Steelers Heading to the Playoffs? Football Outsiders Says Go Ahead And Bet The Farm

by Michael Bean (Blitz) on Sep 26, 2007 10:39 AM EDT in News Comment 6 comments

Wow I love Football Outsiders. The latest gem I've discovered is their DVOA Playoff Odds Report. Basically, FO has played out the season 5,000 times using an equation based on their metrics plus home-field advantage (take the time at some point to read their explanations of how their metrics are derived). Here's what they came up with:

Odds of Pittsburgh making the Playoffs = 94.1%

Odds of Steelers winning the AFC North = 79.7% - um, for those writers out there who aren't convinced that the Steelers are in the same class as the Bengals and Ravens yet? I dare you to take on FO's metrics.

Odds of Steelers securing #1 Seed in Playoffs = 36.9% - interestingly enough, only the New England Patriots are a safer bet to make the playoffs (99.1%), but Pittsburgh is still more likely to claim the top seed (36.9% vs. 34.4%).

A few more interesting nuggest from their projections: San Diego's got work to do. As of now, their likelihood at making the playoffs is a mere 21.8%. The Bengals better get after it as well. Their chances are 16.6% for the moment. Here was also a big shocker that I'd like to spend more time thinking about: Tampa Bay is slightly more likely than Indianapolis (10.5% vs. 9.3%) to win the Super Bowl

Oh, and by the way, Your Pittsburgh Steelers have, at the moment, the the best odds at winning the Super Bowl. Hey, it's just one computer's 'opinion', but I'd rather such prognostications be coming from an algorithem than a human's stream of consciousness.

0 recs | Comment 6 comments

Story-email Email Printer Print

Comments

Very interesting

Do you know if FO accounts for probability of injuries at key positions ? For example, if Brady was to blow his ACL on Sunday, what does that do to NE's chances? (just an example, i'm not inviting bad karma for the Steelers by actually hoping for this)

btw, when you say we are "odds-on" for the super bowl, I assume you mean "team with highest probability" of winning the SB. "Odds on" would imply better than a 50% chance.

by santhome on Sep 26, 2007 11:57 AM EDT reply actions 0 recs

stand corrected

Good point, wrong term there. I will correct it.

And no, they do not account for subjective variables. That would be impossible, and of course, it's the reason these metrics only carry so much water. Applying regression analysis is probably easier with the stock market than it is with sports. There are simply far too many unpredictable variables that can't be accounted for on a weekly basis.

by Michael Bean (Blitz) on Sep 26, 2007 12:04 PM EDT reply actions 0 recs

Hard but not impossible

It depends on whether they can come up with meaningful historical data to model these variables. For example, based on games missed due to injury and sacks per attempt I would say McNabb, McNair and unfortunately Ben are somewhat more susceptible to injury than Manning or Brady. Combine that with league averages for QB injuries and you could quantify it as a composite probability, build a model and test it against historical data to see if it is accurate most of the time.

by santhome on Sep 26, 2007 2:55 PM EDT up reply actions 0 recs

Hmm, how ironic.

Weren't these the same people who, after the Steelers' last Super Bowl victory, calculated them to be the most improbable Super Bowl champions in NFL history?

19 games later, it's amazing how things completely turn around, eh?

by HinesField on Sep 26, 2007 12:17 PM EDT reply actions 0 recs

well, actually

It was Pro-Football-Reference who made that calculation. Different set of metrics used. And, when you consider we had to win 5 in a row I believe to even be the #6 seed, those calculations probably carry more theoretical and mathematical weight than do these extremely early-season projections made by Football Outsiders. Your point of course is valid that this is all conjecture. I presented it more for fun than anything else. I'm not enough of a regression analysis and sabermetrics whiz to comprehensively explain the merits of their DAVE system, but I am aware that huge aspects of their system are arbitraty and that even one single unpredictable variable (like an injury to Big Ben or Willie) would throw absolutely everything out of whack.

by Michael Bean (Blitz) on Sep 26, 2007 12:31 PM EDT reply actions 0 recs

It's still good

Being 3-0 and having FO, the preeminent outside football analysts (think baseball prospectus 10 years ago) say that right now you have a 94% chance of making the playoffs is pretty awesome. I'd say the year started off much better than I could've anticipated.

by Cols14 on Sep 26, 2007 2:04 PM EDT reply actions 0 recs