## Hail Mary Probabilities

Baseball guru Tom Tango noticed the WP graph for the recent Monday night meltdown had the Seahawk's chances at 0.01 prior to the game-ending play and thought it must be too low. He's correct.

There's a distinction between the WP model’s empirical methodology and its automatic output without any intervention or input from a human. When I do a detailed analysis for any specific play, I have the luxury of time and logic to dig directly into the data. The “auto” model that spits out WP estimates without any human input is based on lots of assumptions and interpolation on top of extrapolation etc. There are literally billions of combinations of game states (yd lines, downs, to go distances, seconds remaining, score difference, time outs). It’s just a matter of how much time I can put into coding the calculator to handle special cases like “a team's very last desperation play.”

With all the attention on that final play in the GB-SEA game, I thought it would be useful to look at Hail Mary success rates.

I looked at all the situations in which there were 8 or fewer seconds remaining (enough for one play) and a team needed a TD to tie or win (down by 4 through 8 points), and see how often they got that TD based on field position. This includes any plays that ultimately result in a TD, which would include any defensive penalty that enabled a subsequent scoring play. Even though SEA was specifically down by 5, the situations for being down by 4-8 are effectively indistinguishable for the purpose of estimating the chance they can get the TD.

From the 2000 through 2011 seasons, there were 223 examples in total--a little over 20 per 10-yard bin of field position. The chart below plots the TD success rate in the sample.

The Seahawks were at the 24 yd line, which would correspond to just over a 10% chance of a TD and winning the game.

Notes:

-The one indicated TD from the offense’s own 24 was from a 2003 NO-JAX game. I'm told that's an error. Thanks for the correction.

-It’s possible in short ranges (inside the 20), teams could get off 2 plays, not just 1 with 8 sec left. Limiting the results successively to cases within 7, 6, 5 sec, etc. doesn’t affect the results around the 20-30 bin. Actually, only the 1-10 yd line bin drops, and it becomes measurably lower than the 10-20 yd bin. This might suggest that it’s easier to score from slighter further back from the goal line. (There’s more space for receiver routes.)

### 9 Responses to “Hail Mary Probabilities”

1. Jeff says:

Typo: "I looked at all the situations in which there were 8 or fewer seconds remaining (enough for one play) and a team needed a TD to tie or win (down by 4 through 9 points)...."

Should be "4 through 8 points" at the end there.

2. Unknown says:

The Saints did NOT win the River City Relay game.

3. Andrew says:

The Saints actually lost that 2003 game at Jacksonville. They missed the XP after a miracle, multi-lateral touchdown, and lost by one. The loss eliminated them from playoff contention as well. Probably one of the few final plays of a game even more ridiculous than what happened last night.

Pedantry aside, good stuff, per usual.

4. Anonymous says:

Seems like this implies that the "auto" end game WP model could use some improvements. Maybe using Keith's Markov techniques? Since WP swings are so dramatic near the end of close games, wouldn't this also have a huge impact on WPA-based individual stats?

5. Ian Simcox says:

Interesting to compare this one back to the Redskins 4th-16 v 62yd FG from the other day.

A hail-mary pass has a lot of similarity to a 4th-16 at the end of a game. You are desperate for a lot of yardage and you aren't concerned how you fail in your attempt (so even if there's no great window you'll still throw the ball up and give your receiver a chance, even if it might result in an INT).

In the hail mary case, a 16 yard 'desperation' pass is successful some 20-25% of the time. Suggests that perhaps the Redskins had an even greater chance of converting than you thought.

6. James Sinclair says:

Would it make much of a difference to include games where a team was trailing by three or less and was out of field goal range? I remember the Jaguars winning on a long hail mary a few years ago in a game that probably isn't reflected in your data because the margin was less than four (it might've even been tied, but including ties seems problematic for obvious reasons).

Also, how is the 2003 Jacksonville-New Orleans game an error? They missed the extra point, but that has nothing to do with the preceding touchdown.

7. bytebodger says:

I'm a Jaguars season ticket holder, and yes, that game a few years ago was tied before they hit the Hail Mary. The Jags were right around the 50 yard line and the ball was batted back into the field of play by a Texans DB, where it was caught by Mike Thomas who easily walked into the end zone. I sit in the north end zone about 10 rows up and the play basically happened right in front of me.

8. J.D. says:

Ian - the chance of gaining 16+ yards on 4th-and-16 is generally higher than getting a TD from the 16 yard line on the last play. If the offense has a 4th-and-16 at, say, midfield, the defense needs to protect not only against the offense getting the first down, but also against them going deep. If the offense needs a TD from the 16 yard line, the defense just needs to protect the end zone, and can play much tighter.

9. Unknown says:

Brian;
wouldn't the chance of defensive pass interference (and one more play on the 1 yard line) increase the prob.% a bit as well?