This one is a little late due to life and some cloud sync snafus getting in the way.
2014 Opponents: William & Mary, Ohio State, ECU, GT, W. Mich, UNC
2015 Opponents: Ohio State, Furman, Purdue, ECU, Pitt, NC State
THE BIG TWO: TOTAL O & SCORING O
After that uplifting, rejuvenating win to climb back to .500 in ACC play and overall, both total and scoring offense trended...down? The Hokies amassed 358 yards and 28 points, both of which were below their 2015 averages to date heading into the game. This year's offense is now outscoring last year's by a razor-thin 0.5 points per game. Meanwhile the delta in total offense has narrowed somewhat thanks to a steeper week-to-week downward trend last season. This season's offense is currently gaining 32.8 fewer yards per game than at the equivalent point of last season.
THIRD DOWN CONVERSION PERCENTAGE
We're holding steady at just under 40%, with no indication that significant movement up or down is impending. Through half the season, the greatest week-over-week delta (absolute value) has been 1.2 percentage points, with this week's trend being a very modest +0.3. A steeper negative trend at this point last season has narrowed the gap to -9.1 percentage points, the second-closest delta this season. But it remains clear that this Motley-led offense is inferior to its 2014 Brewer-led counterpart in converting third downs.
RED ZONE EFFICIENCY
Red zone efficiency is a stat I developed specifically for this experiment, so there are no national rankings available. (I'm not going to calculate RZE for every team in the nation, because I'm lazy.) But I can still tell you we're really good at it this season. We are 18th in the nation in red zone touchdown percentage, which is the anchor of my red zone efficiency calculation. We are finishing drives with touchdowns this season, and that is responsible for a big improvement in this area compared to last year. This week sees the biggest week over week improvement thus far this year, and the delta percentage when 2015 is compared to 2014 is now a season-high +11.3%.
YARDS PER
Yards per play held steady at 5.4 from last week. A downward week-over-week trend over the same period last year now has the 2015 offense outgaining last year's squad by 0.2 yards per play.
Brenden Motley did not have a great week throwing the ball against NC State (strictly in regards to YPA; the three second-quarter TDs to Ford were a thing of beauty). As a result, YPA takes its second straight significant week-over-week hit. The passing game is still more explosive this year than last, but 2015 is now outperforming 2014 by 1.1 YPA, after posting a season high delta of +2.4 after week three. If this trend continues, the passing game is in danger of surrendering all of its gains over last season.
Ah, that's better. After a concerning two-week slide in YPC by RBs, the McMillian/Rogers Express showed up in a big way to end the skid. After dipping below 2014's average last week, the 2015 offense is back on top again by 0.8 yards per carry. It is precisely the big play ability of Rogers and McMillian that led to this improvement. All else being equal, you go with the backs who have the ability to break the big run on occasion. Here's hoping they stick with that plan.
SUMMARY
Sometimes stats don't tell the whole story, and the more general the stat, the bigger a culprit of this it can be. Both total and scoring offense trended downward this week, but does anyone feel that the offense did not pull its weight and contribute to the win over NC State? Yet statistically, it wasn't a great week for the offense in the "big" metrics. Make of that what you will. I'll take the win and feel good about it.
At the beginning of the season, I circled third down conversion percentage and red zone efficiency as really critical stats. In terms of RZE, this team has shown big improvement over last year. Meanwhile, we lag pretty far behind the 2014 squad in converting third downs. As Brewer nears being medically cleared for live games, I will be paying particular attention to whatever effect his return has on these stats.
Comments
I hope you do this segment after the transition from Beamer to the new Head Hokie, whoever he may be. *cough*#TomHerman4HC*cough*
and of course, the first couple of years after that to see if there is steady improvement under the new regime. This has been a pretty cool installment...btw. So thanks for that :D
Hard to believe that we more than tripled our offensive output from last week... and our cumulative average STILL went down.
Darn you, Furman, and your upper outlier influence...
Point taken. However, Pitt was a much larger outlier than Purdue was. The 583 yards against Purdue represents 1.24 standard deviations from our average performance, while the 100 yards against Pitt was 1.69 standard deviations. They don't even cancel each other out.
I guess now I know how my students felt over the past few years when they would, for example, carry a 96 average through a semester, but then score a 93 on a test and watch their average plummet.
RE: the matter at hand, can we get some IQR calculations going or something?
See, and I don't know why, but that gives me comfort that Pitt was truly just an off day. I still believe our offense is better than the numbers indicate. I may very well have my orange and maroon glasses on too tight though.
Our offense is actually very, very good. It's just a matter of them remembering how to football once the game starts. We've had some trouble with that over the past several years.
(I say the above partially in jest, but how else would you explain last season?)
Just for reference, VT's opponents defensively this season: 24.2 points/game, 339.5 yards/game, 188 yards passing/game, 151 yards rushing/game. For now they're positive in points scored and yards gained relative to their defense. I say "for now" because I think NC State's going to drop like a stone.
You da man. I heart normalized stats.
We are about ready to hit that point in the season when the 2014 offense dropped like a rock. So, yay, we will catch that offense.
I am still concerned that we don't have a running game to lean on and Loefflers east-west offense takes too long. But we do have good WR/TE's, I think good OL and if Brewer can unseat Motley, that means that Brewer is pretty good (I expect both will play in rotation the rest of the year, best QB playing the 2nd half).
We are a better offense than last year, by a significant margin. But that's a low bar.
#FORESHADOWING
We were terrible in the 2nd half. If I recall correctly, our ridiculously high 3rd down percentage normed, which was either the cause or effect of a terrible offensive 2nd half.
Effect. If your performance on 3rd down>performance on 1st and 2nd down that's generally unsustainable. That's also part of why I don't like using 3rd downs as a benchmark for an offense- it can improve/decline just due to dumb luck.
And it's really small sample stuff. You might have 10 3rd downs in a game. Big difference between 3 of 10 and 5 of 10.
Which makes the stability of our third down conversion rate thus far all the more interesting.
I feel crazy saying this considering my long held convictions about Scot Loeffler, but I really expect our offense to improve greatly down the stretch.
{shudders}
let us hope!
I feel confident after watching the first half of this season that any offensive struggles we have are not schematic in nature. These players obviously get the big picture now and understand the roles they play in the offense. Now it's an issue of, can the individual positions actually do what they are being called upon to do in the scheme? Internalizing the scheme is the first major step, but if the scheme is innately too complex, you'll see a sputtering offense, one that looks great for stretches and then looks inept for stretches. And so far that's kind of what we've seen.
We're about to hit the meat of our schedule, and we'll know a lot more about what Lefty is actually capable of as an established OC come early November.
I still can't stand the fundamental concepts of the running game. Run. Down. Hill. Damnit.
It puts so much pressure on Malleck and Rogers to make nearly impossible blocks on every play.
Is the average length needed on 3rd down readily available? That might say something as well.
Not ignoring this comment. Still looking for a stat site that lists average distance per down.
You do recall correctly, at least for the next few weeks. Last year, we were 4-2 at this point and scored 22 points combined over the next two games.
All we need is a touchdown next week to match the offensive output of the same week last year. The PAT would put us above.
I still think we get too cute with the running game. How about a couple of toss sweeps? Pound rock from the I on an ISO maximizing Teller's brutality. I would like to see NIjiman start at LT and I would like to see more split back sets with 2 tight ends even if one is lined up at WR just to give Phillips and Ford a break.
Can you show how offensive stats for last season to see where the offense then went?
You mean post the chart for all of 2014 week by week?
That would be nice, maybe have the future points greyed out, but present, so we can visually see if we are following the same trajectory that had the wake forest game from last year (though I suspect we already had our wf game this year)
That would be great if it isnt to much trouble
No problem. I'll try to have it posted within the next 24 hours.
Here's a hint, though: it was all downhill from here last season.
thats what I suspected
This ain't no Netflix where you can binge-stat a whole season in day! You will just have to let the suspense eat at you for a week like all the other live-watchers.
Sorry if you've already explained it and not to stir any pots, but who do you consider "running backs" for the purpose of this analysis?
I think the opponents for last year's game 5 & 6 (w. Mich & unc) play heavily into the total O trend discrepancy.
I include both tailbacks and fullbacks in that stat. Basically if Shane coaches them, they're lumped into YPC-RB.
I would love to see rushing O, passing O, and total O tracked based on where we are in the FBS on a week by week basis. If that could be compared to last year as well I think a lot of people will be shocked by how good we are doing in comparison. (With our one game huge exception)
IllHokie, you said you developed RZE just for this exercise, how do you calculate it?
And all of these are partial/running averages?
It's a measure of how many "potential points" the offense scored based on their number of red zone trips. A red zone efficiency of 1.000 would mean a team scored a touchdown on every trip to the red zone. The formula is:
[RZTDs + (RZFGs x 3/7)]/RZ attempts
By assigning FGs a fixed value of 3/7 of a touchdown, I don't have to worry about missed PATs, two point conversions and other rarities. A touchdown is a touchdown, regardless of what happens on the try after, and a field goal is weighted to its practical proportional value vs a touchdown.
Everything on the charts is running averages through that week, vs the running averages through the same number of games last season. Variance in the schedules means there's isn't a perfect comparison last year to this, but that effect diminishes with additional games. Our OOC games were also very similar last year to this. OSU and ECU both years, swapping William & Mary/Western Michigan for Furman/Purdue.
Nice, thank you.