5 yard opportunity are most predictive of a win - more so than points per red zone (20 yard) trip or points per drive.
I always got the impression that "Red Zone" was hype for creating viewer excitement.
I feel like some old coach noticed that getting inside the 20 or so was better for winning but never really looked at the stats to back it up and just started coaching inside the 20 because he won and thus he was correct. So then other people caught on and no one actually looked at stats because well sports don't need needs. So then it caught on and tv talked about it and made graphics over it and no one actually dug deeper until 2018.
The whole "scoring opportunity" thing seems really interesting. If you had great athletes and had two or three plays where you scored on a long pass or break-open run (or multiples of either), it seems like you'd be "penalized" in the metrics. (My brain seems especially foggy today, so hope I'm thinking about that right).
Also, did I miss in the glossary the definition of "money down"?
I wouldn't think of it this way... instead I'd ask yourself what you're looking to learn from the metric:
The ideal way is probably multiple stats, including:
This gives you the best picture of what actually happened.
On a related note, if you ever listened to Podcast Ain't Played Nobody (RIP) back in the day, you'll remember box score bingo with Bill Connelly where they went through the stats and he told you exactly how the game went. If you were not around for OG PAPN, here's a random episode where they play the game.
Money Down = Power Down. Power downs measures the success rate on 3rd or 4th and <2, or when the team is <2 yards away from a touchdown. However, 'Power Downs' ignores plays where the ball is punted away. I think this is dumb, which is why in included the 'conversions per opportunities' stat, and rename the category.
I still think if you have the time, including averages would be interesting. Are we performing above or below how we usually do? Is the opponent performing better or worse than usual? Maybe at the end of the season when we're talking explosiveness rate, money down success rate, and money down explosiveness. That would give each team a full season of data that we can compare our performance to.
I have some ideas. Let me see if I can spin something up for the next game.
Several ideas: comparison against last years averages, maybe something like a running average of last three games this season (so would move) against first 3 games this season, highs this season vs high of last season and low vs low. And if high and low are meaningless, maybe take best game vs best game and worst vs worst (actual choices may be subjective unless you use another metric as a benchmark).
Still very much a work in progress, but how about something like this, where you can see the percentile that a given stat exists in:
So looking at the highlighted row, we see that VT had a Success Rate of ~42% against Wake Forrest. This success rate was better than 67% of our past performances, and is a median success rate compared to all of FBS.
I like the information this shows, but I feel like it's a little overwhelming to look at. Thoughts?
Edit: Also, I'm not positive the last column is correct... I blindly pasted a formula from ChatGPT, so it could be a little sus.
As not a math guy, percentiles wouldn't necessarily be as effective as game % vs. season average %
After Wake, Tech is so close to the Ferentz line. Currently averaging 24.4 pts.
Think we'll beat that?
It's going to be tight. We've proven that in a loss we score 17 or less and it looks like a win is 30+.
If we finish 6-6 🤞 and we get 30 in each of the wins and 17 in each of the losses we finish at 24.58.
If we go 5-7 with all games being 17 or 30, but hang a 50 burger on UVA, we finish at 25.16.
You can obviously slice it a bunch of other ways, but based on what we've seen so far it seems like we're going to be hovering right around that magic 25 number.
So we're gonna hang 50 on UVA, got it!
Love burgers that time of year.
38-0, bro
Ooo, gotta do the math on 6-6 with a 38-0 for Thanksgiving.
Edit: I shit you not, that rounds out to 25.25.
I mean we did that and more in the '83 squeaker my freshman year...48-0 in Lane North AND took the goalpost to boot!
According to SP+, the D ranking of teams left on VT's schedule:
Not murders row, but not too many opportunities to run up the score. For comparison, Pitt is 40, FSU is 16, and WF is 54.
3-2 Baby!
I'd take it as a step in the right direction.
I would even be content with 2-3 (since most of our remaining games are away) so long as we are competitive and one of those wins is Little Brother.