## How Much Time Does It Take to Get into FG Range?

You wouldn't know by reading this study. It sets up a multiple regression model that uses a kitchen-sink approach to estimating the time needed in the end-game to get to the 35-yd line, commonly accepted as FG range. It uses QB rating, time remaining at the start of the drive, number of all-pro players on the offense, time outs remaining, starting field position, home field advantage, and whether the 2-minute warning is still available. The dependent variable is the time taken to reach the 35.

There are numerous fatal problems with this study. First, the model assumes linearity of the effects of predictor variables. I can tell you from my intimate familiarity with the variables involved that they are not linear at all. The model also assumes a normally distributed outcome variable, which is not investigated, and I doubt could be possible because games are bounded by the expiration of regulation time.

The study uses 3 seasons of data, which only yields 92 example situations to analyze.

The authors find enormous multi-collinearity problems with their model, and I'm not surprised. The model specification looks like this:

time taken = constant + field position + time outs+ ...a bunch of other stuff... + game time at drive start + game time when reached 35

But doesn't time taken = time at start - time when 35 reached? Of course. You can't have a regression model where the dependent variable is always the exact sum of two of the independent variables. The model's r-squared is 0.97, because it's one giant tautology.

But even more troublesome is how the study was set up and the selection/survivor bias at the heart of its methodology. It only uses end-game drives in which the offense indeed makes it to the 35. What about all the drives in which time expires prior to reaching the 35? To the authors' credit, they mention this flaw, but I don't think they appreciate how critical that point is to their entire approach.

It would much more useful to look at the success rates of reaching the 35 based on the time remaining and other game variables. That would include all the drives in question, not just the ones that made it to the 35.

Lastly, the authors should have produced one or more graphs or tables of their results--not the coefficients and p-values--but the actual time required to get to the 35 based on key variables, such as starting field position. They could have plot 4 curves, with time required as the y axis and field position as the x axis. Each curve would represent each of the possible timeouts remaining for the offense.

The bottom line is that this study amounts to a very costly waste of time. A simple Google search would have pointed them to the volumes of research at this site and others on the topic. If you're embarking on an academic research effort on something like this, please email me. I will point you in the right direction and save you months of wasted effort (and embarrassment). I've done that many times in the past, from 8th graders through post-doc research.

### 5 Responses to “How Much Time Does It Take to Get into FG Range?”

1. Jeff says:

Thank you for writing this. I was surprised how much circulation this study got despite its flaws.

2. Phil Birnbaum says:

Wow. I guess this study was mentioned in the press somewhere?

3. B. Riemann says:

This is hilarious, thanks for the note.

4. Cas says:

"Research done upon this subject is minimal at best, which
provides the opportunity to customize and pioneer a model relating in-game variables and the total time an offense needs to get downfield."

I think it was very nice of you to offer help, even if it might be construed as somewhat back-handed (8th grade comment...).

5. Brian Burke says:

I should clarify that it's not the authors I have a problem with. They're obviously curious, motivated and thorough young researchers who have a passion for the same topic I and the readers here do. My problem is with their research adviser who didn't bother to notice their model was circular. And it baffles the mind that in a sports management program in the Bay Area any instructor could be oblivious to Moneyball-style research outside the ivy covered walls.

The point is I offer free data, advice, models. And you really have to make an effort to avoid this site if you're poking around for NFL data.

The 8th grade comment was not intended as backhanded. I'm helping young Matthew with his game theory assignment this week. He's a Giants fan and will find that Tom Coughlin runs too often!