2015 West Virginia Football Season Review: The Defense - Part 2 - Advanced Stats

In Part 1 of the Season Review, we looked at the standard, or aggregate, statistics as they related to the defense. The problem with looking only at those statistics is that they do not take into account the competition. Playing in the Big 12 can artificially inflate the total stats to make teams look worse than they may appear. By comparison, playing in a conference where teams do not run many plays can artificially deflate statistics making them appear better than they are.

If you are interested in the advanced stats, I suggest you checkout the Stats Glossary over at FootballStudyHall. Bill C., who runs FootballStudyHall, put together a fantastic post last year about the Five Factors necessary to win a football game. To paraphrase, if you can make more big plays than your opponent, if you can stay on schedule, if you tilt the field, you finish drives, and you fall on the ball you win more games than you lose. We're going to take a look at the defense and compare them on a per play basis.

On average, a Big 12 team runs over 79 plays per game. Not only do the teams in the Big 12 run more plays than the average team but the conference schedule means there is one less non-conference game, which also allows teams to pad their rankings by playing a weaker opponent. Without looking at some of the advanced stats, you truly cannot compare teams who play difference non conference schedules, different conference schedules and play at different paces.

Rushing Defense

So what do all these numbers mean? Let's break it down. I'm not sure the statistics tell us anything we didn't know about the 2015 defense, but they do paint that picture in high definition rather than in brush strokes. S&P+, a stat devised by Bill Connelly, which looks at all of the plays in college football and ranks teams based on four factors: Explosiveness, Efficiency, Field Position and Finishing Drive. An explosive, efficient offense than scores points on every drive is extremely hard to beat. If that same team is given a shorter field, they become that much more dangerous. A defense that does not allow a team to gain long plays and score points gives its team a chance to win. S&P+ ranks the Mountaineers rush defense as 3rd best in the nation based on who they played and how they performed.

That is surprising. I would have told you the Mountaineers had a very good rush defense but 3rd in the nation? Hard to believe.

The next statistic is one I really love and gives us a chance to see how good the defense is. Rushing Success Rate is defined as the number of plays that gain at least 50% of the yards on first down, 70% of the yards on second down, and all the necessary yards on 3rd and 4th down. The lower this number, the more often a team is put in 2nd and 3rd and longs or is forced to punt. The Mountaineers did a very good job of forcing teams into longer field positions and forcing teams to punt.

Adjusted Line Yards is very similar to OPS+ in baseball. It is a normalized stat averaged to 100. Above 100 is good, below 100 is bad. WVU is above 100 and ranked 33 in the nation. Without going into too much detail, this statistic attempts to assign how much credit goes to the running back and how much credit goes to the offensive line. The defensive measure adjusts how much the defensive line effects the running game. For a team with only a "front 6", to be so good on the defensive line is really surprising.

Opportunity Rate is a straight measure of the percentage of running plays that gained at least 5 yards. Based on historical and numerical data, an offensive line is responsible for the first five yards. Yards 5-10 are equally split between a running back and offensive line. The Tony Gibson led defense only allowed 34% of running plays to gain 5 yards or more, well below the national average of 39%.

Our next statistic is Power Success Rate. This stat is a measure of old school, my offensive line is bigger, meaner, badder than your defensive line and we can line up and gain 2 yards when you know we are running the ball. The Big Ten loves this type of football. The true definition is "the percentage of runs on third or fourth down, two yards or less to go, that achieved a first down or touchdown. Also includes runs on first-and-goal or second-and-goal from the two-yard line or closer." As you might suspect with an out-manned front 6, when teams lined up to gain 2 yards, they were more successful than the national average against the Mountaineers. This will always be a weakness of the 3-3-5.

The final rushing advanced stat we're going to compare is Stuff Rate. This measures the percentage of running plays where the defense "stuffed" the run at or behind the line of scrimmage. West Virginia is below average in this area but above average in Adjusted Line Yards. These paint the picture of a swarming defense but not one that gets a lot of penetration on running plays.

Passing Defense

Now we move on to the pass defense. With Karl Joseph, Darryl Worley, Terrel Chestnut, KJ Dillon and Dravon Askew-Henry, we should expect the stats that measure the ball in the air to be in favor of the Mountaineers.

Our first measure is Passing S&P+. This is very similar to the S&P+ measure in rush defense, only measured and accounted for on the passing plays. To be honest, given that this stat is adjusted for opponent, I would have expected an even higher ranking. WVU accounted for 31 turnovers, 26 of which were interceptions. They held teams to an average of less than 50% passing. To be that good in passing metrics while playing teams that often throw the ball around 55% of the time, would have led me to believe we had a top 5 pass defense.

Moving onto Success Rate, we once again measure how often the pass defense allowed a team to gain a specific percentage of yards on downs. The defense was very good with this metric, only allowing a team to be successful 33% of the time on passing plays. That is absolutely outstanding.

The final two statistics do not paint the Mountaineers in a favorable light. The first stat is IsoPPP. This stat attempts to measure how efficiently an offense gains yards along with how explosive they are. When coaches address the media, they often says things like "We need to limit the big play". What they are saying is that you can't let the offense gain 25+ yards. If you can make the team gain 3-4 yards at a time they are more likely to shoot themselves in the foot. IsoPPP takes these factors into account and weights them. This allows us to look at offense in two steps: How consistently successful were you, and when you were successful, how potent were you? With the amazing Success Rate number for the Mountaineers, the defense must have had several breakdowns that resulted in long scoring gains. If you watched the Baylor, Oklahoma or Kansas State game, you would know that is the case. The defensive backs would gamble and allow a busted play to go for a long score.

The final measure should not be a surprise to Mountaineer fans. Adjusted Sack Rate, which gives sacks (plus intentional grounding penalties) per pass attempt adjusted for down, distance, and opponent. Its then normalized to 100. Above 100 is good and below 100 is bad. It should be no surprise that a team with 30 sacks would be less than average in Adjusted Sack Rate.

Total Defense

We've now broken down the two main components of defense, rushing and passing. We need to see how the defense fared as a whole. From the above stats we can tell that the 3-3-5 did a very good job of putting teams in longer down and distance situations (Passing and Rushing Success Rates in the 30 percents, well below the national average) and did an even better job of swarming to the ball (Passing S&P+, Rushing S&P+, Adjusted Line Yards) but if containment failed, it often failed spectacularly for scores (Rushing IsoPPP & Passing IsoPPP).

We can also tell that despite smaller, quicker linebackers and defensive line players, the Mountaineers did not get consistent penetration on running or passing plays (Stuff Rate and Adjusted Sack Rate). This gives us the picture of what the defense did and what it is capable of. How does that compare to the rest of the country and within the conference?

Overall, the Mountaineers were well above average all of the advanced metrics except for Explosive Rate. We've touched on this above, when the defense broke, it broke spectacularly.

The first value is the most convoluted. The DFEI is based off of  The Fremeau Efficiency Index (FEI). It is a college football rating system based on opponent-adjusted drive efficiency. Defensive efficiency (DE) is the scoring value generated by a team's defense per possession. Opponent-adjusted defensive efficiency (DFEI) is scoring value per possession adjusted for the strength of opponent offenses faced.

This metric views the Mountaineers quite favorably. Based on the teams WVU played, how WVU played and how little they allowed teams to score per possession, the Mountaineers defense comes in at fourth overall. I don't know that you can ask for any more than that from a senior laden team. Well done WVU.

Moving onto Defensive Efficiency, this stat attempts to measure how much scoring value the defense added to the entire team. In many ways that is related to not allowing a team a first down when Nick O'Toole pins the opponent at the 5 yard line or intercepting a pass in the opponent's redzone to give the offense a chance at an easy scoring drive. The defense did very well in this metric, ranking 22nd overall and 2nd in the conference.

Now we can move onto less "subjective" statistics and more measurable stats. The first one is Defensive First Down Rate. Its a very simple stat, simply telling us the percentage of opponent drives that resulted in at least one first down.  67% of the time, WVU gave up at least one first down. That sounds like a lot but its still much lower than the national average of 71% of the time. Still what it tells us that for every 3 possessions, West Virginia generated a three-and-out.

Next we move to Available Yards Percentage, or the total number of yards earned on opponent drives as a percentage of the total number of yards available based on starting field position. How many yards does WVU give up based on the total available. If Nick O'Toole pins the opponent at their own goal line, we can reasonable expect the defense to give up 38 yards (38% of the available yards). This is a great stat for WVU. Often bend-but-don't-break defenses will give up more yards and then stiffen up in the redzone. WVU takes the opposite approach and just doesn't give up the yards.

The lone outlier in our rosy picture of the defense is Defensive Explosive Drive Rate. This measures the percentage of opponent drives that earn at least 10 yards per play. The last two words there are very important, 10 yards PER PLAY.  WVU is still above the national average and given that the team plays in the most explosive conference in the country, having a lower number isn't unexpected. 12% of the time, the Mountaineers gave up a drive that resulted in 10 yards per play.

Defensive Methodical Drive Rate attempts to measure the percentage of drives that last at least 10 plays. Less than 10% of the time, the Mountaineers were on the field for sustained drives. That matches well with the Explosive Drive Rate. The higher the explosive drive rate, the lower the methodical drive rate. However, since WVU was not near the bottom of the DEx measure, we can tell that the defense did a good job of simply limiting the opposition and not allowing them to gain sustained drives that wore down the defense.

The final statistic is Defensive Value Drive Rate, the percentage of opponent drives that begin at least 50 yards from the end zone and reach the team's 30-yard line. Essentially, how often did the defense allow the other team to start on their end of the field and reach at least field goal position. This is a hidden gem in terms of advanced stats. Less than 30% of the time, WVU was successful in keeping teams from getting into scoring position at all*. If you rewatch the Oklahoma game, you can see this defense, despite being put in a position where Oklahoma started with the ball at their 25 yard line, the defense kept OU out of the end zone. When OU had to drive the field, save for one quarter, the defense stood tall.