Should the Pats be going for 2-point Conversions?

QuantumMechanic · Sep 17, 2015

Fencer said:
But what I quoted above doesn't exactly say that. If I want to double my money, dividing it in 10 parts and betting each part once is a horrendous strategy, if I stop afterwards, for reasons much simpler and more drastic than what you said -- after all, if I lose ANY of the 10 bets, I fail.

True. I muddled it up and could have been better stated like this:

The bad team is like a gambler who has to go to the casino 16 times and each time has to make $100 worth of wagers on betting evens on roulette. How he makes that $100 of wagers is up to him. If he walks out the casino with more than $100 it's a win, if less than $100 it's a loss.

The casino has a 5.3% edge on double-0 wheels. So on average the gambler will end up walking away with $94.70.

If the gambler takes a low-variance strategy (like 100 $1 bets) his outcomes will tightly group around ending up with $94.70 and he will lose most of his 16 "games".

If he takes a high-variance strategy (a single $100 bet) he has a almost 50% chance of ending up with $200 (and of course a slightly larger than 50% change of ending up with nothing), so he will probably finish close to 8-8 for the year.

So if you have the advantage you want to use a low-variance strategy to make it more likely you finish near your superior average (and the more superior you are the lower variance you want). And if you have the disadvantage you want to use a high-variance strategy to make it more likely you finish above your inferior average (and the more inferior you are the higher variance you want).

aluminum seats · Sep 17, 2015

Interesting topic. I agree w/the poster who pointed out that constant 2-pt attempts expose bushels of red zone plays, so there's that.

I wonder too about the psychological aspect...missing extra points is such a huge let down after the euphoria of a touchdown, going for two every time means you'll have that let down pretty much half the time you score. That would sorta suck.

BTTA · Sep 17, 2015

Generally agree it isn't in the Pats' best interest to change up here, but have been wondering about the Colts game, if as usual they can't stop the run.

PittPatriot · Sep 17, 2015

Oh, F--- it. If we want to tell the rest of the league to go to hell, I say we go for 2 every time. We have the GOAT QB, a great OC that will come up with different looks to get the 2 yards needed, a likely GOAT TE, a great RB by committee...and JE11. We have so many options down there.
Plus, it puts sooo much more pressure on the opposition.

Edit: Perhaps we take XP while the weather is good. Those 33yd field goals with wind, rain, and snow can't be taken for granted.

SlowGettingUp · Sep 17, 2015

QuantumMechanic said:
That's why lesser teams try to slow the game down and limit the number of possessions.

That makes a lot of sense. It also validates the "keep Brady off the field" strategy which I never really understood before given both teams have (roughly) the same number of possessions.

Be interesting to look at statistics and see if highly-favored teams tended to come up short more often in low-scoring games. If that is in fact the case there might be some interesting betting strategies involving a combined under/over bet and a bet on the winner (although I kind of assume there is no free lunch to be had).

ctpatsfan77 · Sep 17, 2015

SlowGettingUp said:
That makes a lot of sense. It also validates the "keep Brady off the field" strategy which I never really understood before given both teams have (roughly) the same number of possessions.

Be interesting to look at statistics and see if highly-favored teams tended to come up short more often in low-scoring games. If that is in fact the case there might be some interesting betting strategies involving a combined under/over bet and a bet on the winner (although I kind of assume there is no free lunch to be had).

Usually when there's money to be had, it's in arbitrage based on different books having different odds/lines on the same bet (e.g., one book is giving –6.5 while another is giving –4.5).

QuantumMechanic · Sep 17, 2015

SlowGettingUp said:
That makes a lot of sense. It also validates the "keep Brady off the field" strategy which I never really understood before given both teams have (roughly) the same number of possessions.

Be interesting to look at statistics and see if highly-favored teams tended to come up short more often in low-scoring games. If that is in fact the case there might be some interesting betting strategies involving a combined under/over bet and a bet on the winner (although I kind of assume there is no free lunch to be had).

Right. Let's say Team A has a 50% chance to score on a given possession and Team B has a 40% chance.

If each team only has a single possession, then it's a 0-0 tie 30% of the time, Team B wins 1-0 20% of the time, Team A wins 1-0 30% of the time, and it's a 1-1 tie 20% of the time. So Team A wins 30% of the team and Team B wins 20% and it's a tie 50% of the time.

But if each team has two possessions then its:

Tie 0-0: 9%
A wins 1-0: 18%
A wins 2-0: 9%
B wins 1-0: 12%
Tie 1-1: 24%
A wins 2-1: 12%
B wins 2-0: 4%
B wins 2-1: 8%
Tie 2-2 4%

which means that A wins 39% of the time, B wins 24% of the time, and it's a tie 37% of the time.

Both teams had their chance of winning increase (as the chance of a tie drops) but A's chance of winning increased by more than B's chance of winning did.

Or if you look at it from the perspective of an inferior team that is trying to avoid a loss, a 1-possession game gives it a 70% chance to not lose but a 2-possession game only gives it a 61% chance to not lose. And as you add more possessions, B's chance of not losing will continue to shrink.

Obviously this is a vast oversimplification, but it does give some insight into why inferior teams want to slow the game down.

TruthSeeker · Sep 17, 2015

Interestingly enough, it can often be right to go for 2 when down by 14 or higher multiples of 7 later in the game. If you make it, then kicking the extra point next time gives you a 1 point edge. If you don't convert then you have close to a 50% chance to get back to even next TD.

Assuming 48% for 2 point conversion and 95% for 1 point conversion Values approximate.

Overall
- 45% chance of 1 point edge (48% x 95%)
- 23% chance of tie
- 32% chance of 2 point deficit
Total win % = 61%

This compares with
- 90% chance of tie
- 9+% chance of 1 point deficit
- <1% chance of 2 point deficit
Total win % = 45%

Worth thinking about.

pdangle · Sep 17, 2015

All good points. And exactly the reason they probably moved it further back. As discussed here (and just briefly) it adds a complete new element of strategy to the game. But it does take away a bathroom break.

NormZauchin · Sep 17, 2015

QuantumMechanic said:
Good teams playing lesser teams want to pick a low-variance strategy since that increases the odds their superiority will win out. That's the same reason why a good team wants there to be lots of possessions -- the more possessions, the more chances for their superiority to win out.

Conversely, bad teams want high-variance strategies because it increases the odds of an outlier event happening -- which is what they need to be able to win. That's why lesser teams try to slow the game down and limit the number of possessions.

It's analogous to going to a casino and playing a game where the odds are close to even (say betting evens on roulette) and trying to turn $100 into $200. If you bet the $100 on a single bet you have close to a 50/50 chance of achieving your goal. If you make 10 $10 bets you are virtually guaranteed to fail to make your goal since with repeated trials the house edge will grind you down. And 100 $1 bets would be even worse.

Or think of it in terms of bell curves (admittedly not directly applicable, but gives a visualization). Say NE averages 28 points a game and team X averages 17 points a game. If both teams play very low-variance strategies (i.e. the bell curves are very narrow around the averages) they will finish relatively close to their average performances. Obviously good for NE and bad for team X. But if X plays a high-variance strategy (average stays the same but the curve gets a lot "fatter") it increases the odds it can finish above its average (and of course increases the odds it can finish below its average) and therefore gives itself a better chance to win.

Many years ago I read an article on the difference between the number of tennis tournaments that the top tennis players win and the number of golf tournaments that the top golfers win. It illustrates a point that you make in your comment. The winner of a golf tournament strikes the ball roughly 280 times. The winner of a tennis tournament will strike the ball that many times in one set. Given many more chances at hitting the ball, the best tennis player is more likely to win than the best golfer. (It probably also hinders a golfer that he must defeat the entire field, not a limited number of opponents, as in match play.)

ctpatsfan77 · Sep 18, 2015

pdangle said:
All good points. And exactly the reason they probably moved it further back. As discussed here (and just briefly) it adds a complete new element of strategy to the game. But it does take away a bathroom break.

Belichick pushed for moving it back on the grounds that it was essentially automatic, and therefore essentially useless.

Should the Pats be going for 2-point Conversions?

QuantumMechanic

Burn it all down!

aluminum seats

Pro Bowl Player

BTTA

He/Him

PittPatriot

PatsFans.com Supporter

SlowGettingUp

2nd Team Getting Their First Start

ctpatsfan77

PatsFans.com Supporter

QuantumMechanic

Burn it all down!

TruthSeeker

PatsFans.com Supporter

pdangle

Rotational Player and Threatening Starter's Job

NormZauchin

PatsFans.com Supporter

ctpatsfan77

PatsFans.com Supporter

Similar threads