## Drive Results

I had intended to post this a few weeks ago, but moved on to other things and forgot all about it. Continuing my data-dump series of posts of things that may only interest me, here are three graphs that illustrate the results of drives based on field position. Note that these are league-wide baselines, averaging all drives from the 2000 through 2007 seasons. Only drives that ended due to the expiration of time were excluded.

Each graph is based on first down field position. For example, for all 1st downs at a team’s own 35 yard line, offenses go on to score touchdowns 20% of the time. It doesn’t matter how the team got to the 35 yard line with a first down. They could have started the drive there or converted a first down from their own 20.

Field positions are defined as distance to the opponent’s end zone. A team’s own 20 is the “80 yard line.”

These graphs are part of my real-time win probability site. As the field position, down, and distance change during a game, my site continually updates the likelihood the offense will score either a touchdown or field goal.

The first graph depicts how often NFL drives result in scores.

The second graph depicts how often drives result in turnovers.

The third graph combines the two previous graphs and adds punts. It also groups the data into 5-yard increments so the lines are less noisy and easier to read.

### 8 Responses to “Drive Results”

1. Daniel says:

These are great graphs right here. Its interesting to see that an interception is basically twice as likely as a fumble recovered by the opposition.

Also, the turnover graph shows that the first half is roughly linear and increasing to about the fifty yard line, at which point it levels out.

Perhaps teams are more conservative with the ball as they get closer to the goal because they have "guaranteed" points in a field-goal?

Or they take more risks when they have more yardage between them and the goal line (like deep passes getting picked off).

On another note, I was wondering if you could tell me what programs you use to make graphs, harvest and organize data, etc.

I'd like to get more informative and data-driven with my posts and research on my own blog; I've been mainly doing anecdotal evidence and analysis so far.

Thanks!

2. Brian Burke says:

I'm not sure we can tell anything about getting conservative or other play-calling tendencies. The turnover rates are decreasing mostly because there is less distance to travel. Just to be clear, the way to read the graphs is: a team has a first down on the X yd line. How will this drive end?

On the other hand, turnovers are less common when pinned back on your own goalline even though offenses would have the longest distances to drive. This would indicate conservatism.

I use Excel (the latest version, which is a significant upgrade) for graphs and for most data analysis. I depend heavily on Excel's pivot tables feature. I use Gretl for regression and other advanced statistical stuff. I do a lot of coding using PHP for compiling the stats and building databases. MySQL is the database system I use. Apache/PHP/MySQL is what runs the win probability site.

3. Anonymous says:

Just found this site from a comment on PFR. Love it.

I'm fascinated by the fact that the probability of a FG never exceeds the probability of a TD. I'd have thought at some point FGs become a more likely result, but clearly not. Who thinks when you start a drive on your own 15 that a TD is more likely than FG? But, it is.

I'd like to see a line for (No TD/FD), i.e. drives stalls. Everything wouldn't add up to 100% anymore, but it'd be interesting to see if the chance of advancing the ball is relatively constant over the field or if there is evidence that teams get more conservative in certain areas and how big those areas really are.

Responding to the TOs leveling off, I suspect that's because TO per yard or play are roughly constant, hence the linear trend. TOs/drive will increase as drive length increases and drive length increases as the drive start position gets farther away from a TD. Eventually, though, the average drive length stops increasing, because TDs are no longer the controlling event. TOs/drive should correlate strongly with drive length. It'd be good to see a plot of drive length v. drive start, too.

4. Tarr says:

Very cool graphs.

If we break it out in 1-yard increments again, do we see a spike in "turnover on downs" at the 1 yard line?

5. Will says:

Good stuff. Everything looks sensibly monotonic except field goals, which makes sense since they get less likely as the touchdown becomes more likely near the opponent's goal line, and turnovers, which decrease deep in a team's own end, possibly as a result of conservative play. The really interesting one may be "downs/other," with two maxima, one near 30 yards and then the uptick right near the offense's own endzone. If it's significant, that increase deep in one's own end probably reflects the drives that end in safeties. But how do you explain the other bump, around 30 yards? Why does a 1st down at the 30 result in a failed 4th down conversion more often than a 1st down at the 20 or the 40? Interesting...

6. Will says:

Just answered my own question. Around the 30 you have the "no-man's land," too far for a field goal and too close for a punt. Duh.

7. Anonymous says:

Will,

I bet the reason has to do with missed field goals instead of being in "no-man's land". About 2 of 7 attempted field goals of 40+ are missed. The bump in downs/other correlates with field goals made.

8. Brian Burke says:

Just saw the last comment. If I remember correctly, the line for FGs is for FG attempts. Sorry I wasn't clear.