Longtime reader Eddy Elfenbein wrote me last night about the continuing demise of running in the NFL. From his own site:
The season is still young so this trend may not last, but with two weeks on the books, the number of rushes-per-game is down 4.8% from last year.
But what’s really dramatic is that average yards-per-carry is down by 10.7% (from 4.262 to 3.805).
Combine the two effects, fewer runs going not as far, and the rushing-yards-per-game figure is down 15% from last year. That number has been fairly steady for the last 25 years.
On the passing side, attempts are up 7.8%, completions are up 10.0% and passing yards are up 10.5%. The league-wide passer rating now stands 87.3. In 1994, that would have qualified as fourth-best in the league.
I'm surprised league-wise rushing yards per carry is so low. I don't recall anything below 4.1 YPC, even this early in the season. Aside from that, everything else is part of a continuing trend--something game theory predicts will continue.
You guys seem like nice guys and trying to do right things. I want to caution to you that math and football is like mashed potatoes and Gravy; they don't mix.
I played high level youth football until found out I got asthma. Coach told my mom I would probably have gone pro because of my good talent. Football is played on the gridiron and not on a computer.
I think it is best for you to leave the talk of football to the experts like those of us who played the game.
Easy there, Uncle Rico.
By the way, you're wrong. I'm not a nice guy.
This is really interesting to me, 'cause from the games I've seen so far this season, I keep noticing that the play calling is affecting the running game. It's like, the pass attempts are not doing anything for spreading out the defense. Also, I notice runners seem to be runnning to the same spots too much, making it very predictable. And often, they seem to be trying to go through a D's strong side.
I'm no football expert but that's how it appeared to me in the parts of 4 games I've seen this month. It's been quite frustrating to watch that. I thought it was just in my head.
IIRC wasn't last season one of the best seasons ever in terms of YPC?
If we can assume that all 64 YPC results so far are statistically independent, I have p=0.2% for the statement that we are looking at a real reduction in YPC from last year. That is based on var_YCP_random = 1.445 for a single game from here: http://redd.it/1b6oyb
So the standard deviation in error after 64 results is sqrt(1.445/64) = 0.150. p = 2*( 1 - norm_cumulative(x=3.805, mean=4.262, stdev = 0.150) ) = 0.002.
I should not have said you can not talk about football. you can do whatever you wanting of course, its true. I will be patient for you, to give time to change your mind about trying to understand football with a calculator.
My middle school coach played semi-pro; he'd always tells us to run the football down their throat.
I don't think one could design a better troll than John Black.
Thank you SportsGuy. it's nice to know that just because i'm leading the a cause against what your group believes, doesn't mean we can't have respect. You guys aren't bad people; you just don't know what football is because you couldn't play it. that's not your fault. https://twitter.com/FootballHunch
I wouldn't not have been so easygoing a couple decades ago during my hay day though, however.
3.8 YPC is not so shocking; you only have to go back to 1996 to find a season with league-wide 3.8 or lower YPC. It is more surprising if you consider the recent improvement of offense, both passing and rushing. 2011 and 2012 were both historic high points for YPC.
What stands out most to me, though it is still early, is the pass attempts/rush attempts ratio--the number of times a team calls a pass play for each rush. So far it is 1.48 for 2013. The next highest is 1.31 in 2012 and 1995. Maybe a result of lots of close games?
"High level youth football" - lol!
John-I hope you comment on every post here.
You guys should see his tweets to me.
/break/
TB- I agree that it might be the result of a lot of close games in the first couple weeks. Maybe more goal line or short yardage situations too.
Thank you Brian I will try. i'll have more time after my discrimination suit against my former employer is over.
I's glad twitter wasn't around when I was ballin, because I would have lit up some fools.
John -
Why are reading articles at a site called "Advanced NFL Stats"?
I think we're having our leg pulled...
John, I'll take the bait. What happened with your former employer?
TPM, youll having to ask Brian why his site is called Advanced NFL Stats.
Brian, my former employer wrongfully fired me for making supposed "intentionally false" statements on my hiring documents.
[on-topic]
What's interesting to me is that running efficiency (YPC) and carries have BOTH decreased. Game theory predicts that as carries go down, YPC will go up as defenses compensate... I suppose we should predict that defenses just haven't compensated yet, and soon (later this year or next year) running efficiency will go up and passing efficiency will go down when defenses finally do compensate.
But there's always the terrifying possibility that the evolution of football offenses point toward an equilibrium of 100% passing...
Adam, true but it's a chicken/egg question. Game theory says as running effectiveness decreases, attempts should also decrease. So what we are seeing (if it's a real effect) makes sense.
We aren't seeing much in the way of read option from RG3 or Kaepernick yet. Those guys boosted the run totals all by themselves LY. Am waiting for articles detailing this season's countermeasures up at Grantland.
> Game theory says as running effectiveness decreases,
> attempts should also decrease.
That's only true for an somewhat unusual notion of effectiveness -- it may make game theoretical sense to use running plays, even if the expected result is a loss of yardage. Consider, for comparison, that bluffs in poker are plays which themselves have negative expected outcome, but improve the expected outcome of the strategy as a whole.
Nate-that may be true, but how often do you need to bluff for the desired strategic effect? Should a poker player bluff 45% of the time? Wouldn't he lose a heck of a lot of chips along the way on called bluffs?
Bluffs in poker are often positive expectation. For instance if I am betting $1 into a $5 pot and I believe you will fold 40% of the time, then my expectation is +$1.40.
In terms of the deeper point, it might be true that you could come up with a theoretical payoff matrix in which running negative expectation plays would sometimes be optimal, but if we stipulate a further reduction in payoff to those plays, wouldn't that always result in a lower optimal frequency?
In other words, if we imagine that the only change is that running has gotten harder, the one thing we should be able to say for sure is that teams should not run more (and probably they should run less.)
Reflecting on things a little - for any simple game, against a Nash Equilibrium strategy, all viable options must have equal expected payoffs. (Otherwise it's not an equilibrium strategy.)
Things like the negative EV bluff really only show up as facets of larger strategies, or when all the other options are similarly -EV.
That said, along with a bunch of other questionable assumptions, it's rather silly to think of NFL play calling to be done with any kind of NE strategy. It looks like passing is under-defended, so teams should keep taking to the air more until defenses adjust.
But what does John Black think about Nash Equilibria?
The Nash Equilibria would be the name of the Tennessee Titans if you had anything to do with it