Crazy thought of day: Cowboys should have missed extra point on purpose with 2:32 left so they'd be up 3, not 4

dingle wrote:

Consider this situation: A psychopathic killer with bad hair tells you that you live if a coin he flips comes up heads and then gives you a choice between one that is fairly weighted or one that is weighted to come up heads 53% of the time. I'm going with the unfair coin even though I only get one flip.

That's not exactly a fair analogy since Rojo's hypo is between playing for tie (and going to overtime) versus increasing one's odds to actually win.

But I think malmo's point is that statistics divorced from reality are worth much. It is like the kids who play implied odds in No-Limit without playing position.

Obviously, if the other factors for payout were identical, you'd choose the higher win percentage option. That's not the case here.

Would you participate in a drawing with a 90% chance to win $1000 and 10% chance to lose $9000? You only get to draw once.

If your income is small (like a coach's stock in public opinion), you may not recover from that $9000 hit.

It is actually a fair analogy because the odds that rojo gave were for winning the game, not going to overtime. I'm skeptical that they are the right odds, but if you assume they are true, then you play the best odds even though you don't experience the whole population of outcomes.

I wasn't exactly sure what malmo's point was, that was why I was asking.

rojos scenario is on the face of it wrong. If the other team knows the same probabilities. Clearly if you are down 3 with 2:32 left and playing for a field goal gives the other team a worse chance of winning, then you would play for a touchdown. I'm also wondering if the calculator he used had ties as an outcome.

douuuuuu wrote:

Obviously, if the other factors for payout were identical, you'd choose the higher win percentage option. That's not the case here.

Would you participate in a drawing with a 90% chance to win $1000 and 10% chance to lose $9000? You only get to draw once.

If your income is small (like a coach's stock in public opinion), you may not recover from that $9000 hit.

If I had to choose between that game and a 87% win $1000 and and a 13% lose $9000, then yes.

If rojo was right and the chance of winning was higher by missing the extra point, then you have to make and argument that the outcome of winning or losing by that strategy is worse than the strategy of making the extra point.

That happens twice in Cormac McCarthy's No Country for Old Men.

There's also a ton of other variables that I am sure rojo's statistical model isn't properly accounting for - wind speed, field conditions, number of timeouts remaining, home vs. away, etc.

Even accepting the 74% vs. 82% figures at face value is a big leap of faith.

dangle wrote:

You still get just one flip and your chances are 50/50

False, you'd get one flip and your chances would be 53%.

What people seem to be forgetting is up 3, the defense still gets a chance to make a stop. In fact, the Cowbos defense did just that. It stopped Detroit before Detroit got in field goal range.

I calculated the odds by assuming that Detroit would start the ball at the 20 after the kickoff. I just put Detroit down 3 or down 4, with 2:32 left and the ball on their own 20 into their win probabilty model.

I've written the owner of the site to see if he'll write an article confirming how I'm correct or if he'll make a statement one way or the other if he doesn't have time for an article.

dingle wrote:

It is actually a fair analogy because the odds that rojo gave were for winning the game, not going to overtime. I'm skeptical that they are the right odds, but if you assume they are true, then you play the best odds even though you don't experience the whole population of outcomes.

I wasn't exactly sure what malmo's point was, that was why I was asking.

Here are the potential outcomes in Rojo's scenario:

(1) Team intentionally misses the PAT; opposing team ties with a FG; team wins in overtime

(2) Team intentionally misses the PAT; opposing team ties with a FG; team loses in overtime

(3) Team makes PAT; opposing team scores a TD; opposing team wins in regulation.

(4) Team makes PAT; opposing fails to score a TD; team wins in regulation.

Rojo's point was that intentionally missing the PAT was statistically more likely to lead outcome #1 or #4.

What this statistic fails to appreciate is the situational risk in outcome #2. This is classic game theory, by the way.

rojo wrote:

[quote][

I calculated the odds by assuming that Detroit would start the ball at the 20 after the kickoff. I just put Detroit down 3 or down 4, with 2:32 left and the ball on their own 20 into their win probabilty model.

I've written the owner of the site to see if he'll write an article confirming how I'm correct or if he'll make a statement one way or the other if he doesn't have time for an article.

My question for you, rojo, is that if Detroit has a better chance of winning when down 4, why wouldn't they adopt the strategy of going for a touchdown? This is a zero sum game, after all. Obviously, the odds would change depending on the situation, but the model has to assume bad strategy by the losing team at some point if their odds of winning are lower when the score is closer.

From what I can read, rojo just typed the situation into a outcome calculator and the outcome calculator gave the odds of winning in that scenario. It doesn't exclude any outcome.

rather than miss the extra point on purpose, wouldn't it make the most sense to go for 2 so you'd be up 5 instead of 3 or 4??

based on the win probability calculator on the internet 5 its is better than 4 or 6 and slightly better or same as 3

so never miss the extra point on purpose, but go for 2

that all said statistically, the Lions had the win statistically for about 90% of the game and the Cowboys for about 5%, the refs had the other 5%, you play for the final score so I think kicking the extra point was smart given the Lions had Prater kicking indoors with a career long of 64 and a season best of 52 this year

Anyone else remember the Super Bowl a couple years ago when in the closing minutes, the Giants RB tried not to score a TD to burn up more clock, and the defense stood there and let him score because their best strategy was to get the ball back with time on the clock even though now behind. It was surreal. The defense and offense did opposite what they would normally do in such situations. The defense wanted to be scored on and the offense was trying not to score, too soon. And the RB who was trying not to score didn't expect this response from the defense and accidentally scored anyway.

WPC results for Detroit down -3 :

Win Probability: 0.18

Adjusted Win Prob.: 0.17

Expected Points: +0.34

First Down Prob: 0.67

TD Prob: 0.15

FG Prob: 0.08

WPC results for Detroit down -4:

Win Probability: 0.25

Adjusted Win Prob.: 0.24

Expected Points: +0.34

First Down Prob: 0.67

TD Prob: 0.15

FG Prob: 0.08

There is no difference in TD Prob and FG Prob under both scenarios. Difference in Win Prob due to added factor of Lion's Win Prob in OT in the -3 Scenario if they kick game tying FG in regulation. You kick the extra point.

go for 2 wrote:

based on the win probability calculator on the internet 5 its is better than 4 or 6 and slightly better or same as 3

This makes me believe that the win probability calculator includes ties in its not wins.

Sorry Rojo, but I think you should stick to running related analysis.

Has anyone else played with that calculator? Seems fishy. I was able to re-create rojo's numbers using these inputs:

2:32 left

4th period

Own 20

1st and 10

Pre-Game 0.50

Don't Know

Score Diff = -3

This yields win probability of 18% for Detroit. Here are the other probabilities:

0 - 61%

1 - 46%

2 - 39%

3 - 18%

4 - 26%

5 - 21%

6 - 26%

7 - 19%

These numbers make no sense. According to this calculator a team is more likely to win if it's down 7 rather down 3. That seems crazy because when you're down 7 you need a touchdown, while being down 3 means you can score a touchdown or settle for a field goal. But OK, just to play along with the argument from earlier this thread, maybe it means the team down 7 will go for it on every 4th down. Fine. But then why aren't the odds basically the same for being down 4 or 5 or 6 or 7? In all 4 scenarios your fourth-behavior should be identical, so why do the odds vary so much?

Without knowing where these numbers come from, how can we place any weight in them?

McTed wrote:

Sorry Rojo, but I think you should stick to running related analysis.

No. He tried to do a probability analysis earlier this year on a running event (I forget which) that was demonstrably wrong.

malmo wrote:

rojo wrote:

If I ever get rich enough to own an NFL team, you can damn bet I'll have stat boys running the time outs, 4th down decisions.

That's already being done. Welcome to 1990.

Stats guys have shown coaches don't go for it enough. Look at it yesterday 4th and 1 in the 4th on about the 40 and the Lions punted. I'm almost certain a computer would say to go for it.

dingle wrote:

rojos scenario is on the face of it wrong. If the other team knows the same probabilities. Clearly if you are down 3 with 2:32 left and playing for a field goal gives the other team a worse chance of winning, then you would play for a touchdown. I'm also wondering if the calculator he used had ties as an outcome.

The coaches have shown time and time again they are risk averse. They are likely to kick the field goal and take their chances in Overtime. Look at Detroit yesterday on 4th and 1, very risk averse.

wejo wrote:

malmo wrote:

That's already being done. Welcome to 1990.

Stats guys have shown coaches don't go for it enough. Look at it yesterday 4th and 1 in the 4th on about the 40 and the Lions punted. I'm almost certain a computer would say to go for it.

dingle wrote:

rojos scenario is on the face of it wrong. If the other team knows the same probabilities. Clearly if you are down 3 with 2:32 left and playing for a field goal gives the other team a worse chance of winning, then you would play for a touchdown. I'm also wondering if the calculator he used had ties as an outcome.

The coaches have shown time and time again they are risk averse. They are likely to kick the field goal and take their chances in Overtime. Look at Detroit yesterday on 4th and 1, very risk averse.

Huge difference between being risk adverse and not going for a 4th down play and intentionally missing an extra point to put the other team within a field goal instead of a touchdown.