For all the optimism coming out of Thursdsay nights’ 51-17 rout of UNLV, there was one unmistakble blemish in the box score for the Badgers. Specifically, the Badgers allowed a rather shocking 146 rushing yards against the Runnin’ Rebels, a team which only broke that number on two occasions in 2010 (and required over 40 carries to do so, as opposed to 38 against Wisconsin). After removing the asinine NCAA rule that sacks count as rushing yardage as opposed to getting their own separate category, the Badgers gave up 165 rushing yards on 35 attempts, a 4.7 yard rushing average.
Even when we take into account that sacks probably had a huge impact on UNLV’s horrid 2.96 yards-per-rush total last season — that number jumps to 3.58 when quarterback “rushes” are removed from the equation — the Badgers still allowed an inordinately high amount of rushing yards Thursday night. Specifically, it was sophomore back Bradley Randle who tore through the Badgers defense, picking up 65 yards on only 12 carries (5.4 yards per carry). Randle, not coincidentally, was decent in his limited action with the Rebels last season, picking up 4.4 yards per carry. The Badgers also had some issues stopping Randle’s fellow sophomore Tim Cornett, who averaged 4.4 yards in his 14 carries.
The Rebels ranked 113th of 120 teams in terms of rushing average and 112th in terms of success rate. When viewed through this lens, of course the Badgers’ performance was poor. But this kind of simple analysis ignores the concept of regression towards the mean — that is, with no other information, we should expect teams to move towards the average as time goes on (as the sample gets bigger). This concept applies for a number of reasons — bad teams tend to expel the poor players who caused the incompetence (like UNLV graduated RB Channing Trotter and his 3.3 YPC) and they tend to feature younger players who gained experience over the past season (like Randle and Cornett), among other reasons.
Even with an adjusted perception of UNLV’s running game, it’s still difficult to proclaim the defense’s performance as “good.” Observe, the UNLV’s success rate and yards per attempt in the running game from Thursday night (with sacks omitted as, you know, logic would suggest).
(Click to embiggen)
The best way to describe this performance would be “bend but don’t break.” The Badgers bent at a few points — specifically, we can see three runs of three successes or more in a row on the above chart — but they never gave up a huge play (no rushes greater than 20 yards) nor did UNLV have consistent success, spending much of the game with a success rate between 40 and 50%. If a run defense is going to give up a relatively high yards-per-carry, this is the way to do it — the few bigger rushes a team gives up don’t have immediate impact, and the opposing team isn’t able to sustain enough success to make it hurt. We saw this in the first half of Thursday’s game — the Rebels managed to drive into Badger territory, but couldn’t keep their drives going long enough to get more than field goal attempts.
We must consider, as well, the fact that this was Chris Borland’s first game in a calendar year. We must consider the fact that UNLV’s pistol look was completely alien to the Badgers’ defense.
Certainly, Badger fans will have more confidence if the defense can shut down the Oregon State running attack — a unit that racked up 266 yards on 43 attempts (6.2 per carry) despite losing to Big Sky member Sacramento State — this coming Saturday. For now, let’s just remember that although the Badgers should have and could have performed better against UNLV’s running attack last week, it is not a trend and by no means should the run defense be considered a major weakness at this point.
Photo courtesy Garrett Craig