Finding the sweet spot for Missouri’s offensive pace

• Going fast is good for Missouri’s offense. Anybody can see that. Just point to the 740-yard performance against Tennessee last season in which the Tigers ran 3.53 plays per minute.
• Going fast is bad for Missouri’s offense. Anybody can see that. Just point to the 265-yard performance against LSU last season in which the Tigers ran 3.44 plays per minute.
• Going slow is good for Missouri’s offense. Anybody can see that. Just point to the 455-yard performance against Florida last week in which the Tigers ran 2.28 plays per minute.
• Going slow is bad for Missouri’s offense. Anybody can see that. Just point to the 340-yard performance against Auburn earlier this season in which the Tigers ran 2.73 plays per minute.

So...what am I trying to say here? Going fast is good...and bad. Going slow is good...and bad.

What’s the sweet spot? What’s the tipping point, where it’s fast enough to be just good without getting out of control?

Well, that’s hard to answer. And, really, the answer doesn’t lie in merely plays per minute.

First, allow me to introduce you to my new best friend: Missouri’s pace of play spreadsheet.

It takes the 15 games the Tigers have played against Power-5 competition in the past two seasons under coordinator Josh Heupel and divvies them up into a number of categories:

Rush Attempts, Yards, Average and Play Distribution
Pass Attempts, Yards, Average and Play Distribution
Total Plays, Yards and Average
Time of Possession (TOP)
Plays per Minute (PPM)
Seconds per Play (SPP)
Dead-Ball Plays (DBP...incomplete passes minus interceptions)
Live-Ball Plays per Minute (LBPPM)
Difference between PPM and LBPPM (LB Diff)
Plays for 0 or Negative Yardage (0 Plys)
Pct. of 0 Plays

Besides giving you more ridiculous acronyms to parade around (you down with SPP? well, yeah, maybe), I hope all these categories give you a more complete view of how the Tigers’ offense works when it does and doesn’t when it doesn’t.

You should notice a couple of differences between last year’s offense and this year’s:

— This year the Tigers are a little worse at running and a little better at passing
— This year the Tigers’ efficiency numbers are better even though their bulk numbers are worse.
— This year the Tigers are going 3.5 seconds per play slower than they did last year.

Missouri ran fewer than three plays a minute in only two of nine outings against Power-5 teams last year. This year, the Tigers are 3-for-6.

The Tigers’ fastest pace this year, 3.19 plays per minute — against Purdue and Kentucky...again, the dichotomy of pace — would have tied for sixth-fastest in nine Power-5 games last year.

So what is the best predictor of Missouri’s offensive success?

We looked at this in a couple of different ways: total yards and yards per play.

We ran regression tests for four independent variables — plays per minute, live-ball plays per minute, live-ball play differential and percentage of zero plays -- to see which one predicted the two dependent variables (total yards and yards per play) for the entire population of games, just 2016 games and just 2017 games.

Then we got expected values for those dependent variables based on the regression tests and saw which predicted results had the lowest standard deviation from the actual results to see what was the best predictor.

But before you scream, “SHUTUP NERD,” you can just disregard those last three grafs and think about it this way.

We made projections using four different factors and saw which factor’s projections got closest to reality.

Here’s what we got:

Whole Population
Live-ball plays per minute was the best predictor for total yards in a game, with a standard deviation of 107.6 yards.

The five closest matches were:

1. 2016 Arkansas (Actual: 399; Expected: 397.6; Difference: 1.41)
2. 2017 Auburn (Actual: 340; Expected: 346.0; Difference: -5.96)
3. 2016 South Carolina (Actual: 465; Expected: 454.4; Difference: 10.6)
4. 2017 Georgia (Actual: 312; Expected: 299.5; Difference: 12.5)
5. 2017 South Carolina (Actual: 423; Expected: 405.3; Difference: 17.7)

So the sweet spot for repeatable success is between 1.96 and 2.56 live-ball plays per minute (or 2.40 and 3.19 total plays per minute).

Live-ball play differential was the best predictor for yards per play, with a standard deviation of 0.89 yards per play.

The five closest matches were:

1. 2017 Georgia (Actual: 6.37; Expected: 6.41; Difference: -0.04)
2. 2016 Georgia (Actual: 6.28; Expected: 6.20; Difference: 0.08)
3. 2016 South Carolina (Actual: 6.12; Expected: 5.97; Difference: 0.15)
4. 2016 West Virginia (Actual: 4.62; Expected: 4.83; Difference: -0.21)
5. 2017 Florida (Actual: 6.79; Expected: 7.03; Difference: -0.21)

So the sweet spot for repeatable success is between 0.17 and 1.12 live-ball plays per minute (or 2.28 and 3.88 total plays per minute...or the lower and upper bounds of this study...lol).

Not very helpful, that.

2016 Season
Live-ball plays per minute was also the best predictor for total yards in a game, with a standard deviation of 90.3 yards.

The three closest matches were:

1. Florida (Actual: 363; Expected: 354.6; Difference: 8.42)
2. South Carolina (Actual: 465; Expected: 434.7; Difference: 30.3)
3. West Virginia (Actual: 462; Expected: 519.2; Difference: -57.2)

So the sweet spot for repeatable success is between 2.38 and 2.75 live-ball plays per minute (or 2.94 and 3.88 total plays per minute).

And live-ball play differential was again the best predictor for yards per play, with a standard deviation of 0.54 yards per play.

The three closest matches were:

1. South Carolina (Actual: 6.12; Expected: 6.11; Difference: 0.01)
2. West Virginia (Actual: 4.62; Expected: 4.49; Difference: 0.13)
3. Georgia (Actual: 6.28; Expected: 6.44; Difference: -0.16)

So the sweet spot for repeatable success is between 0.53 and 1.12 live-ball plays per minute (or 3.19 and 3.88 total plays per minute).

2017 Season
Zero-play percentage is the best predictor for total yards in a game, with a standard deviation of 113.6 yards.

The three closest matches were:

1. Auburn (Actual: 340; Expected: 368.5; Difference: -28.5)
2. Florida (Actual: 455; Expected: 487.4; Difference: -32.4)
3. South Carolina (Actual: 423; Expected: 365.4; Difference: 57.6)

So the sweet spot for repeatable success is between 14.9 and 34.8 percent zero plays (or 2.28 and 3.08 total plays per minute).

Zero plays were also the best predictor for yards per play, with a standard deviation of 1.07 yards per play.

The three closest matches were:

1. Georgia (Actual: 6.37; Expected: 6.02; Difference: 0.35)
2. South Carolina (Actual: 6.13; Expected: 5.67; Difference: 0.46)
3. Florida (Actual: 6.79; Expected: 7.33; Difference: -0.54)

So the sweet spot for repeatable success is between 14.9 and 34.8 percent zero plays (or 2.28 and 3.08 total plays per minute).

---

So...what has the point of all this been? I don’t really know.

The answer is it probably behooves the Tigers to go faster against some teams, slower against others. It certainly has behooved this year’s team to go about 16-percent slower than last year’s team, on average, and last week’s slow it down, ground it out strategy vs Florida would seem to be a good way to go against the final three opponents this year...all of whom are bad at run defense.

If we take those nine most representative games over the past two years from those three categories (whole population, 2016 season, 2017 season) above, the average plays per minute is 2.96 — one every 20.3 seconds — and 2.38 live-ball plays per minute.

So, if Missouri is holding the ball for 20 minutes, look for 59 plays and 12 incompletions.

If Missouri holds the ball for 25, look for 74 plays and 15 incompletions.

If Missouri holds it for 30, look for 89 plays and 17 incompletions.

You get the picture. Not too fast, not too slow. Just right.

And, if you’re not asleep yet, you can look at my work: