BBTF's Newsblog Discussion :: IDS News: IU researcher designs football strategy device

Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Friday, April 28, 2006

IDS News: IU researcher designs football strategy device

nmsnc points out this “sabermetricish stuff in football”.

Senior scientist Charles Bower from the IU physics department and partner Frank Frigo of Louisville have created ZEUS, an “off-the-shelf laptop” that simulates different conclusions to football games based on the decisions made during plays, according to a press release. It’s designed to help football coaches make quick play decisions during games.

...Bower provided an example of a play in which ZEUS can be applied to help a coach make a decision based on solid statistical data. A coach weighs the outcome of a crucial decision during a play and has to consider the consequences. For example, at fourth down and a few yards, should the team punt the ball or go for the first down? This tool allows the coach to see what the chances of success of each decision are.

ZEUS can help predict what the team will gain or lose from the number of decisions available to the coach at that point in the game.

Repoz Posted: April 28, 2006 at 12:24 PM | 33 comment(s) | Login to Bookmark
Related News: General, Sabermetrics, Products

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 1 of 1 pages

1. jmac66 Posted: April 28, 2006 at 12:38 PM (#1993907)

ANY statistical analysis of football (going back to Hidden Game and many others) has claimed that going for it on 4th down (rather than punting or trying a FG) is usually preferable

the chances of coaches adopting that stategy approaches zero

2. Russ Posted: April 28, 2006 at 12:52 PM (#1993922)

ANY statistical analysis of football (going back to Hidden Game and many others) has claimed that going for it on 4th down (rather than punting or trying a FG) is usually preferable

It's weird, but I rarely if ever punted when playing football video games... it just felt like if I had a better than 50% chance of getting the first and I wasn't going to give a guy more than a field goal, it was better to go for it.

Sometimes I wonder if the punt on 4th down thing is just a vestige of the game remaining from when you couldn't throw a forward pass.

3. DiggerP Posted: April 28, 2006 at 12:53 PM (#1993924)

Charlie Weis went for it on 4th down an incredible number of times last year. Obviously he had a good offense, but he also had a decent kicking game. I always assumed that he is just so confident that he believed he could make the call that would convert any 4th down. I wonder if he was aware of this kind of study on some level.

4. Designated Sitter (GGC) Posted: April 28, 2006 at 12:54 PM (#1993925)

ANY statistical analysis of football (going back to Hidden Game and many others) has claimed that going for it on 4th down (rather than punting or trying a FG) is usually preferable

Depends on field position and kickers (are Aussie style kickers more proficient at the coffin corner kick?), but teams should go for it more often than they do. OTOH, you're looking at it from the perspective of maximizing gains, while alot of coaches are more interested in minimizing their losses. The game theory term for this escapes me at the moment.

5. Designated Sitter (GGC) Posted: April 28, 2006 at 12:59 PM (#1993932)

Weis did come from the Pats, who have a rep for being football scholars.

6. Dr Love Posted: April 28, 2006 at 01:06 PM (#1993946)

OTOH, you're looking at it from the perspective of maximizing gains, while alot of coaches are more interested in minimizing their losses. The game theory term for this escapes me at the moment.

It's called "covering your ass."

7. Greg Maddux School of Reflexive Profanity Posted: April 28, 2006 at 01:12 PM (#1993949)

It's weird, but I rarely if ever punted when playing football video games... it just felt like if I had a better than 50% chance of getting the first and I wasn't going to give a guy more than a field goal, it was better to go for it.

John Madden thought it was unrealistic that people weren't punting, so he demanded that fourth downs be made harder to convert in an attempt to force players to use conventional strategy.

8. jmac66 Posted: April 28, 2006 at 01:12 PM (#1993950)

It's called "covering your ass."

don't resort to technical terminology in here

9. jmac66 Posted: April 28, 2006 at 01:15 PM (#1993954)

John Madden thought it was unrealistic that people weren't punting, so he demanded that fourth downs be made harder to convert in an attempt to force players to use conventional strategy.

if true, that's astonishing (I've never played these games, so I don't know anything about them)

he demanded that they convert true statistical probabilities into "everybody knows it's true" probabilities?

there should be a Greek word for that

10. Russ Posted: April 28, 2006 at 01:24 PM (#1993966)

there should be a Greek word for that

Semper ubi sub ubi.

11. Designated Sitter (GGC) Posted: April 28, 2006 at 01:37 PM (#1993978)

BTW, as scoring increases, it makes more sense to go for it on fouth down IIRC. Way back in the old days, they were teams that would punt on first down if they had poor field position. Of course, that was when the ball was easier to kick and players were smaller.

12. Craig in MN Posted: April 28, 2006 at 02:36 PM (#1994039)

Way back in the old days, they were teams that would punt on first down if they had poor field position.

I don't get that at all. If you are playing for field position, doesn't it make more sense to try to grind out 5 yards and then punt? It's like getting an extra 5 yards of carry on your punt. I guess if you were fighting the clock it might be worthwhile to just punt. But you need to have a better punter than the opponent (or have the wind to help you), or else they could just punt it back to you and you've gained nothing.

13. kthejoker Posted: April 28, 2006 at 02:48 PM (#1994051)

Madden may be crazy, but conventional wisdom also suggests that 4th down turns into an intense battle of wills, whereas most video games treat it more or less as just another down. It is rare to see video games adopt an increased level of intensity as games enter into distinct high-pressure situations. Some games have had some gimmicky attempts at this, by letting users control crowd noise which can affect a player's stress, etc, but in general, it's a non-factor in most video games.

I would think that from a video game standpoint, increasing the intensity and general tension of the player in conventionally highly intense moments (bottom of the 9th, 2 outs - 4th and 1, time ticking down, etc) would be a *good thing*, even if statistics indicate those situations are no different than the same situation in the 1st inning or 1st quarter.

But that's more an aesthetic than statistical viewpoint.

14. Vaux, A.B.D. Posted: April 28, 2006 at 03:25 PM (#1994087)

It always annoyed me that computer football games fostered strategies so different from what actually happens in real games. Statistical probabilities are what they are, of course, but I like to do in a game what I would have to do if I were really coaching a team. Teams play different defenses on fourth downs, quarterbacks are apt to panic (like LaTroy Hawkins in the ninth inning)... if I were coaching, I'd probably go for it a lot on 4th and 4 or fewer, but more than that are tougher to get on fourth than any other down for those human reasons, I'm sure.

15. Slivers of Maranville (SdeB) Posted: April 28, 2006 at 03:25 PM (#1994088)

he demanded that they convert true statistical probabilities into "everybody knows it's true" probabilities?

I think that the results of a play on Madden have very little to do with "true statistical probabilities."

16. Fridas Boss Posted: April 28, 2006 at 03:33 PM (#1994099)

while alot of coaches are more interested in minimizing their losses. The game theory term for this escapes me at the moment.

Maximin strategy.

17. Jack of Arcades Posted: April 28, 2006 at 03:43 PM (#1994109)

It's like when Jim Mora punted last year in OT in his own territory and played for a tie, even though a tie all but eliminated them from the playoffs whereas a win would give them a much better shot. Of course, they lost the game anyway.

18. Will Young Posted: April 28, 2006 at 03:49 PM (#1994118)

I don't get that at all. If you are playing for field position, doesn't it make more sense to try to grind out 5 yards and then punt? It's like getting an extra 5 yards of carry on your punt. I guess if you were fighting the clock it might be worthwhile to just punt. But you need to have a better punter than the opponent (or have the wind to help you), or else they could just punt it back to you and you've gained nothing.

Not necessarily. The punters and kickers back then were typically the quarterback or other player rather than a specialized position. This way, a team could line up a common formation (like the single-wing) and punt the ball when the defense had no one deep and hope that it would roll/bounce more than five extra yards. Michigan with Brian Griese and the Eagles with Cunningham have both used the quick-kick recently to some success.

On a related note, the new Viking uniforms are absolute hideous. Just awful.

Back to baseball...

19. Vrhovnik Posted: April 28, 2006 at 03:51 PM (#1994120)

Semper ubi sub ubi.
Not Greek, but it reminded me of my Latin I class and "Civis Romanus."

Oh, and the IU football team needs all the help it can get. Good to see the academics are pitching in.

20. pkb33 Posted: April 28, 2006 at 04:02 PM (#1994129)

Belichick has a paper some math guy did which goes through the 'correct' move on fourth down from all points on the field, score situations, and down/distance situations. That is part of why he goes for it more often than most.

I assume Weis has that same info, though its worth noting that Weis last year also had a very poor defense, too

21. Moloka'i Three-Finger Brown (Declino DeShields) Posted: April 28, 2006 at 04:32 PM (#1994163)

Eagles with Cunningham . . . used the quick-kick recently to some success.

I distinctly remember the Eagles doing this (not on first down, of course) with Cunningham in a 1989 game at the Meadowlands. Cunningham boomed the ball and got a tremendous Astroturf roll.

Ninety-one-yard punt. Yeah, I'd call that . . . successful.

22. Moloka'i Three-Finger Brown (Declino DeShields) Posted: April 28, 2006 at 04:35 PM (#1994167)

(And it turns out the punt was the fourth-longest in NFL history.)

23. Bad Doctor Posted: April 28, 2006 at 06:03 PM (#1994336)

I distinctly remember the Eagles doing this (not on first down, of course) with Cunningham in a 1989 game at the Meadowlands.

Nah, that wasn't a quick kick. Cunningham just waved the punter off the field on 4th down, but it was clear that a punt was coming. I think Dave Meggett was back to field it, but he totally misplayed it.

There was a hell of a wind that day too.

24. Moloka'i Three-Finger Brown (Declino DeShields) Posted: April 28, 2006 at 06:32 PM (#1994389)

Hmmm. I guess I didn't remember the play as distinctly as I thought.

Strange, though, as I remember the year, the location, the yardage, and what I was doing precisely at that moment. But I didn't remember the specific situation. Funny the way memory works.

25. Designated Sitter (GGC) Posted: April 28, 2006 at 06:43 PM (#1994413)

Was that the game covered in the book No Medals For Trying?

I never knew that Randy was Sam Cunningham's brother.

26. Moe Greene Posted: April 28, 2006 at 06:50 PM (#1994421)

My memory of that punt is that the punter was hurt, so Cunningham had to do it. Don't remember him needing to wave the punter off.

27. Los Angeles Waterloo of Black Hawk Posted: April 28, 2006 at 07:02 PM (#1994442)

John Madden thought it was unrealistic that people weren't punting, so he demanded that fourth downs be made harder to convert in an attempt to force players to use conventional strategy.

Of course, the way this was implemented was completely hamhanded. Instead of just making the defense tougher on 4th downs, they make the offense more inept -- e.g., you're more likely to throw a perfect pass to a wide open receiver who drops it than you are to have a situation where all of your receivers are well-covered.

Anyway, I have the version from a few years ago, where that is the case. That edition also has very poor re-creations of goal-line situations. Dunno if that's been improved in more recent editions.

28. Designated Sitter (GGC) Posted: April 29, 2006 at 03:36 AM (#1995526)

Nah, that wasn't a quick kick. Cunningham just waved the punter off the field on 4th down, but it was clear that a punt was coming. I think Dave Meggett was back to field it, but he totally misplayed it.

There was a hell of a wind that day too.

Bad Doctor is correct. It was fourth down and Max Runager, the Iggles punter was more known for his accuracy than his distance, so they had Cunningham punt. Cunningham booted the ball 60 yards in the air and it bounced another 31.

Thanks go out to Jerry Izenberg and No Medals For Trying.

29. Jim A Posted: April 29, 2006 at 01:08 PM (#1995652)

Belichick has a paper some math guy did which goes through the 'correct' move on fourth down from all points on the field, score situations, and down/distance situations. That is part of why he goes for it more often than most.

You're referring to this Bellman Equation paper by David Romer: http://emlab.berkeley.edu/users/dromer/papers/nber9024.pdf

Most coaches know about this paper, but Belichick is probably more likely to understand it, given his economics background.

Similar analysis has been done by William Krasker on footballcommentary.com.

A more extreme view recommends never punting because the play calling strategies associated four down mindset will result in a more efficient offense:
http://actualturnoverratio.blogspot.com/

30. Greg Franklin Posted: April 29, 2006 at 04:48 PM (#1995744)

I think teams nowadays do go for it more than in the old days. I've heard it credited to the changed field goal rule, where a missed FG was spotted at the location of the kick instead of the line of scrimmage. The NFL was trying to kill its reputation as the National Field-Goal League.

The rule change convinced coaches to attempt fewer long field goals, and as a byproduct attempt more conversions. The maximin mentality is there, but not as strong as it used to be.

31. Designated Sitter (GGC) Posted: April 29, 2006 at 07:08 PM (#1995917)

This has nothing to do with this article, but it is football related so I'll mention it here.

I've long thought that there were stadium effects in football; similar to park effects in baseball. I was fooling around for an hour or so with the results in the New York Giants games for the past 10 seasons. It might mena little, but the Giants games in the Meadowlands averaged 39.90123457 ponts per game (both teams) while their away games averaged 37.16455696 points.

I was surprised, I figured that Giants Stadium would keep the scores down; at least in the latter part of the year. But it's not like football schedules are anywhere near balanced as baseball skeds.

32. Kiko Sakata

Posted: April 29, 2006 at 07:40 PM (#1995955)

the Giants games in the Meadowlands averaged 39.90123457 ponts per game (both teams) while their away games averaged 37.16455696 points

I love that you present these numbers to eight decimal places.

33. Designated Sitter (GGC) Posted: April 29, 2006 at 08:16 PM (#1995993)

I love that you present these numbers to eight decimal places.

I just copied them and pasted them from the spreadsheet.

Page 1 of 1 pages

You must be Registered and Logged In to post comments.

<< Back to main

Friday, April 28, 2006

IDS News: IU researcher designs football strategy device

Reader Comments and Retorts

Support BBTF

Bookmarks

Hot Topics