A brief look at “advanced” statistics from The Citadel’s 2016 SoCon campaign

This is a post primarily about the “Five Factors” of college football.

What are the Five Factors? I’ll let Bill Connelly of SB Nation explain:

…I’ve come to realize that the sport comes down to five basic things, four of which you can mostly control. You make more big plays than your opponent, you stay on schedule, you tilt the field, you finish drives, and you fall on the ball. Explosiveness, efficiency, field position, finishing drives, and turnovers are the five factors to winning football games.

  • If you win the explosiveness battle (using PPP), you win 86 percent of the time.

  • If you win the efficiency battle (using Success Rate), you win 83 percent of the time.

  • If you win the drive-finishing battle (using points per trip inside the 40), you win 75 percent of the time.

  • If you win the field position battle (using average starting field position), you win 72 percent of the time.

  • If you win the turnover battle (using turnover margin), you win 73 percent of the time.

Those percentages were based on 2013 FBS data. It’s now 2017, but they probably still apply. Connelly has made adjustments to some of the formulas that go into the five factors, but the basic principles remain the same.

What I wanted to do was see how The Citadel’s 2016 SoCon season looked when the Five Factors were taken into account. I’ve already gone over a bunch of stats in my annual post on per-play numbers, conversion rates, etc., but this is something I haven’t tried to calculate before.

It wasn’t easy, either. FCS statistics for the categories mentioned above basically don’t exist online (at least, I certainly didn’t find any of consequence). The fact the Southern Conference does not have league-only online stats didn’t help.

However, I put together a small package for The Citadel’s season. It is far from perfect, and may not mean much to some people (perhaps for good reason).

There are still almost two months before the opening kickoff, though. So at the very least, it’s better than not talking about football at all.

I’m going to go over the Five Factors now. Afterwards, there are three other statistical categories of note I wanted to briefly discuss. One of them in particular struck me as worth mentioning.

First things first: a spreadsheet! The spreadsheet includes individual game statistics for all of these categories.

Again, a reminder — these stats are for SoCon games only. Also, overtime statistics are not included.

Also, I’m going to use FBS numbers for comparison purposes throughout this post, mainly because there are no FCS equivalent stats online. I’m guessing that if FCS stats were available, they would be similar to those from FBS. At least, I hope so…

Field position

I think field position is possibly the easiest of the Five Factors to understand. The one thing to think about with field position is this: you measure an offense’s effectiveness (in terms of field position) by the starting field position of its defense (and vice versa).

Also, special teams play is obviously important. Net punting, kickoff coverage, the return game — all of that matters.

The FBS national average for starting field position in 2016 was the 29.7 yard line.

-Average starting yard line of offensive drives-

The Citadel Opponent Margin
(Home) 32 26.5 5.5
(Road) 32.25 29 3.25
Total 32.125 27.75 4.375

The Citadel won the field position battle in five of eight games. One of the things that helped the Bulldogs the most in this aspect of the game was the “three-and-out” differential.

Simply put, The Citadel’s defense did a good job of forcing the other team off the field in three plays or less (the “less” occurring when the Bulldogs’ D created a turnover). The offense tended to have longer drives than its opponents, and that usually tilted the playing field in The Citadel’s favor.

The Citadel’s offense had a “3-and-out+” rate of 27.5%, while Bulldogs opponents had a rate of 35.2%. That 7.7% differential was substantial. It would have ranked in the top 35 in FBS, for example.

The top 3 defenses in FBS in 3-and-out+ differential in 2016 were Temple, Clemson, and Ohio State; each had a differential of more than 17%. Those three teams had a combined record of 35-7, with two league titles and two CFP bids (including the playoff winner).

The Bulldogs also benefited from good special teams, particularly kickoffs.

The net punting does not show up quite as well; I find that frankly puzzling, because The Citadel had a generally solid performance from its punt team all season (with the exception of a blocked punt against Wofford).

My guess is that because there wasn’t as much field to work with a lot of the time (as the Bulldogs often had a territorial advantage in individual contests), that there were only so many net punting yards to be had.

Also of note, The Citadel had significantly better net punt/kickoff numbers at home.

Efficiency

With efficiency, we’re talking about a statistic called “Success Rate”. Here is its definition, via Football Outsiders:

A common Football Outsiders tool used to measure efficiency by determining whether every play of a given game was successful or not. The terms of success in college football: 50 percent of necessary yardage on first down, 70 percent on second down, and 100 percent on third and fourth down.

The FBS national average for Success Rate in 2016 was 40.9%.

-Success Rate-

The Citadel Opponent Margin
(Home) 48.20% 43.60% 4.60%
(Road) 42.10% 36.20% 5.90%
Total 45.40% 39.90% 5.50%

As was the case with field position, The Citadel won the efficiency “contest” five out of eight times in 2016 league play.

Incidentally, for the Western Carolina and ETSU games, only first-half statistics were calculated for Efficiency and the next category (Explosiveness). That is because both games were essentially over at halftime.

Bill Connelly, in his book Study Hall, expounds on this line of reasoning:

…The goal of the game for one team has changed from winning to making the game end as quickly as possible…the game is, in effect, over, and what happens after ‘garbage time’ begins is no longer truly evaluative of the teams at hand.

Defining when a game is no longer competitive can be tricky. After all, we’ve all seen big comebacks (The Citadel’s 2011 victory over Chattanooga comes to mind). Still, I think it is fair to consider the WCU and ETSU contests as no longer being in doubt after the first half.

The most efficient opposing offense against The Citadel in 2016 was Samford, which ran a successful play 50% of the time against the Bulldogs’ D. Then there was the game at Wofford, where The Citadel’s offense was only successful on 25% of its plays, by far the lowest percentage for the team all season in league action.

Explosiveness

How is this category defined? Well, with something called “IsoPPP”, and believe me, I had no idea what that was myself until I started researching this topic.

IsoPPP is the Equivalent Points Per Play (PPP) average on only successful plays. This allows us to look at offense in two steps: How consistently successful were you, and when you were successful, how potent were you?

This definition doesn’t really benefit The Citadel, because a lot of the Bulldogs’ successful plays last season were of the “move the chains” variety — five yards on first down, two yards on 3rd-and-1, etc. Every now and then, someone would bust a big play, but that was counterbalanced by all the “smaller” good plays The Citadel had.

This is reflected in the numbers, as the Bulldogs only came out ahead in this category in one of eight league games. Even ETSU had slightly higher “explosiveness” despite being out of the game at halftime.

That doesn’t mean this statistic doesn’t matter as far as The Citadel is concerned. Of course it does.

The Bulldogs need more big plays on offense. They can win without them (as they did in the Chattanooga game, when The Citadel’s longest play from scrimmage was Dominique Allen’s 15-yard gain on the offense’s first play of the game), but it’s much easier to move down the field in large chunks.

The FBS national average for Explosiveness was 1.27.

-Explosiveness (IsoPPP)-

The Citadel Opponent Margin
(Home) 0.924 1.109 -0.185
(Road) 1.068 1.132 -0.064
Total 0.985 1.119 -0.134

Finishing Drives

This category calculates points per trip inside the opponent’s 40-yard line. It’s more or less an elongated version of the “Red Zone” concept.

The FBS national average for points per trip inside the opponent’s 40-yard line in 2016 was 4.47.

-Finishing Drives-

The Citadel Opponent Margin
(Home) 4.5 4.1 0.4
(Road) 4.5 4.7 -0.2
Total 4.525 4.407 0.118

The Citadel had the edge in this category in six of its eight SoCon games. Of course, the Bulldogs also had many more opportunities to add to their “finishing drives” totals than their opponents; The Citadel had 40 such drives in league play, while their opposition had 27.

Turnovers

First, a table of the actual turnovers:

The Citadel Opponent Margin
(Home) 3 5 2
(Road) 4 9 5
Total 7 14 7

A net margin of 1.0 turnover per league contest is a good way to win a lot of games. In FBS, Washington and Western Michigan tied for the national best in turnover margin per game, at 1.29. Only six FBS squads had a net of 1.0 turnover per game or higher.

This next table is the “adjusted” or “expected” turnovers:

The Citadel Opponent Margin
(Home) 5.82 6.9 1.08
(Road) 3.58 6.7 3.12
Total 9.4 13.6 4.2

The difference is that The Citadel had a bit more “turnover luck” than its opponents. The expected turnovers stat is based on A) the idea that recovering fumbles is a 50-50 proposition, and B) that a little over 1/5 of passes that are “defensed” are intercepted. In other words, if a defensive back breaks up four passes, the fifth one he get his hands on probably should be a pick.

In case anyone is interested, I calculated the “passes defensed” interception rate at 22%.

Just because the Bulldogs may have had a bit of good fortune in the turnover department last season, that doesn’t mean a regression is imminent. They start on the same playing field as everyone else this year.

Now, let’s see how The Citadel did in the Five Factors on a game-by-game basis in league play:

  • at Mercer: The Citadel won 3 of the 5
  • Furman: The Citadel won 4 of the 5
  • at Western Carolina: The Citadel won all 5
  • Chattanooga: The Citadel won 2, UTC won 2, and there were no turnovers
  • at Wofford: The Citadel won 2 out of 5
  • ETSU: The Citadel won 3 out of 5, but 4 of 5 in the decisive first half
  • Samford: The Citadel won 0, Samford won 4, and there were no turnovers
  • at VMI: The Citadel won 2, VMI won 2, and each team had one turnover

As for the “what happened?” results, a few explanations:

– Chattanooga: a close game, obviously, that The Citadel won at home

– Wofford: went to OT; field position doesn’t account for the “Pitch 6”

– at VMI: the Keydets’ turnover resulted in a defensive TD; also, The Citadel crushed the “Efficiency” category

– Samford: went to OT, and, uh…

Two of the four categories that favored Samford were very close (Efficiency and Explosiveness). I think one takeaway from that game might be that when one team runs a lot more plays from scrimmage (86-64), it could have a “hidden” edge in efficiency no matter the numbers.

As it was, Samford was up 10 points with six minutes to play in the game. Then the tide suddenly turned on a Cam Jackson run on third-and-long. One TD later, one three-and-out later, one quick field goal drive later, and the game was headed to OT.

We move on from the “Five Factors” (well, at least I’m moving on) and wrap this up with three other statistical categories that I think could be of some interest.

-First down yardage gained per play-

The Citadel Opponent Margin
(Home) 6.69 6.06 0.63
(Road) 5.68 5.11 0.57
Total 6.21 5.59 0.62

To be honest, I was inspired to look these numbers up while perusing Athlon’s 2016 college football annual, which included statistical tidbits for all 128 FBS teams. Some highlights:

  • Western Kentucky’s offense led FBS in average yards gained on first down, with 8.9. That was well ahead of second-place South Florida (8.1).
  • The worst FBS squad in this category was Fresno State (just 4.4 yards gained on first down on average).
  • Minnesota’s defense topped FBS in allowing first down yardage, with its opponents averaging 4.3 yards.
  • I don’t know which team was worst in FBS, but FIU’s defense was 126th out of 128, allowing 7.5 yards per opponent first down. Butch Davis needs to bring in some players.

I also went back and took a look at The Citadel’s 2015 conference numbers in this area, for comparison.

  • The Citadel’s defense allowed an average of 6.09 yards on first down in 2015, including 5.47 yards per rush on first down and 6.80 yards per pass attempt on first down
  • The Citadel’s defense allowed an average of 5.59 yards on first down in 2016, including 3.29 yards per rush on first down and 7.98 yards per pass attempt on first down

The Bulldogs’ D just shut down the running game on first down in 2016. It allowed a bit more per pass attempt, but not enough to prevent an improvement from the year before of a full half-yard.

Okay, we’re leading up to something that is not on the spreadsheet, but which is important.

  • The Citadel’s offense averaged 6.49 yards on first down in 2015, including 5.94 yards per rush on first down and 11.04 yards per pass attempt on first down
  • The Citadel’s offense averaged 6.21 yards on first down in 2016, including 6.14 yards per rush on first down and 6.65 yards per pass attempt on first down

This is something that needs to change in 2017. The Citadel doesn’t throw often, but when it does, it has to make it count. That is especially true on a “standard down”, i.e. a down in which the opponent would not normally expect the Bulldogs to pass. First-and-10 is definitely one of those downs.

Averaging 11 yards per attempt is outstanding, but it is also something that you would almost expect to see in a well-oiled triple option offense. In a typical game, the Bulldogs may throw the ball on first down 2 or 3 times. With the element of surprise, at least one of those passes needs to go for long yardage.

-3rd down distance to gain (in yards)-

The Citadel Opponent Margin
(Home) 5.54 6.32 0.78
(Road) 5.85 10.33 4.48
Total 5.68 8.17 2.49

Air Force’s offense led FBS in yards to go on 3rd down, needing on average 5.5 yards to move the chains. I don’t have the complete list (or even a partial list), but I would suspect that 5.68 would put a team somewhere in the top 15 range, maybe the top 10.

Massachusetts had the worst offensive numbers in this category, needing on average 8.4 yards to make a first down.

Definition of passing downs: 2nd down and 8 yards or more to go for a first down, 3rd/4th down and 5 yards or more to go for a first down

-Passing down success rate: offense-

Rushes Pass Attempts Success rate
(Home) 64 17 40.74%
(Road) 57 22 36.71%
Total 121 39 38.75%

I think it is safe to say that not many teams in D-1 ran the ball 76% of the time on “passing downs”. That success rate may not look good, but it combines fairly well with the Bulldogs’ effort on defense.

-Passing down success rate: defense-

Rushes Pass Attempts Success rate
(Home) 17 60 35.06%
(Road) 42 45 25.29%
Total 59 105 29.88%

The Citadel’s defense was very good at stopping a receiver from picking up the first down after the catch, assuming he was still short of the sticks when he received the ball. This explains why opponents only had a success rate of 45% even on completed passes.

I’m not going to pretend to be an expert on any of what I just posted. This is the first time I’ve tried to perform some of these calculations; it’s possible I may not be 100% correct on everything.

However, if I thought it was all a bunch of garbage, I wouldn’t have posted it. I do have some standards, mediocre as they may be.

Any comments, suggestions, or corrections are appreciated. Also, if someone could hit the fast-forward button to football season, that would be nice.