Being Toxic Is a Good Thing

As we go into the season, I’ll be posting roughly once a week about some of my favorite statistical metrics, which I will then update periodically with new data as the season progresses.  First up is a little number known as toxic differential.  For those of you who are unfamiliar with it, toxic differential is a combination of two other stats, turnover differential and explosive play differential (TOXic = TurnOver + eXplosive plays).  This stat is a great barometer for gauging a team’s chances of making the playoffs — the higher your number, the better the odds are you’ll still be playing come January.

Now, I’m not big into doing lots of math — hell, back in my college days I switched majors from engineering to English after a 3D calc class made me realize just how miserable working equations made me — but I love me some stats.  There’s nothing better for giving your assumptions a reality check than a dose of cold, hard numbers.  But there’s a caveat here, and it’s this: you have to make sure you’re looking at the right numbers.  Feed in the wrong data or crunch your figures the wrong way, and your stats will cheerfully lead you over a cliff into some painfully wrong conclusions.

For toxic differential, working out a team’s turnover differential is simple enough (there’s only so many ways to tally interceptions and fumble recoveries), but explosive play differential can be an issue.  The problem is that the commonly understood definition of explosive plays — any offensive play that results in a gain of twenty or more yards — is, technically speaking, utterly batshit.

For one, it means you’re judging run plays and pass plays by the same criteria, and any longtime football fan will tell you that doesn’t really work.  Think about it, gaining five yards per attempt will get most running backs a pro bowl berth, but for a quarterback a 5 yds/att average just gets you ridiculed and benched in short order.  Similarly, gaining 100 yards in a game is a good day’s work for an RB, but means things went horribly wrong for a QB.

I originally started working with toxic differentials after reading Brian Billick’s book Developing an Offensive Game Plan.  Today he’s better known for his stint as head coach of the defensive-powerhouse Ravens, but few people seem to remember he landed that job as the hotshot offensive coordinator of one of the greatest offenses of the last twenty years, the 1998 Vikings.  In his book, Billick defines explosive plays as runs of twelve or more yards and passes of sixteen or more yards, and after a few years’ worth of crunching stats for fun I have to say I think his numbers paint a much more complete picture of the overall ability of a given team than the NFL’s one-size-fits-all definition.  Don’t get me wrong, I think tallying total chunk yardage plays (i.e. 20+ yards) has its uses, but those stats operate at only the extreme end of the explosive play range.

Anyway, enough background — here’s how every team ranks in toxic differential through week five:

And here’s the raw data for the turnover and explosive play differentials with the teams arranged alphabetically for ease of reference:

The 2013 Seahawks led the league in toxic differential with an incredible +59 for the season, which is inhumanly good; it’s the sort of stat line you’d expect to see if you were playing Madden’s franchise mode on the lowest difficulty setting.  Obviously, the 2014 Hawks are not on pace to match that number again, but the sky is not falling here.  The 2013 Hawks had the exact same +7 explosive play differential after four games (20 passes & 12 runs on offense, 16 passes and 9 runs allowed on defense), and had this year’s offense had a few more drives against San Diego this number could be even higher; despite limited snaps, Wilson threw more explosive passes versus their defense (5) than against any other team he’s faced so far.

The difference here has been turnover differential, more specifically the team’s lack of takeaways.  Seattle has actually been better at protecting the ball this year, with just 3 turnovers as compared to 6 after four games in ’13, but they’ve racked up just 3 takeaways — last year’s squad tallied 11 through four games.  One could argue that low takeaway number is the result of a few calls and a few bounces not going the Hawks’ way, and if that’s the case the law of averages should smooth things out in that column as the season wears on.

No one is denying that the Seahawks have been playing some sloppy ball this year, but I prefer to look at the situation this way: even when they aren’t playing up to par, the Hawks are still a top ten team in toxic differential.  They haven’t been just padding their stats versus pushovers, either — as you can see, three of the four teams they’ve faced so far are also currently rank in the top ten.

To end with, here’s a couple of quick observations on some other teams’ stats:

• DeSean Jackson may be Washington’s shiny new deep threat, but he isn’t the sole reason their offense is fourth in the league in explosive passes.  He isn’t even the leader for the team in that category — TE Niles Paul is, with 9 catches of 16+ yards to Jackson’s 7.
• A big part of Jacksonville’s woes has been their lack of production on the ground.  Of the four explosive runs they’ve managed, two were by QBs, one apiece for Henne and Bortles.  Their new feature back Gerhart has just one run of 12+ yards to his name this year.
• If Jason Garrett keeps his job past this season, he should start signing over half of every paycheck he receives to DeMarco Murray. Of the Cowboy’s league-leading 18 explosive runs, Murray is responsible for 15 of them.