Two point conversions?

[SLU MD]

Watching the Pats vs SD game today they showed that the NFL conversion rate is 51.2% on 2 pt conversions. So if you went for two every time, wouldnt you come out ahead in the long run?

fyi, i dont agree with the statement, but a friend and I were having this discussion. anyone with some cold hard facts that can solve our argument?

 

[her209]

You need a larger sample size.

 

[IceBergSLiM]

some games need to played more conservatively than others.

 

[MartyMcFly3]

I'd rather go for the 1 point and get it 95% of the time than go for a 2 pt. conversion and miss it 50% of the time (practically)

 

[SVT Cobra]

If you went everytime Defenses would defend against it better.

Not to mention it is the same as a 4th and ~2, although the team is coming off of the momentum of a TD.

 

[CellarDoor]

Wasn't there some Princeton professor (or something) that came up with the theory that it's always better to go for it on 4th down, regardless of field position? I would guess he'd agree that it's always better to go for 2. I remember a game a few years back (49ers at Rams) where the Rams' field goal kicker was injured so they went for 2 every time. They made almost all of them I think.

Edit: He said on average you should go for it on 4th down in any short yardage situation, and in many long yardage situations. He's from Berkeley, not Princeton.

http://espn.go.com/nfl/columns/garber_greg/1453717.html

 

[Alkaline5]

Based on that statisitic if a coach knows beforehand that his team is going to score an even number of touchdowns, then he may as well go for it every time. The points will even out in the end.

But if the team is only going to score an odd number of TDs, the safe thing to do is take the PAT every time. That way you're virtually gauranteed of scoring after every TD.

So since no one can ever know how many TDs a team will end up scoring, the best odds are to take the PAT every time until you absolutely need 2 points.

 

[SVT Cobra]

What all these equations do not take into account is momentum. Sorry.

 

[SLU MD]

so no real answer?

 

[ElFenix]

Originally posted by: CellarDoor
Wasn't there some Princeton professor (or something) that came up with the theory that it's always better to go for it on 4th down, regardless of field position? I would guess he'd agree that it's always better to go for 2. I remember a game a few years back (49ers at Rams) where the Rams' field goal kicker was injured so they went for 2 every time. They made almost all of them I think.

Edit: He said on average you should go for it on 4th down in any short yardage situation, and in many long yardage situations. He's from Berkeley, not Princeton.

http://espn.go.com/nfl/columns/garber_greg/1453717.html

a mathematician proved that you'd win more baseball games if your best hitter was your leadoff hitter, followed by second best, third, and so forth until the worst hitter. and yet, every manager in baseball has speedy guys and then power guys, even if the power guys are better in every other stat.

 

[meltdown75]

7 pts is better than 6. If you go for it and don't get it, the other team can tie you with 2 FGs or just take the lead with a TD and a regular PAT.

To me, 2-pt conversions are only useful when you can come within 7 or 3 pts of the opposing team, to bring the game within 1 score. Also, say for example that your TD makes the score 21-20 for your team. It makes more sense to go for 2 at that point to give you a 3-pt lead instead of a 2-pt lead, especially if the game is late in the 4th and you might be able to hold the other team to a FG.

 

[Garet Jax]

Originally posted by: MartyMcFly3
I'd rather go for the 1 point and get it 95% of the time than go for a 2 pt. conversion and miss it 50% of the time (practically)

But the expected value is higher in the two point conversion case.

1X.95 vs 2X.5

 

[rsd]

Originally posted by: SLU MD
so no real answer?

Mathematically it can make sense sure, but you aren't accounting for RISK.

You calculate expected payoff etc, but there is a bigger RISK involved between the two. Unless it is needed (and there are certain times it is), you are better off going for 1.

 

[Garet Jax]

Originally posted by: rsd

Originally posted by: SLU MD
so no real answer?

Mathematically it can make sense sure, but you aren't accounting for RISK.

You calculate expected payoff etc, but there is a bigger RISK involved between the two. Unless it is needed (and there are certain times it is), you are better off going for 1.

There really is no risk (other than not getting the points). If the other team intercepts the ball, they can't return it 98 yards for a 2 point conversion of their own. The play is dead. Assuming the conversion ratio is truly 50%, then you will out point PAT (since PATs are porbably 99% converted).

The problem here is that I don't believe the conversion ratio would be 50% if all teams started doing it. I think it would drop close to 35-40%.

 

[thelanx]

While it is true that 2 point conversions are relatively rare and thus unexpected, they often come at a time when a team needs it or it is strategically advisable (ie leading by 1 after a touchdown). Thus many 2 point conversion attempts occur in situations where you expect such an attempt, which reduces the unexpected nature of 2 point conversions. However, I still agree that if people tried more 2 point conversions, then the success rate would drop. I'm curious though about whether a team would benefit from randomly attempting a few more 2 point conversion attempts in their games.

 

[Alkaline5]

Originally posted by: Garet Jax
There really is no risk (other than not getting the points). If the other team intercepts the ball, they can't return it 98 yards for a 2 point conversion of their own. The play is dead.

Unless the rules of pro ball are different from college ball, this isn't true. It's entirely possible that the other team returns the ball for 2 pts of their own. The chances of that would be small, but it's still something to consider.

 

[IAteYourMother]

Originally posted by: Alkaline5

Originally posted by: Garet Jax
There really is no risk (other than not getting the points). If the other team intercepts the ball, they can't return it 98 yards for a 2 point conversion of their own. The play is dead.

Unless the rules of pro ball are different from college ball, this isn't true. It's entirely possible that the other team returns the ball for 2 pts of their own. The chances of that would be small, but it's still something to consider.

if that happens, the other team may as well forfeit

 

[LS20]

i tried going for 4th down everytime on playstation. doesnt work that well

 

[JulesMaximus]

Originally posted by: Garet Jax

Originally posted by: rsd

Originally posted by: SLU MD
so no real answer?

Mathematically it can make sense sure, but you aren't accounting for RISK.

You calculate expected payoff etc, but there is a bigger RISK involved between the two. Unless it is needed (and there are certain times it is), you are better off going for 1.

There really is no risk (other than not getting the points). If the other team intercepts the ball, they can't return it 98 yards for a 2 point conversion of their own. The play is dead. Assuming the conversion ratio is truly 50%, then you will out point PAT (since PATs are porbably 99% converted).

The problem here is that I don't believe the conversion ratio would be 50% if all teams started doing it. I think it would drop close to 35-40%.

Not to mention that you might be successful 50% of the time but expecting to do that 3 or 4 times in one game is unrealistic. You might get all of them but then again, you might get none of them. A field goal is a much safer bet.

Generally, a team will only go for two if it will bring them within one score of an opposing team or if it will give them a lead that would require the other team to score either a field goal or touchdown to tie the game.

 

[Alphathree33]

I had no idea football was this complicated...

 

[LS20]

Originally posted by: Garet Jax

But the expected value is higher in the two point conversion case.

1X.95 vs 2X.5

but you dont score that many times in a game for the odds to "normalize"

if you score once, and make 2pt, the other team will try a 2pt to tie you
if you score once, and miss the 2pt, you are fcked

there arent enough opportunities for the odds to work itself back...

 

[Viper GTS]

Expected value is only really useful if the outcome of any one event doesn't matter. That implies an essentially infinite number of repetitions.

Would you score more points if you had a positive expected value? Maybe.

Would you win more games? Doubtful.

Viper GTS

 

[spidey07]

I'm pretty sure the coaches have charts that tell them if they should go for 2 points or not depending on the score.

For example, let's say you only score one touchdown or an odd numbers of ones. Then you're screwed by going for 2.

 

[Lonyo]

 

[Garet Jax]

Originally posted by: Lonyo

You reserving the 37th post of this thread? Any significance?