Two Plus Two For Two (Points)

[PairTheBoard]

Pretty Sure DS meant "down by 8" after the touchdown has been scored for 6 points reducing the 14 point deficit to 8. So at the moment the decision is made on which conversion attempt to make the team is still down by 8.

PairTheBoard

[optimus_prime]

Anyone have a link to the math behind this argument? (just curious)

[PairTheBoard]

[ QUOTE ]
Anyone have a link to the math behind this argument? (just curious)

[/ QUOTE ]

DS has Another Thread going on the topic where the math is discussed.

This Link gave the most complete mathematical analysis of it.

PairTheBoard

[Daliman]

[ QUOTE ]
The debate is about whether to go for 2 when down by 8 after scoring a touchdown (i.e. you were down by 14 before scoring). Assuming that you'll score again later, is it better to kick the XP twice for a near guarantee of overtime or to try for 2 after the first score? If you make it, you'll win in regulation. If you miss it, you go for 2 again on the second score and still have a chance to force OT. Based on the assumptions that 2-point conversions succeed at a constant rate of 42% and each team is 50/50 to win in OT, going for two is mathematically correct.

[/ QUOTE ]
Ah. ok. I'd say that the deflating factor of missing the 2 pt conversion is the main reason why people don't go for the 2 in these situations. DS being a cyborg would not fully understand or appreciate this.

[Bjorn]

Personally I'd much rather see an argument and a campaign for correct 4th down strategy. The situation mentioned here simply doesn't come up often enough to be of much use and even when it does it's not that huge a difference maker.

Seen over a season routinly punting on 4th and one probably is about a hundred times worse an error.

In both cases current coaching "wisdom" is mathematically wrong but only one (the 4th down thing) i a big leak, simply because of the difference in frequency.

I also think the 4th down thing is a much easier "sell", many football fans (even of the non-math variety) allready want their teams to go for it more on 4th down and even even more hope that their opponents won't go for it on 4th and short.

/Bjorn

[shemp]

Is there any disparity in the success rates of leading and trailing teams (in the data)? Also, is the 40-45% likelihood of success roughly similar for all teams (conversely, 55-60% defensive failure rate)?

It seems the model in use here is crude enough that it should be one ingredient to the decision, but far from the only one.

[George Rice]

I think that some coaches already do this a lot, especially when just out of field goal range. Bill Parcels comes to mind. Of course your success at running the ball thusfar has an impact on that decision.

But this can be taken too far. Yesterday the Giants went for it 4th and 1 at about the 24 yard line. The game was tied at 7 with 1:30 left in the first half. This would have been a probable 41 yard field goal for the lead going into the half. They lost 3 yards and gave Dallas the ball at the 27 yard line with 1:13 to go (Dallas would up kicking a field goal before half).

The Giants gave up a good chance of scoring the go-ahead score before the half. And even if they made the first down, there's no gurantee they'll score a touchdown, although they will run some more time off the clock.

But otherwise, I think the main factors are your kicker's range and the score. Most coaches realize there's no point in trying a low prob FG and giving the other team the ball with good field position. But many don't give making the first down enough weight compared with punting with what frequently amounts to a 10 or 15 yard net gain (if there's a touch-back or decent run-back).

[Bjorn]

Personally I liked the Giants decicion to go for it there because of the time management situation. If you go for the field goal Dallas will usually get at least a field goal themselves. Going for it here and succeding gives you a decent shot at 14-7 halftime or at the very least 10-7, kicking the field goal will usually result in a 10-10 tie at best.

Essentially its not just about your gain of EV (in points) but also the fact that you potentionally get to deny Dallas a possesion by converting.

Give or take a minute (maybe even half a minute) of clock time and my oppinion is often different.

I do however strongly disagree with the playcalling, runing an outside run play straight at perhaps the defence's best player on a critical 4th down play doesn't seem very smart to me.

/Bjorn

[jah7_fsu1]

[ QUOTE ]
Seen over a season routinly punting on 4th and one probably is about a hundred times worse an error.

[/ QUOTE ]

Care to expand on this? I think their are way to many factors involved to make such a statement. The type of game, field position, etc, etc, etc. Or were you approaching it from a mathmatical standpoint?

[Piemaster]

The problem is one of utility maximisation. If coaches were robots programmed to help their team win at all costs, then they would likely go for 2 points when trailing by 8. In real life, coaches want their team to win, but they want other things too.

- They want to be seen to be seen to be a 'good coach'
- They want to be able to sleep soundly at night.
- They want nice things written about them in the media
- They want to maximise their career earning potential
- They want to make the Hall of Fame

Of course, not all of these things are going though their mind when they are making a crucial decision on the field, but they are in the back of their mind and they 'program' them to think in certain ways. They will tend to act in a risk averse way in a close decision, because a risk averse decision can rarely be proved wrong.

For example, you are 8 behind and decide to go for the 2 pointer. You miss it, then you score the second TD and miss the 2 pointer again. You are now 'that dumb coach who got greedy rather taking the easy point'. People start to question your abilities as a coach (especially if you were only mediocre in the first place). You star in Vic Carruci's Bad Ideas column. Your start to foster a reputation for making bad decisions under pressure, because fans and the media are extremely results orientated.

Now let's say you kick the points instead. Cindy the girly kicker pops one over, then you score another TD and the kicker scores again. In overtime you lose the toss, and the opposition dink their way down the field and score a field goal. You've lost the game again, but nobody is going to question your decision to kick the PAT earlier in the game. First of all that was long ago and nobody will really think about it. Secondly the unknown factor prevents people from really identifying it as a bad move (meh, we would probably have missed anyway. At least we got a shot at winning in overtime). It would take a very astute analyst (almost an oxymoron in football) to identify the coaches error in not going for two.

So basically kicking the PAT is slightly -EV for the team, but maybe slightly +EV for the coach. Hence I would be all in favour of this campaign if it would help analysts and fans to see sense and hence help coaches make the right decisions. Unfortunately I live in the UK and so nobody would have a clue what my bumper sticker was talking about.