The Two Point Conversion When Eight Down

[PairTheBoard]

So the real average 1 point conversion rate is more like 96%, throwing the decision even more in favor of Going For Two right away.

Also, none of this factors in the case of slightly more time on the clock giving a small chance of actually getting two more possessions in which 2 field goals could conceivably be kicked. That's got to weight the decision even more toward Going For Two.

PairTheBoard

[NickMPK]

I think the image thing may be more important than people are giving it credit for.

The fact of the matter is, the "go-for-two" play is so unorthodox that the coach always looks bad when his team misses, even when they come back to tie and win in overtime. People will question why they put themselves in such a desperate situation that they had to go for two a second time.

However, the coach will only look really good if the team makes the first conversion and later scores the second touchdown to win the game. If the team never scores the second touchdown (or the other team scored again to make it irrelevant), the play will just look like an oddity that had no real effect on a losing effort, and will be forgotten. Since it is more likely than now that the trailing team never actually gets into a position to win the game, the coach looks bad much more often than he looks good even if his team makes the conversion more than 50% of the time.

[SHOwnsYou]

Why is the calculation for successful 2 point conversions done on an over all basis as opposed to on a team vs team basis?

For example, just because the Cowboys average a 48% success rate in 2 point conversions over all, does not mean they will have a 48% success rate versus the Colts, it may be drastically lower. I think it is incorrect to not take into account the average on a per team basis.

Furthermore, what if they cowboys have only attempted a 2 point conversion against the colts 3 times with their current roster giving them a painfully small sample size.

I think it would be easy to dismiss by saying that in the long run they will be at 48% to hit their 2 point conversions, but I say that in the long run they are getting 2 point conversions more against teams they were already favored against.

Also - what about variance being a biased estimater in regards to sample vs total size? That may come into play also.

[plj8624]

David, the math is so clear on this point it's indisputable. 95% of coaches (and owners, and fans) don't know the math. I'm not sure what you expect of the average population, most of whom are mathematically illiterate.

I have thought for a while that coaches don't go for it on 4th down enough. Some of the best coaches (Belichick, Parcells) take more risks than others.

What do think of the frequency of making tricky/risky special teams plays? I'm thinking of fake punts, fake field goals, and surprise onside kicks. We probably agree that coaches don't take these risks often enough; probably for the same reason that they don't understand the risk/reward profile. I watch a fair amount of pro football and I don't think I have seen ANY this season.

[stinkypete]

[ QUOTE ]
Furthermore, what if they cowboys have only attempted a 2 point conversion against the colts 3 times with their current roster giving them a painfully small sample size.


[/ QUOTE ]

it's extremely rare to even have that sample of 3 attempts. you have to make an educated guess and GAMBOOOOL

[Wheel Gunner]

Since I don't think anyone has posted the complete solution to this example using David's assumptions I thought I would take the liberty. You need to multiply the probabilities for each instance. Fellas, this is why God created spreadsheets.

Go for Two the first TE

2 on the first, 1 on the second, win .4116 W
2 on the first, miss the kick, win in OT .0042 W
same as above lose in OT .0042 L

0 on the first, 2 on the second win in OT .1218 W
same as above, lose in OT .1218 L
O on the first, O on the second, lose .3364 L

total win .5376
total lose .4624

A similar calculation for going for the tie and only going for two if you miss the first kick gives

Win: .4844
Lose .5156

The math for kicking the extra point on the first TD and going for 2 on the second to win or tie gives:

Win: .4158
Lose: .5842

[dabluebery]

[ QUOTE ]
Since I don't think anyone has posted the complete solution to this example using David's assumptions I thought I would take the liberty. You need to multiply the probabilities for each instance. Fellas, this is why God created spreadsheets.

Go for Two the first TE

2 on the first, 1 on the second, win .4116 W
2 on the first, miss the kick, win in OT .0042 W
same as above lose in OT .0042 L

0 on the first, 2 on the second win in OT .1218 W
same as above, lose in OT .1218 L
O on the first, O on the second, lose .3364 L

total win .5376
total lose .4624

A similar calculation for going for the tie and only going for two if you miss the first kick gives

Win: .4844
Lose .5156

The math for kicking the extra point on the first TD and going for 2 on the second to win or tie gives:

Win: .4158
Lose: .5842

[/ QUOTE ]

I agree with your math, but can anyone tie this in to the real world chances of winning a football game?

The 5% edge in going for two the first time, over kicking the PAT, is because you're holding constant the chances that the opponent won't score AND that you will score a second touchdown. Those factors decrease the chances of winning by an unknown margin. I know nothing about football, maybe someone who does could elaborate.

Your math shows an additional expection of .07 wins, roughly one extra win in fourteen games holding those other factors constant. If you add in the additional factors, maybe it only adds an extra win every 30 or 40 games. (Guessing)

It's possible that an extra win over a time trial of a few seasons might not be worth it in the long run if a coach could likely be fired as the result of this going wrong, against "conventional" wisdom. Sort of like not risking your tournament life on close edges, etc.

Rob

[andyfox]

Sports is a very conservative milieu. Much of the baseball "book," for example, was "written," by John McGraw in an era when runs weren't scored in nearly the same numbers that they are now and almost all of it, therefore, is wrong for today's game. When Bill James first presented the math to a major league general manager, his comment was the paperwork would make a good doorstop. He didn't understand it and didn't want to understand it. So I don't doubt it when you say, as you did in another thread, that up until a few years ago (when the sabermaticians started to have some influence) you would have been the greatest baseball manager ever. Hell, I think I would have been it.

On the psychological deflation issue, I think there might be the opposite effect if the team is successful with the first two point conversion. The players might say to themselves, "All right, we're playing to WIN, not for a tie! This guy [the coach] has COJONES!" (Since they probably won't know the math either, they'd see it as a ballsy decision, rather than a logical one.) That might well inspire them to score the second touchdown more often than they would if they went for one.

[RacersEdge]

Is the 42% backed with data? I can see it at mid-field, but it seems high for the 2 yard line.

[odellthurman]

[ QUOTE ]
Sports is a very conservative milieu. Much of the baseball "book," for example, was "written," by John McGraw in an era when runs weren't scored in nearly the same numbers that they are now and almost all of it, therefore, is wrong for today's game. When Bill James first presented the math to a major league general manager, his comment was the paperwork would make a good doorstop. He didn't understand it and didn't want to understand it. So I don't doubt it when you say, as you did in another thread, that up until a few years ago (when the sabermaticians started to have some influence) you would have been the greatest baseball manager ever. Hell, I think I would have been it.

[/ QUOTE ]

Being a great baseball manager is about a lot more than just lineup decisions and decisions made during games. And being a great football coach is bout about a lot more than deciding when to punt, when to kick, and when to go for two.