Comments on Advanced NFL Stats: No Man's Land

Jim A-Nope, no bias in the data. For between the 3...

2008-12-09T20:44:00.000-05:00

Jim A-Nope, no bias in the data. For between the 31 and 40 yd-lines, the avg field position for 4th down and every to go distance up through 4th and 20 is the 35.6, with almost no variation.

So I bet you're right--there's some irrational effect pushing coaches to try the field goal more often with shorter to go distances given equal field positions.

Interesting. Sort of like an impulse that says "w...

2008-12-09T09:53:00.000-05:00

Interesting. Sort of like an impulse that says "we came so close to getting a 1st down--we need to come away with some points."

But maybe there is some bias in the data. Perhaps 4th and 1s tend to happen closer to the 30 than the 40 for some reason. I doubt that's the case, but it could explain the difference. I can dig a little deeper to see.

This data is mostly predictable, but there's somet...

2008-12-09T09:04:00.000-05:00

This data is mostly predictable, but there's something fascinating about the 31-40 yardline play selection bars. Once a coach decides not to go for it, the decision to attempt a FG vs. punt appears to be dependent on how many yards to go. On 4th-and-1, teams kick a FG slightly more often than punt. On 4th-and-2, teams punt slightly more often than kick a FG. On 4th-and-3 and 4th-and-4, teams punt much more often than attempt the FG.

Unless there's some bias in the field position data, this doesn't seem to be a rational decision. Perhaps there's a psychological effect of being reluctant to punt on 4th-and-short that's driving these numbers.