Hate the Average, not the game
I've always had a little bit of an issue with the way defensive units are usually evaluated and ranked in terms of passing defense and rushing defense. While identifying the best team in each category is certainly a noble cause, and useful when breaking down matchups, the use of yards per game in either of those categories is quite a poor barometer of a teams success using or against the pass or the rush.
Remember the mid to late 1990s WAC? I'm talking even pre-ESPN2 10:00 p.m. Saturday night games featuring Pam Ward. It wasn't unusual to see 100 combined points and 5 hours of play in the intra-conference games. Quarterbacks like Josh Wallwork (Wyoming), Jon Denton (UNLV), Billy Blanton (SDSU) and Steve Sarkisian (BYU) were at the top of the NCAA Total Offense list, but no one had heard of any of them, and most still haven't. The defenses were flat out horrible, yet it wasn't impossible to give up 42 points in a game and only 130 rushing yards, which would have landed a team near the top 25 rushing defenses in the country.
The problem? The variable of rushing or passes attempts per game completely invalidates the use of the yds. per game statistic in terms of evaluating team units against the run and the pass. I'm not sure why the media, schools or whomever is in charge of affecting statistic usage trends, haven't began using the statistic of using yards per carry or per pass. The only place this average is really used widely is in evaluating individual offensive skill players. It only makes sense to extend it to team defense, given the fact that the colege football world can be regionalized when it comes to passing/rushing frequencies.
While one could make the same argument for team offenses, I think the ypg and ypc/ypa stats both have a place when it comes to that side of the ball.
An example of the defensive evaluation problem:
Last year, Boise St. had the #10 rushing defense in the nation according to the ypg method. Although the WAC isn't as singleminded about passing as it once was, Boise St. still saw fewer rushing attempts than all but one team in the nation (only Cal saw less). As we saw in the Liberty Bowl, once the Broncos saw a strong (and frequent) running team, they weren't able to put up the great per-game numbers.
If we considered YPC numbers when calculating the Rushing Defense numbers, Boise St. moves from #10 in the nation to #33.
So whenever I refer to defensive rushing/passing unit rankings here, I'll be using the ypc/ypa method.
Just for fun, here are the teams that were most inaccurately classified last year, for better or for worse, thanks to the use of an innacurate measuring stick (2004 ypg rank indicated):
2004 Rushing Defense
Overrated
Boise St. (#10)
South Carolina (#43)
Underrated
Miami (#67)
Clemson (#41)
2004 Passing Defense
Overrated
San Jose St. (#18)
Mississippi St. (#11)
Underrated
Iowa (#39)
Ohio St. (#41)
Remember the mid to late 1990s WAC? I'm talking even pre-ESPN2 10:00 p.m. Saturday night games featuring Pam Ward. It wasn't unusual to see 100 combined points and 5 hours of play in the intra-conference games. Quarterbacks like Josh Wallwork (Wyoming), Jon Denton (UNLV), Billy Blanton (SDSU) and Steve Sarkisian (BYU) were at the top of the NCAA Total Offense list, but no one had heard of any of them, and most still haven't. The defenses were flat out horrible, yet it wasn't impossible to give up 42 points in a game and only 130 rushing yards, which would have landed a team near the top 25 rushing defenses in the country.
The problem? The variable of rushing or passes attempts per game completely invalidates the use of the yds. per game statistic in terms of evaluating team units against the run and the pass. I'm not sure why the media, schools or whomever is in charge of affecting statistic usage trends, haven't began using the statistic of using yards per carry or per pass. The only place this average is really used widely is in evaluating individual offensive skill players. It only makes sense to extend it to team defense, given the fact that the colege football world can be regionalized when it comes to passing/rushing frequencies.
While one could make the same argument for team offenses, I think the ypg and ypc/ypa stats both have a place when it comes to that side of the ball.
An example of the defensive evaluation problem:
Last year, Boise St. had the #10 rushing defense in the nation according to the ypg method. Although the WAC isn't as singleminded about passing as it once was, Boise St. still saw fewer rushing attempts than all but one team in the nation (only Cal saw less). As we saw in the Liberty Bowl, once the Broncos saw a strong (and frequent) running team, they weren't able to put up the great per-game numbers.
If we considered YPC numbers when calculating the Rushing Defense numbers, Boise St. moves from #10 in the nation to #33.
So whenever I refer to defensive rushing/passing unit rankings here, I'll be using the ypc/ypa method.
Just for fun, here are the teams that were most inaccurately classified last year, for better or for worse, thanks to the use of an innacurate measuring stick (2004 ypg rank indicated):
2004 Rushing Defense
Overrated
Boise St. (#10)
South Carolina (#43)
Underrated
Miami (#67)
Clemson (#41)
2004 Passing Defense
Overrated
San Jose St. (#18)
Mississippi St. (#11)
Underrated
Iowa (#39)
Ohio St. (#41)
posted by JohnnyMitchell at 4:37 PM
0 Comments:
Post a Comment
<< Home