I've been dropping advanced box scores for the last few games, so I decided to put everything into an 'advanced report card' that summarizes games 1-6, hopefully to see trends. Here are screenshots:
Glossary:
- Scoring Opportunities = Trips inside your opponents 40 yards line
- Success Rate = Percentage of plays where you get 50% of first down yardage OR 70% of second down yardage OR 100% of third or fourth down yardage
- Standard Down = any down where the defense does not know if you will run or pass (*insert offensive coordinator joke*), such as any first down, any second down with 7 or less yards to gain, any third or fourth down with 4 or less yards to gain
- Passing Down = any down where the offense needs to pick up significant yardage to stay 'on track' - this includes any offensive snap that is NOT a standard down
- Power Down (AKA 'Money Down') = 3rd or 4th down with 2 yards or less to gain OR any goal line play 2 yards or less to go.
- Explosiveness is calculated based on Collegefootballdata.com's PPA metric (same thing as EPA). Basically:
- Every possible down, distance, and yard line is assigned a point value based on historical data (I think 5 years of every FBS snap goes into this particular model), this is called Predicted Points (or Expected Points). Here's a good piece if you're interested in the specifics of calculating PP.
- The delta between your starting PP and ending PP is PPA. It can be positive or negative
- Explosiveness is the average of the positive PPA; a small PPA represents a minor situational improvement. A big PPA represents a significant situational improvement. For context, against Marshall, Tuten had 27 yard rush on 2nd & 9 from our own 42 that had PPA = 2.73. In the same game, Jaylin Lane gained 7 yards on 1st & 10 from the Marshal 39, resulting in a PPA = 0.41.
- Line Yards attempts to measure the number of running yards which are attributed to the offensive line. Yards are weighted as follows:
- losses - 120%
- 0-4 yards from LOS - 100%
- 5-10 yards from LOS - 50%
- 11+ yards from LOW - 0+
- Second Level Yards = the average yards per carry that are between 5 and 10 yards from the LOS.
- Open Field Yards = average yards per carry that comes from over 10 yards past the LOS
Callouts:
- Sacks are counted as pass plays (because typically if a QB is sacked, they are trying to throw downfield.
- None of this is opponent adjusted (like SP+, for example)
- On Money/Power Downs - the service providing the national average doesn't include punts. I think that's dumb, because IMO a punt is not a 'success'. That's why the VT average is significantly lower than the national average. I'll update this soon, but I'm annoyed by it.
- I'm creating this from scratch from play-by-play data. I'm seeing some (small) differences between sites (ESPN, Collegefootballdata.com) - specifically in Yardage, and Success Rate, etc
Other:
- This is my first time putting all this data into this 'Report Card' format, so I'm open to feedback.
- If anything looks fishy, let me know and I'll take a look. There's a lot of spreadsheet mapping going on, it's very possible mistakes were made.
- If you want to poke around the master spready, you can view it here.
- I really wanted to use conditional formatting to color code based on how far VT is compared to the national average, but that was too tedious for now
- I honestly haven't had a chance to digest/reflect on this information, so no takeaways yet, but I will at some point this week. Looking forward to others' thoughts.
Forums:
DISCLAIMER: Forum topics may not have been written or edited by The Key Play staff.
Comments
My big takeaway is that we are abysmal at Money Downs. Seems to track with all of our backbreaking mistakes throughout the season
See my callout above:
So let's say you have 10 money downs, convert 4, and punt on 5. I would count that as a 40% success rate. CFBData counts that as 80%.
I'll update soon so it's standardized.
Ah I did miss that part. So are we getting burned on that stat by going for it on 4th and not converting instead of just punting away? Even if we were on the opposing 35 and most analytics point to that being the smart play.
I wouldn't say we're getting 'burned', I'm just defining the stat differently.
Okay - so I broke this section out on it's own. Hopefully this paints a better picture.
On offense, we're slightly below average (again - not accounting at all for strength of opponent).
On defense, we're performing noticeably better than the national average (again, not accounting for strength of opponent).
I've really enjoyed all of these breakdowns you've done throughout the year and this one is no exception. I'll be revisiting this one throughout the day I'm sure.
I have a couple comments/questions:
1 - there seems to be a phantom point for Rutgers in the VT on Defense table
2 - how come the national averages for offense and defense are different? Shouldn't it be zero sum? If you're truly looking at national averages, then the average scores gained by the offenses should equal the average scores given up by the defenses. The only thing I can think of is that you're averaging only FBS teams and teams that play against FCS teams get padded stats, skewing those averages for both offense and defense.
3 - are Second Level Yards kind of indicative of a poor job of the OLine getting to LBs?
It appears our defense is pretty mercurial. We're better than average in most categories (all of basics except points/opp, all of success rate except rushing). Interestingly, our defense is better than average on explosive rush plays, which doesn't align with the eye test for me. Otherwise, we suck at explosive plays and rushing. That much checks out.
I think it's also interesting that, offensively, we're better than average for Yards Per Carry, but worse than average for Line Yards, Second Level Yards, AND Open Field Yards. How did we manage that?
This is good work, bar! Thanks for putting this together and sharing. It's been fascinating to follow over the season. The way I see it, our defense is slightly better than average and our offense is slightly worse than average. Explosive plays are killing us on defense and we're not getting enough explosive plays of our own on offense.
Yeah, of we aren't last in untouched 60 yard rushing touchdowns allowed on total clown breakdowns, I'd hate to be a fan of the team that is. FFS.
Good catch! Fixed.
You are correct, this was a copy/Paste formula error - fixed!
It's always tough to attribute success/failure of a specific stat to a specific match up/position group, but I think it's fair to say that poor Second Level Yards means that the defense is winning battles in 'the second level'.
That could be a variety of things:
In general, these stats are great for evaluating how a team does in a given situation; not necessarily how one player/unit performs.
Hopefully my answer isn't too vague.
Good question - for context, I'm taking play by play data for VT games and calculating these numbers, BUT for national averages, I'm just taking it from CollegeFootballData.com (or possibly doing a crude calculation). So, it's possible that there's a mistake on my end.
Another possibility is that there are just a lot of negative/short yardage runs - every rush that results in negative yardage is -120% line yards. The rationale for this is that, if the ball carrier is tackled behind the LoS, it's not their fault; it's the line's fault. In no scenario are positive yards represented by >100%. Looks like we're averaging over 4 negative runs per game (that does not include sacks).
I'm pretty confident it's the latter, because when I calculated rushing values for the FSU game, it matched CFBD's exactly, but, I could be missing something.
The big red flag on defense IMO is points per opportunity. An opportunity is any time that a team gets inside the 40 yard line. The fact that our defense is averaging >7pts/opp indicates that teams are scoring before they cross our 40 yard line.
When you take that factoid, and match it with what you see in the game (explosive plays everywhere) it makes sense.
Now, that value is skewed heavily by the Pitt and FSU games - Pitt had 2 scores, but only made it past our 40 once. That's really unusual. FSU was a rout vs a top 5/10 team, so it's not absurd that we got dominated there. This is where averages can be a bit misleading - the issue exists, but it's not as extreme as the average suggests.
Circling back to this comment:
I'll double check my math here.
right, I saw that and figured it was yet another indicator that we give up lots of explosive plays
Well, I think Shelton answered your question way better than I did/could have:
I was going to ask if it was some kind of bi-modal situation on runs. Either we're doing pretty well stopping them early, or we're getting runs housed against us. That would track with the big runs getting at least somewhat swamped out in the explosiveness metric when it all gets averaged.
Yep - what I see in the numbers and the games (I say this as someone with zero X's and O's knowledge) is a break-don't-bend defense. We bend a little more and break a lot more than a foster defense. We force plenty of 3 & outs, and have a complete shutdown/shutout quarter at least every other week. But then, we're good for giving up at least one long run each week
I'm not exactly sure which of my questions this is answering...
but it does make a lot of sense. My eye test tells me that we're okay against the run, generally, until we miss a gap fit and let someone run 60+ for a TD. Those explosive running plays are totally killing us. Take those plays away, and our running defense is pretty stout.
I have a question about "Money Downs."
Let's say that team A has the ball for three "Money Downs" and team B is on defense.
Let's say that team A converts on two out of three "Money Downs." I know that team A's "Money Down" percentage is 66.67%. Is team B's defensive "Money Down" percentage 33.33% because team B was successful 1 out of 3 times, or is team B's defensive "Money Down" percentage 66.67% because they're allowing success on 66.67% of the time?
Think of it this way:
Make sense?
so it'd be 66.6% for both
Yes - Team A's Offense and Team B's Defense would each have a 66% success rate.
Thank you for explaining that. I really enjoy these breakdowns.
Based on the pattern of scoring we should expect to drop 37 points on Louisville.
I'm cool with that.
and, hopefully, we hold Wake to 14, or fewer
Sure love to see a game by game breakdown of the breakdown of points scored by various units during the heyday of Beamerball. The defense and special teams averages would seemingly have been WAY higher...
a few excerpts from the 2009 media guide- roughly 20 years into the Beamer era:
-defense scored 79 TDS- 51 on pick sixes; 26 scoop and scores and two fumble recoveries in the end zone
-special teams scored 41 TDs on 15 blocked punts, 16 punt returns, 5 kickoff returns, four blocked field goals, and
one fumble recovery
-74 different players on special teams or defense had scored- 20had 2,5 had 3, Andre Davis had 4, Macho Harris had 5, and Deangelo Hall had seven!
-from 1993/2008 VT was 66-12 when scoring at least one non-offensive TD.
-total blocked kicks to that point(through 2008 season)-122 (60punts,38 FG, 24 PAT's)
I really miss being good at football.
Looking forward to seeing if VT will beat Ferentz's climb to 325.
VT seems to be trending towards average on most metrics, remarkable IMO considering our start.
When we score 17 or less lose. When we score more than 17 we win.