Tuesday, June 19, 2007

Turnovers: Now With Context!

First, a note about the NCAA's numbers. In going back through their database, I noticed some holes in their turnover data from the 2000 and 2001 seasons. At least I think they're holes. An example:

Do the blank spaces denote zeros or a lack of data? I'm not sure, but after looking through some old box scores, I suspect the former. Even so, I'm going to omit the 2000 and 2001 seasons just to be safe. Put a mental asterisk next the Pack's 2000 season turnover figures in my previous post, though fortunately there aren't any holes in the Pack's 2001 data.

Moving on...

A commenter on yesterday's post wondered if maybe teams were playing more conservatively against the post-Rivers Wolfpack because they knew they didn't need to score as much, which might could explain why we've had trouble forcing fumbles. What follows is a breakdown of our opponents' play selection by year and their respective fumble rates.
        Pass   Rush   Rush%   Fum/100Ru
2002 437 525 54.6 5.9
2003 516 470 47.7 5.7
2004 272 429 61.2 3.3
2005 380 461 54.8 3.7
2006 269 481 64.1 2.7
Opponents have been running the ball more frequently over the last three seasons, which would suggest a more conservative style, but all I can do is guess at the impact that might have on their fumble rates (if there's an impact). It'd be helpful if the NCAA listed fumbles for each player, because in that case I could look at fumble rates both per rush and per pass, which might lead us closer to figuring out what play selection means for a team's overall fumble rate.

In order to add some context to the Wolfpack's turnover rates, I compiled those numbers for the rest of the conference. In the table below are each school's average rates from 2002-2006. I've included all 12 current members despite the fact that some obviously weren't in the league for all five years. Sorted by opponents' fumbles per 100 rushes:
                Opp_Fum/100Ru  Opp_INT/100PA  Fum/100Ru  INT/100PA
Va Tech 6.0 5.0 4.5 3.8
Wake Forest 5.6 3.5 3.4 2.9
Florida State 5.4 3.4 5.0 3.3
Duke 5.1 3.4 5.0 3.9
Miami 5.0 4.0 4.8 3.5
Ga Tech 4.9 3.2 4.3 4.1
Boston College 4.9 4.0 4.0 3.1
Average 4.7 3.5 4.6 3.4
Clemson 4.4 3.9 5.4 3.2
NC State 4.3 3.0 5.1 3.2
Virginia 4.2 3.3 3.7 2.6
Maryland 3.8 3.0 4.9 3.9
North Carolina 3.4 2.0 5.0 3.5
-- Virginia Tech's defense has been a fumble-causin', pass-yoinkin' machine over the last five years. Both of their rates top the league.

-- On the flip side are John Bunting's loveable collection of guys what suck at football. Cause fumbles? Catch the ball so as to prevent it from reaching its intended target? Why bother? Only two of the twelve teams have caused fewer than 100 turnovers since 2002: Maryland (97) and Carolina (77).

-- Wake's line offers more tangible evidence of Jim Grobe's coaching prowess. The Deacs are a league-best +43 in the turnover department over the last five seasons. Grobe's been able to put together a defense that, despite its limitations, is solid at forcing interceptions and quite good at causing fumbles. They may lack size and speed, but they still create a lot of problems. The offense, meanwhile, has been incredibly secure.

-- Clemson is the fumblingest school in the ACC. Back in 2002, they fumbled 39 times in 13 games (8.5 per 100 carries).

-- Rare is the team that throws fewer than 2 INTs per 100 pass attempts over the course of a full season; that's been done a mere five times since 2002 (twice by UVA, once by NC State, Miami, and Maryland). The lowest figure belongs to the 2003 Wolfpack, which threw 1.4 INTs per 100 passes.

-- There is one team that's above average in all four categories: Tom O'Brien's Boston College Eagles. Clearly, this man is a terrible football coach. I hope you ruin us that good, TOB.