On playing for OT and game theory

[hawkfan340]

Much like the OSU game from 2 years ago, today Iowa sat on the ball (with time outs in hand) in an opponent's stadium to force the game into overtime. And just like against OSU, it was the wrong strategy -- plain and simple.

From a game theory perspective, the idea of sitting on the ball is to reduce the risk of loss due to turning the ball over. However, the reality is that you have not reduced this risk. You have only changed the time in which you will incur this risk.

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

Along with simply moving the risk from regulation to OT, you obviously also forfeit the strategical advantage of being able to kick the winning FG in regulation.

The only way you could argue that sitting on the ball is better is if you feel the probability of a turnover in regulation is much greater than the probability of a turnover in overtime. This marginal difference in turnover probability must be greater that the probability of making the gaming winning FG drive.

Plain and simple, it's hard to argue that someone you're 3-5 times more likely to turn the ball over just because it's regulation. It just doesn't make sense.

** Note that all this doesn't mean that KF is a bad coach or should be fired. It just means he's not really playing the percentages if he's sitting on the ball waiting for OT, especially when on the road.

 

[Mac50702]

Much like the OSU game from 2 years ago, today Iowa sat on the ball (with time outs in hand) in an opponent's stadium to force the game into overtime. And just like against OSU, it was the wrong strategy -- plain and simple.

From a game theory perspective, the idea of sitting on the ball is to reduce the risk of loss due to turning the ball over. However, the reality is that you have not reduced this risk. You have only changed the time in which you will incur this risk.

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

Along with simply moving the risk from regulation to OT, you obviously also forfeit the strategical advantage of being able to kick the winning FG in regulation.

The only way you could argue that sitting on the ball is better is if you feel the probability of a turnover in regulation is much greater than the probability of a turnover in overtime. This marginal difference in turnover probability must be greater that the probability of making the gaming winning FG drive.

Plain and simple, it's hard to argue that someone you're 3-5 times more likely to turn the ball over just because it's regulation. It just doesn't make sense.

** Note that all this doesn't mean that KF is a bad coach or should be fired. It just means he's not really playing the percentages if he's sitting on the ball waiting for OT, especially when on the road.

Hey best post of the day.

 

[bitmap]

Awesome post! Expand that into the third OT. I keep hearing how "conservative" KF is in his play calling. The truly conservative approach on the gift 4 and 1 would have been to run Coker up the middle or have JVB do a QB sneak. Let me explain - Get the yard and we get at least four more downs. We had their defense on the field at the bottom of the 2nd OT and scored on them. Now at the top of the 3rd OT we get to run against them for a second series in row. Wear them down some more, the advantage is ours at this point and we also rest our D by staying on the field. Nothing fancy, just good high percentage plays. If we are willing to settle for the FG, why not put it off for two more minutes?

The risky play call would have been to go for the win, the all or nothing approach calling a deep strike down the field. Not KF's philosophy for good reason. If the offensive execution had been there all day, maybe, but then again if good execution had been there all day we wouldn't be in OT, right?

Settling for the FG was the stupid call....

 

[latrans]

Much like the OSU game from 2 years ago, today Iowa sat on the ball (with time outs in hand) in an opponent's stadium to force the game into overtime. And just like against OSU, it was the wrong strategy -- plain and simple.

From a game theory perspective, the idea of sitting on the ball is to reduce the risk of loss due to turning the ball over. However, the reality is that you have not reduced this risk. You have only changed the time in which you will incur this risk.

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

Along with simply moving the risk from regulation to OT, you obviously also forfeit the strategical advantage of being able to kick the winning FG in regulation.

The only way you could argue that sitting on the ball is better is if you feel the probability of a turnover in regulation is much greater than the probability of a turnover in overtime. This marginal difference in turnover probability must be greater that the probability of making the gaming winning FG drive.

Plain and simple, it's hard to argue that someone you're 3-5 times more likely to turn the ball over just because it's regulation. It just doesn't make sense.

** Note that all this doesn't mean that KF is a bad coach or should be fired. It just means he's not really playing the percentages if he's sitting on the ball waiting for OT, especially when on the road.

I would guess that is exactly why KF does what he does. That being said 1:17 left with 2 timeouts seems like a good time to try and win the game.

 

[hawkfan340]

I would guess that is exactly why KF does what he does. That being said 1:17 left with 2 timeouts seems like a good time to try and win the game.

But for this to be true, you'd have to be like 5 times more likely to have a turnover in regulation than in OT. I just don't think that's realistic. Maybe there's a slight increase because you could be pressing, but it's not like all the sudden you're playing against 15 guys or something.

The point is, I wanted to look at it from an unemotional, economic viewpoint. We're giving up something of value (chance for the winning FG) for something of relatively little value (protecting ourselves from the marginal difference of turnover risk in regulation vs. OT). That doesn't make sense when you boil it down. We're not playing it correctly.

 

[MoseSchrute]

I'm with you 100%. Very well said, sir.

 

[DaBears]

Don't need a theory, just no faith in Vandy. Not having faith in your QB says a lot about the confidence the coaches have with this offense.... NONE.

 

[1hawkeye1]

I approve this message.

 

[KCHawkeyes]

I have asked this question many times before, "Do we practice the 2 minute offense"? We have no clue how to run it no matter who the QB is.

 

[MJSCK]

I have asked this question many times before, "Do we practice the 2 minute offense"? We have no clue how to run it no matter who the QB is.

It's deja vu. We've talked about the lack of time management and a two-minute drill for years.

KF and O'Keefe simply must not teach a solid two-minute offense. I suppose you could say he doesn't trust his team or the talent of the personnel, so he thinks playing in OT is better. That doesn't jive, though. If that's the case, why play the game at all if you're simply playing not to lose?

You have to attack! Especially with this defense. And who would know that better than anybody else? The coaches. The guys who see the D is soft this year and incapable of being leaned on as they have in the past.

They don't have a two minute drill.

 

[pacoyal]

Not that the majority is always correct, but when has anybody else seen a team play for OT on the road. I still can't get over that.

On a side note, when will the D line figure out how to keep contain and stay in their lanes? Missed tackles suck, but when a sack is missed and all the QB has to do is run to the other side which has no defenders, you give up a free pass/run/1st down/almost anything.

 

[SeattleHawkeye]

It's deja vu. We've talked about the lack of time management and a two-minute drill for years.

KF and O'Keefe simply must not teach a solid two-minute offense. I suppose you could say he doesn't trust his team or the talent of the personnel, so he thinks playing in OT is better. That doesn't jive, though. If that's the case, why play the game at all if you're simply playing not to lose?

You have to attack! Especially with this defense. And who would know that better than anybody else? The coaches. The guys who see the D is soft this year and incapable of being leaned on as they have in the past.

They don't have a two minute drill.

I wholeheartedly agree w/ your point on attacking a defense, specifically deep, but it has to be done intelligently, not just randomly. For example, on the Coker fumble, I think that if earlier they had actually tried to go deep or at least run deeper routes, there wouldn't have been as many ISU defenders waiting for Coker. (Make no mistake, there's no excuse for Coker's continued fumbles.) ISU dropped 8 guys into the box all day, and I don't recall IA ever going over the top, specifically deep down the middle. If they would have at least tried, ISU backs off. (And if you go vertical and ISU doesn't back off, keep doing it.) Instead, we don't challenge them vertically, and it makes running the ball that much harder. Also, the route combos are often at the same level, usually shorter routes. If you don't threaten your opponent w/ a vertical passing game, or at least consistently multilevel routes, a D will gradually creep up and plug running lanes and challenge your shorter passing game. At times IA's play calling plays right into the hands of opposing D's.

 

[bitmap]

I would guess that is exactly why KF does what he does. That being said 1:17 left with 2 timeouts seems like a good time to try and win the game.

Wow! Searching through some old threads and just saw this again. Knowing what we know now, about the "no huddle" offense the Hawks are capable of, this strikes me as being even more right on the mark. All we needed was a FG and Meyer was perfect that day.

 

[homerHAWKeye777]

Much like the OSU game from 2 years ago, today Iowa sat on the ball (with time outs in hand) in an opponent's stadium to force the game into overtime. And just like against OSU, it was the wrong strategy -- plain and simple.

From a game theory perspective, the idea of sitting on the ball is to reduce the risk of loss due to turning the ball over. However, the reality is that you have not reduced this risk. You have only changed the time in which you will incur this risk.

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

Along with simply moving the risk from regulation to OT, you obviously also forfeit the strategical advantage of being able to kick the winning FG in regulation.

The only way you could argue that sitting on the ball is better is if you feel the probability of a turnover in regulation is much greater than the probability of a turnover in overtime. This marginal difference in turnover probability must be greater that the probability of making the gaming winning FG drive.

Plain and simple, it's hard to argue that someone you're 3-5 times more likely to turn the ball over just because it's regulation. It just doesn't make sense.

** Note that all this doesn't mean that KF is a bad coach or should be fired. It just means he's not really playing the percentages if he's sitting on the ball waiting for OT, especially when on the road.

It would appear that you need to learn more game theory. For one thing, you weren't accounting for all the possibilities that could have occurred. Similarly, you weren't accounting for "inertia" in the game. Also, you weren't accounting for the fact that every new set of downs IS NOT uncorrelated.

Not only was there the concern of turning over the ball, there was also the concern of having a bunch of quick incompletions ... and having them in poor field-position.

While some folks like to bring up how GOOD Vandenberg has looked in our no-huddle O ... what those same folks are forgetting is that Vandenberg really wasn't in much of a rhythm through that ENTIRE game! Thus, if you're playing the percentages ... then you quickly realize that at the end of the game, if you come out aggressive and passing ... the most likely scenario would be a rash of incompletions. What's more ... the ISU kicker proved that he had plenty of leg to hit a 50 yarder!

The whole strategy of NOT playing for OT is reliant on field position! That's a critical element that folks have absolutely FAILED to note.

 

[CP87]

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

As long as this has been bumped back up, I have a disagreement with your logic.

The proper comparison is not the risk of a turnover at the end of regulation vs. a risk of a turnover in OT. The comparison of interest is the risk of a turnover at the end of regulation vs. the risk of something positive happening at the end of regulation.

Just because you do not turn the ball over, that does not mean you win the game. You still have to move it at least 50 yards down the field and line up and successfully kick a long FG.

As an extreme example, would you run a play from your own 15 yard line with 5 seconds on the clock? Probably not, but not because your risk of a turnover is greater in this instance, but rather because your risk of something positive happening is even smaller than your risk for a turnover (especially if your QB is Marquis Gray and you are playing a school from the Dakotas).

The Iowa coaches have traditionally felt that it is more likely that their offense will screw up than that it will do something positive in end-of-game/end-of-half situations.

I am not justifying the decision to sit on the ball at the end of the ISU game; I hated it at the time, and I hate it more in retrospect. It does not show any confidence in the offensive half of your team. The only "positive" is that you cover your butt as a coaching staff by doing something that hundreds of coaches have done throughout the ages. Thus, if it doesn't work you don't catch as much hell is if you had thrown a pick-6 at the end of regulation.

Time will tell what will happen the next time the 2011 Hawks find themselves in such a situation.

 

[whammer33024]

to add to homer some, i feel that the 2 minute offense forces players to rush some things, and rushing leads to more mistakes. in OT, there's no clock(aside from the playclock).

 

[austinhawk]

Much like the OSU game from 2 years ago, today Iowa sat on the ball (with time outs in hand) in an opponent's stadium to force the game into overtime. And just like against OSU, it was the wrong strategy -- plain and simple.

From a game theory perspective, the idea of sitting on the ball is to reduce the risk of loss due to turning the ball over. However, the reality is that you have not reduced this risk. You have only changed the time in which you will incur this risk.

What I mean is that a turnover in OT is no different than a turnover in regulation. A turnover at either time essentially seals your fate. You have not reduced the risk of turnover, only changed the time you will incur that risk from regulation to overtime.

Along with simply moving the risk from regulation to OT, you obviously also forfeit the strategical advantage of being able to kick the winning FG in regulation.

The only way you could argue that sitting on the ball is better is if you feel the probability of a turnover in regulation is much greater than the probability of a turnover in overtime. This marginal difference in turnover probability must be greater that the probability of making the gaming winning FG drive.

Plain and simple, it's hard to argue that someone you're 3-5 times more likely to turn the ball over just because it's regulation. It just doesn't make sense.

** Note that all this doesn't mean that KF is a bad coach or should be fired. It just means he's not really playing the percentages if he's sitting on the ball waiting for OT, especially when on the road.

True Dat!!!

 

[KnightHawk9]

Somewhere in an alternative universe, Kirk decided to go for it, JVB threw a pick six, and people 2 years later still debate on Hawkeye message boards why we didn't just sit on the ball and play for overtime.