Advanced Football Stats: Explosiveness

In a June series for WKUInsider, we're exploring the world of advanced football statistics. Read inside for the first in our five-part series.

In a June series for WKUInsider, we're exploring the world of advanced football statistics.

Similar to tempo-free stats for basketball, these numbers are more complicated than the average metrics used, but tell the story of a team (or player,) much more accurately.

SBNation's Bill Connelly, an expert on advanced football stats, says there are five factors that makeup college football's most important measurements.

Today, we'll look at his strongest one--"explosiveness".


According to Connelly, if a football team wins the "explosiveness" statistic in a given game, that team wins an astonishing 86 percent of the time.

Measured by PPP (Points Per Play,) this measurement is exactly what it says--How many points a team gains per play. Yards per play can also be used.

Taking out the field position factor, and looking strictly at yards per play game data from 2013, if a team outgained their opponent by just the slim margin of .5 to 1 YPP, they won 72 percent of the time.

Wider margins only increase the win probability. Teams who outgained their opponent between 2 to 3 YPP (happened in 14.8 percent of 2013 contests,) won 95 percent of the time.

How does a team increase their yards per play? Obviously with big plays, and/or lengthy gains—Thus, "explosiveness."

A key to remember is big plays reduce the number of mistakes a team can make. A 12-play, 50-yard scoring drive might demoralize an opponent, but leaves several opportunities to give up a turnover, or take a sack that decreases the odds of getting the next first down, or forces a field goal instead of touchdown. A big play reduces the chances you can make a mistake, and gets points on the board.

Here's a breakdown of Western Kentucky's 2013 schedule for reference. YPP from the actual games played are indicated in parenthesis. Note-Some of the "estimated win probability's" are my best guesses, so could be off by a few percentage points.


WKU (6.5), Kentucky (7.0) :

Margin = -.5

Estimated win probability = ~35 percent.

Actual result = Win


WKU (5.6), Tennessee (6.0) :

Margin = -.4

Estimated win probability = ~40 percent.

Actual result = Loss


WKU (5.8), South Alabama (6.3) :

Margin = -.5

Estimated win probability = ~35 percent.

Actual result = Loss


WKU (8.6), Morgan State (5.1) :

Margin = +3.5

Estimated win probability = 100 percent

Actual result = Win


WKU (5.6), Navy (3.2) :

Margin = +2.4

Estimated win probability = greater than 95 percent.

Actual result = Win


WKU (7.0), ULM (5.8) :

Margin = +1.2

Estimated win probability = ~85 percent.

Actual result = Win


WKU (7.0), ULL (5.4) :

Margin = +1.6.

Estimated win probability = greater than 90 percent.

Actual result = Loss


WKU (6.0), Troy (6.3) :

Margin = -.3

Estimated win probability = ~42 percent.

Actual result = Loss


WKU (5.9), Georgia State (6.0) :

Margin = -.1

Estimated win probability = 45 percent

Actual result = Win


WKU (5.2), Army (6.0) :

Margin = -.8

Estimated win probability = ~30 percent

Actual result = Win


WKU (6.1), Texas State (2.2) :

Margin = +3.9

Estimated win probability = 100 percent

Actual result = Win


WKU (6.2), Arkansas State (4.4) :

Margin = +1.8

Estimated win probability = greater than 92 percent

Actual result = Win


So YPP (not as standard as PPP,) still managed to get 8 of WKU's 12 games correct. Just the one stat by itself, not even in its most accurate version. Pretty impressive.

The four games predicted incorrectly--Kentucky, ULL, Georgia State and Army—all have unique stories to tell.

For example, WKU dominated most of the game against UK, leading 35-17 with 13:32 left in the 4th quarter.

After Leon Allen's 14-yard touchdown run gave the Toppers their late 18-point lead, WKU led the YPP stat 7.0 to 5.5. A 1.5 YPP margin—-Big gap.

In (essentially) the garbage time that followed, the Wildcats moved the ball 154 yards in 20 plays—a 7.7 YPP average—and the Toppers managed only 51 yards on 16 plays—a 3.1 YPP average. That was enough to give Kentucky the lead in the stat by game's end, still eventually losing 35-26.

The biggest outlier was the Louisiana-Lafayette loss. Despite outgaining the Ragin' Cajuns by 1.6 YPP, giving the Toppers over a 90 percent probability to win, they fell 37-20. How?

WKU led 10-0 in the second quarter before giving up a 99-yard interception return for a touchdown. That same quarter, ULL also ran back a fumble recovery for 45 yards, from their own 20-yard line to WKU's 35-yard line. They scored a second touchdown quickly after, taking a 13-10 lead without having to move the ball much.

WKU had two other turnovers in the game to help ULL's efforts, and that doesn't even include finishing 0-for-4 on fourth down.


So "explosiveness"-—while very accurate-—is certainly not the "end-all, be-all" stat in a complicated game.

But as far as the box score goes, it's usually a great place to start.

Connelly says "explosiveness" makes up about 35 percent of his "five factors" team success formula. In our next analysis, we'll look at his second-most important measurement—-Efficiency.

For more from Connelly, check out his book, available for order on Amazon.com.

Stay tuned with WKUInsider.com for the latest in WKU recruiting and team news.


WKU Insider Top Stories