Writers of Pro Football Prospectus 2008

15 Sep 2012

First-and-10 Inside the 10: So You Wanna be a Hall of Famer?

Using Pro Football Reference's game play finder, Jason Lisk uncovers a glaring, yet minimally exploited, strategic advantage for any coach who dares to throw when backed up inside his own 10-yard line. If only they called plays optimally, we all might be Hall of Famers.

Posted by: Danny Tuccitto on 15 Sep 2012

11 comments, Last at 19 Sep 2012, 8:04pm by zenbitz


by Bright Blue Shorts :: Sat, 09/15/2012 - 5:17am

Another intellectual notch in the belt of Bill Belichick?

The Giants got a safety when Tom Brady dropped back to pass on their first play in the Super Bowl last year ... and while it was an intentional grounding penalty I think Brady held the ball too long rather than being under massive pressure.

by BJR :: Sat, 09/15/2012 - 6:20am

Good article. But I'd ask, how much extra advantage does throwing instead of running on 1st&10 inside the 10 yield in comparison to throwing instead of running on any down and distance, anywhere on the field?

by Mr Shush :: Sat, 09/15/2012 - 6:48am

It doesn't seem to me at all obvious that this is a case where teams should seek equilibrium of outcomes. The optimal strategy might perfectly well be to increase the pass:run ratio to just short of whatever point would stop opposing coaches from selling out against the run. Of course, identifying that point could be tricky.

by Anonymousjk (not verified) :: Sat, 09/15/2012 - 8:53am

Two things: a) does this account for safeties caused by penalty (seems to only discuss sacks) and b) does it account for the fact that good QBs will have a disproportionate number of plays in this situation?

If you have Brandon Weeden, you're not even going to try from the 2. Not as much Aaron Rodgers.

by andrew :: Sat, 09/15/2012 - 10:42am

It will be nice to see Gus Frerotte's bust in canton, alongside the wall.

by dbostedo :: Sat, 09/15/2012 - 10:51am

And facing the wall I assume, perhaps with a slight tip forward so the forehead is making contact...

by JonFrum :: Sat, 09/15/2012 - 1:02pm

Unless you're Jake Locker - Chandler Jones strip-sacks you, Dont'a Hightower picks it up and returns it for a TD.

by BMFS :: Sat, 09/15/2012 - 7:10pm

Of all things -- the embedded video cost me a fantasy matchup! My opponent had both Frerotte and Berrian, and that play was worth approximately 35 points with the long-TD bonus.


by erniecohen :: Mon, 09/17/2012 - 8:30am

I confess I'm a little disappointed that the stats don't support the notion that it's better to throw on first and 11 at the 1 than on first and 10 at the 2, just so that there would be a statistical argument for intentionally taking a penalty.

Taking penalties very close to your own goal line is perhaps my favorite tactic in football, one that probably has many potential applications (other than taking the opportunity to get a free shot at opposing linemen when you have nothing to lose and they can't retaliate). I always wondered why a defensive team, trying to kill time with the ball practically on their own goal line, doesn't just line up all 45 guys on the field on defense.

by NoraDaddy :: Mon, 09/17/2012 - 11:24am

I think this would fall under Rule 12, Section3, either Article 2 or 3:

Article 2 The defense shall not commit successive or continued fouls to prevent a score.
Penalty: For continuous fouls to prevent a score: If the violation is repeated after a warning, the score involved is awarded to the offensive team.

Article 3 A player or substitute shall not interfere with play by any act which is palpably unfair.
Penalty: For a palpably unfair act: Offender may be disqualified. The Referee, after consulting his crew, enforces any such distance penalty as they consider equitable and irrespective of any other specified code penalty. The Referee could award a score. See 15-1-6.

by zenbitz :: Wed, 09/19/2012 - 8:04pm

It's not clear to me that this isn't already at Game Theory equilibrium.

YPA is so high because defenses play the run because offenses run >60% of the time. If they passed more, defense would play the pass more and YPPA would go down and YPRA would go up.

If you multiply it out -
7-9 yl: .35x7.85 (2.74) ~= .65x4.2 (2.73)
4-6 yl: .30x8.99 (2.67) > .70x3.2 (2.24)
<3 yl: .27x10.2 (2.75) > .73x3.08 (2.24)

If you call a turnover -30 yards, then 4-6 evens out quite a bit (+2% TO)... and it looks like teams should run MORE on 1st down from the 7-9 YL. That's probably not right as we "know" (from between the 20s data) that if you go to 50-50 pass/run you can maintain 7.4 yppa while only going up to 4.5 r/p. (more like 5.5 if you count turnover differential).

Also you would conclude (work not shown - I think this has been pointed out before) that teams run too much on 1st down between the 20s.

I wonder if you just derived a function of YPP (pass & run) as a function of frequency you could estimate what to change your ratio to?

If you guesstimate at a 65:35 P/R ratio you can get 7 YPPA and 6 YPRA (5+0.9 for net turnovers@30 yards per) with turnover ratio unchanged... you get about 7 YPP total. Essentially the 2011 Saints offense.

At some point (60/40?) the defense just goes into permanent nickle which should (in principal) reduce the Y/A for an average team passing to more like 6.

Incidentally, the Saints in 2012 throwing 75% of the time have ~6 YPP and >5 YPR (not including turnovers)