Going for two each time Theory - Page 2

#1 12-05-2005, 06:34 PM

I think this can be put in the same category as a segment I saw on ESPN a month ago.

Some mathematician calculated the "odds" of going for it on fourth down and said that most teams should go for it way more often than they usually do. When they presented this evidence to some head coaches, their reactions were similar:

"Sounds like a good theory, but has this math guy ever had his name on the front page of the sports section because he decided to go for a fourth down early, missed it, gave up a field goal, and lost by two points?"

ScottieK

Mason Hellmuth · #2 12-05-2005, 06:35 PM

[ QUOTE ]
Some mathematician

[/ QUOTE ]
Was his name David Sklansky?

Also, hot damn the mods are quick today.

Buckmulligan · #3 12-05-2005, 06:37 PM

[ QUOTE ]
as soon as you drop the stat to 49% it makes sense why they wouldn't try that. you're further from a lock that you'll make two half the time, than you are from a lock that you'll make one 100% of the time.

[/ QUOTE ]

You don't think the colts could average 1.1 or 1.2 points per two point try?

jasonHoldEm · #4 12-05-2005, 06:37 PM

It doesn't seem like it would matter very much...what I mean is for those teams who have a highly capable offense that can score 2pt conversions more than the league average is the extra point or two they'd pick up really going to matter? It seems like most of these teams win by a larger margin than that most of the time anyways.

Hope that makes sense.

J

Buckmulligan · #5 12-05-2005, 06:41 PM

[ QUOTE ]
It doesn't seem like it would matter very much...what I mean is for those teams who have a highly capable offense that can score 2pt conversions more than the league average is the extra point or two they'd pick up really going to matter? It seems like most of these teams win by a larger margin than that most of the time anyways.

[/ QUOTE ]

Success is all about thinking at the margin. I'm sure that if you could convince any NFL owner that going for two each time would get them two more points in a easy, all kickers would be getting their walking papers.

bobman0330 · #6 12-05-2005, 06:40 PM

Two reasons:
1. 6 points only ties 2 FGs, which is a common scoring increment. As you've established, the 2pt is roughly equal to the PAT, but loses big against 2 FGs.
2. Assume that each team scores 1 TD. If the first team goes for 2, it will lose a lot more than 50% of the time. 50% of the time it misses and loses. 50% of the time it makes. Then, the other team knows to go for 2, and ties the game half the time. Obviously, the real life situations aren't so formalistic, but the longer there is remaining in the game, the more this effect costs the converting team.

Buckmulligan · #7 12-05-2005, 06:44 PM

[ QUOTE ]
Two reasons:
1. 6 points only ties 2 FGs, which is a common scoring increment. As you've established, the 2pt is roughly equal to the PAT, but loses big against 2 FGs.
2. Assume that each team scores 1 TD. If the first team goes for 2, it will lose a lot more than 50% of the time. 50% of the time it misses and loses. 50% of the time it makes. Then, the other team knows to go for 2, and ties the game half the time. Obviously, the real life situations aren't so formalistic, but the longer there is remaining in the game, the more this effect costs the converting team.

[/ QUOTE ]

It sounds like you have something worth-while to say but I'm not really following. Could you rearticulate?

bobman0330 · #8 12-06-2005, 06:17 PM

[ QUOTE ]
[ QUOTE ]
Two reasons:
1. 6 points only ties 2 FGs, which is a common scoring increment. As you've established, the 2pt is roughly equal to the PAT, but loses big against 2 FGs.
2. Assume that each team scores 1 TD. If the first team goes for 2, it will lose a lot more than 50% of the time. 50% of the time it misses and loses. 50% of the time it makes. Then, the other team knows to go for 2, and ties the game half the time. Obviously, the real life situations aren't so formalistic, but the longer there is remaining in the game, the more this effect costs the converting team.

[/ QUOTE ]

It sounds like you have something worth-while to say but I'm not really following. Could you rearticulate?

[/ QUOTE ]

1. Points don't have linear utility. Increasing your point total by 8 is better than increasing it by 7, because you'll beat a TD+PAT rather than tying it. But increasing your point total by 7 comparatively even better than increasing it by 6, because you'll tie a TD+PAT and beat 2 FGs. Even though both differences are 1 point, the downside risk of a 2pt conv. is higher than its upside.

2. Going for 2 early in a game gives the other team informational advantages. Whenever they have to make a similar decision, they can act knowing whether or not you made your conversion. If you did not, they can play conservatively to win. If you did make it, they can gamble and sometimes tie you. The benefit is substantially decreased because the other team can adjust if you succeed.

MyTurn2Raise · #9 12-06-2005, 07:44 PM

[ QUOTE ]
1. Points don't have linear utility. Increasing your point total by 8 is better than increasing it by 7, because you'll beat a TD+PAT rather than tying it. But increasing your point total by 7 comparatively even better than increasing it by 6, because you'll tie a TD+PAT and beat 2 FGs. Even though both differences are 1 point, the downside risk of a 2pt conv. is higher than its upside.

2. Going for 2 early in a game gives the other team informational advantages. Whenever they have to make a similar decision, they can act knowing whether or not you made your conversion. If you did not, they can play conservatively to win. If you did make it, they can gamble and sometimes tie you. The benefit is substantially decreased because the other team can adjust if you succeed.

[/ QUOTE ]

right on...I almost posted something like this, but my economics is quite rusty. I think I did a brain flush the second I dropped out of my PhD work.

I think much of these ideas do not account for the fact that the plays are not held in isolation. A football game is a dynamic system with each play inter-related to the next. Pushing to maximize the output of any one play, or "event," can have very detrimental consequences towards maximizing the output of the entire game, or "system."

While I think that football coaches are too conservative, I do not see the fault in going for two not frequent enough. I see coaches go for two far too frequently. I also think they should go for it on fourth more often, but not nearly as often as suggested in the paper by the Professor out at Berkeley.

What I have seen in football is the total breakdown of risk-taking on the offensive side of the ball. The mid to long range passing game is becoming extinct. I think everyone realized how important turnovers are, but over-reacted. An interception 35 yards downfield is different than a fumble at the line. Oh well...I have a whole notebook full of observations of things I'd change if I were a coach. Maybe after I make millions at poker, I'll work my way up from the low levels of coaching.

MyTurn2Raise · #10 12-05-2005, 06:45 PM

awful, awful idea to go for two every time

These are not independent trials. The effect of going for two cannot be viewed in isolation. It directly changes the way the rest of the game is played.

missing does not cost just 1 point. There are hidden points lost later based on game situations. A team might have to later go for two to make up for a missed one earlier. It might become advantageous for an opponent to later go for two, and, if they make it, the missed opportunity cost 2 points then.

As the previous poster pointed out...making it might cause the opponent to go for two later where they can make it up.

EDIT: This is an example of what happens