Is it better to receive or defer with the new OT rules (1 Viewer)

[Hipple Long Ware & Peete]

You always receive, if you don't then you might not even get the ball. I don't care how good your D is, you take the ball.

As last year's Denver/Pitt game showed, all it takes is 1 play.

As this year's KC/Pitt game showed, all it takes is 1 play.I don't have access to enough statistical data to really provide the analysis. The calculation is this combination of probabilities:

A1) Probability of scoring TD on opening drive

A2) Probability of scoring FG on opening drive and stopping opponent on 4 downs

A3) Probability of winning when game is tied after each team's first drive, with possession of the ball

vs.

B1) Probability of holding opponent to FG and scoring a TD

B2) Probability of holding opponent to no score and scoring a FG or TD (last night's scenario)

B3) Probability of winning when game is tied after each team's first drive, without possession of the ball

A1 is larger than B1, A3 is larger than B3, and B2 is larger than A2. My intuition is that the difference between B2 and A2 is larger than the combination of the differences between A1/B1 and A3/B3, but without the numbers it's just speculation.

I'll just note that the knee-jerk "always receive" reaction is like Karl Rove's reactions on election night. There is a correct answer here: it might be that kicking is slightly better, it might be that receiving is slightly better, it might be close enough that it's a judgement call based on secondary factors. But it's clear that there's not a big advantage to be had from receiving; even with the old rules the receiving team won only a hair over 50% of the games.

No matter how you parse the data, the probabilities of these things happening all change based on the quality of the qb & conditions.

 

[SSOG]

Ask Marty Mornhinweg how he feels about deferring.Old rules... new rules... doesn't matter. Always take the rock.

Ask Mike Shanahan and Hank Stram, too, while you're at it.

Game dependent, but leaning towards "always defer and onside kick".Also, I'd always defer the opening kickoff in favor of receiving to start the second half.

That coaches don't do this baffles me. Defer the kick that is.

I believe getting the kickoff in the first half is actually a (very, very slight) statistical advantage. Points scored in the first quarter correlate better to victory than points scored in the third quarter. There's a huge advantage to being the first team to take a lead, because you knock the other team off their gameplan. Think about Manning's Colts- their entire strategy basically just amounted to "get a lead, then unleash Freeney and Mathis".

 

[Mr. Irrelevant]

Did we really get all the way to Page 2 of this thread without someone taking a crack at something resembling a statistical analysis of this question? Sigh.

Well, because I am a stat geek, I wanted some hard numbers to back up my gut feeling that you should never defer.

Football Outsiders to the rescue! They've got offensive and defensive drive stats, by season, by team, going all the way back to the 2000 season.

As mentioned above, there are only 6 meaningful scenarios at play here ...

1) Opponent scores a TD

2a) Opponent scores a FG, you fail to score

2b) Opponent scores a FG, you score a FG

2c) Opponent scores a FG, you score a TD

3a) Opponent fails to score, you fail to score

3b) Opponent fails to score, you score a FG or TD

... in which you lose in (1) and (2a) and win in (2c) and (3b). For the remaining two, (2a) and (3a), your opponent gets the ball back in sudden-death format. In true sudden-death, the odds of your winning in that situation are right around 40%, but let's shade that up to 42% because I'm both feeling generous and a Douglas Adams fan.

Let's assume you have an excellent defense and a pedestrian offense - for instance, you're the 2011 Baltimore Ravens.

As per the FO link above, the Ravens' D yielded .120 TD's per drive and 1.29 points - by calculation, this implies .150 FG's per drive. (Both figures led the AFC, BTW.)

Meanwhile, their offense scored .200 TD's and 1.90 points (thus .167 FG's) per drive - both just slightly above the 2011 NFL average.

Our weighted chances of winning by scenario are then:

1) .120 * 0 = 0

2a) .150 * .633 * 0 = 0

2b) .150 * .167 * .42 = 0.011

2c) .150 * .200 * 1 = 0.030

3a) .730 * .633 * .42 = 0.194

3b) .730 * .367 * 1 = 0.268

-----

Total odds of winning = 0.503

Hmmmmmmm. One self-serving of crow, comin' right up!

Seriously: you can quibble about the exact assumptions I'm using, but deferring the kick in this situation appears to be much more worthy of consideration than I would have ever suspected.

 

[Mr. Irrelevant]

You know what? I'm an idiot.

Because we're not interested in our odds of winning the game if we defer the kick; we're interested in our odds of winning if we defer compared to what they are if we receive.

And I'm sure, using ths same set of statistical assumptions as above, that our odds of winning by receiving off the bat are gonna be quite a bit higher than 50.3%. I'll take a stab at the numbers later tonight.

But, yeah, I'm probably an idiot.

 

[Sarnoff]

Game dependent, but leaning towards "always defer and onside kick".Also, I'd always defer the opening kickoff in favor of receiving to start the second half.

That coaches don't do this baffles me. Defer the kick that is.

I believe getting the kickoff in the first half is actually a (very, very slight) statistical advantage. Points scored in the first quarter correlate better to victory than points scored in the third quarter. There's a huge advantage to being the first team to take a lead, because you knock the other team off their gameplan. Think about Manning's Colts- their entire strategy basically just amounted to "get a lead, then unleash Freeney and Mathis".

I get the statistics behind that, I'd believe if there is an advantage it's surely slight, but I think it would be evened out or outweighed by the unable-to-quantify strategic advantage offered by having more data about the situation when receiving in the second half. Is the game nearly even or dangerously close to getting out of reach? Is it high-scoring or a defensive battle? Are there game-changing injuries that need to be considered? Stuff like that can all be weighed during the break between halves and a strategy can be worked out behind closed doors rather than on the sideline in the heat of the moment. The decision can be evaluated, debated, reconsidered, etc. among all the coaches and then given to the players. Do we go for a long, clock-killing drive to keep their offense off the field and cold, or do we hurry up and try to get a quick score before things get completely out of hand? If the drive stalls just over midfield, are we better off punting, attempting a long FG, or are we compelled to go for it? Etc.

 

[CalBear]

You know what? I'm an idiot.

Because we're not interested in our odds of winning the game if we defer the kick; we're interested in our odds of winning if we defer compared to what they are if we receive.

And I'm sure, using ths same set of statistical assumptions as above, that our odds of winning by receiving off the bat are gonna be quite a bit higher than 50.3%. I'll take a stab at the numbers later tonight.

But, yeah, I'm probably an idiot.

Three things:1) (This is not just for you): At the beginning of overtime, you can't defer. You can elect to kick or elect to receive. At the beginning of the game, you can defer your decision to the second half.

2) I don't see how the odds of winning if you kick can be anything but the inverse of the odds of winning if you receive--except for the small factor of tie games. So the odds of winning if you receive should be .497 minus the probability of a tie game.

3) But, your stats are limited to drives; from .503 you have to subtract the possibility of a kickoff return TD.

4) And, your probability of winning on the third possession is probably higher if you received to start the OT, so you have to knock .503 down a little bit more.

In the end, it's gonna be pretty close to a dead heat.

 

[mbuehner]

You cant compare OT directly to the average of the game i think. We know running games for instance get stronger later in games as defenses wear down. What are the odds of scoring in classic OT first drives vs earlier kick off drives? I suspect significantly higher. Use that factor to calculate some idea of what the baltimore defense might be expected to give up in OT vs regular time.

And that kickoff return is certainly non-trivial, and again is unique in OT with tired defenders. You have to figure if that Baltimore team baseline had superior kickoff coverage vs the average to determine a reasonable starting field position in OT.

 

[SSOG]

You cant compare OT directly to the average of the game i think. We know running games for instance get stronger later in games as defenses wear down. What are the odds of scoring in classic OT first drives vs earlier kick off drives? I suspect significantly higher. Use that factor to calculate some idea of what the baltimore defense might be expected to give up in OT vs regular time.And that kickoff return is certainly non-trivial, and again is unique in OT with tired defenders. You have to figure if that Baltimore team baseline had superior kickoff coverage vs the average to determine a reasonable starting field position in OT.

IIRC, only one overtime game in history has been won on a kickoff return TD. I would suggest that the possibility certainly is trivial.

 

[mbuehner]

You cant compare OT directly to the average of the game i think. We know running games for instance get stronger later in games as defenses wear down. What are the odds of scoring in classic OT first drives vs earlier kick off drives? I suspect significantly higher. Use that factor to calculate some idea of what the baltimore defense might be expected to give up in OT vs regular time.And that kickoff return is certainly non-trivial, and again is unique in OT with tired defenders. You have to figure if that Baltimore team baseline had superior kickoff coverage vs the average to determine a reasonable starting field position in OT.

IIRC, only one overtime game in history has been won on a kickoff return TD. I would suggest that the possibility certainly is trivial.

You dont need to score to get into scoring position. Were discussing field position, if the average OT kick return is better than average it can easily blow the field position argument.

 

[Mr. Irrelevant]

2) I don't see how the odds of winning if you kick can be anything but the inverse of the odds of winning if you receive--except for the small factor of tie games. So the odds of winning if you receive should be .497 minus the probability of a tie game.

Because if you're averaging 0.61 points per drive more on offense than you allow on defense, which the 2011 Ravens did, you're by definition going to have a greater than 50% chance of winning the game - or at least any subset of a game which consists of an equal number of possessions by both teams.If your team averaged exactly the same number of TDs and FGs per drive on O as you allowed on D, then yes, the two scenarios would have to average out to 50%. But then this whole argument is moot, because for such a team it can't possibly be advantageous to sacrifice the chance at an additional possession by kicking off to start OT.

 

[Zow]

You know what? I'm an idiot.

Because we're not interested in our odds of winning the game if we defer the kick; we're interested in our odds of winning if we defer compared to what they are if we receive.

And I'm sure, using ths same set of statistical assumptions as above, that our odds of winning by receiving off the bat are gonna be quite a bit higher than 50.3%. I'll take a stab at the numbers later tonight.

But, yeah, I'm probably an idiot.

Please do this because the link I showed above suggests it's only %47.

 

[flc735]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

 

[sporthenry]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

 

[Mr. Irrelevant]

You know what? I'm an idiot.

Because we're not interested in our odds of winning the game if we defer the kick; we're interested in our odds of winning if we defer compared to what they are if we receive.

And I'm sure, using ths same set of statistical assumptions as above, that our odds of winning by receiving off the bat are gonna be quite a bit higher than 50.3%. I'll take a stab at the numbers later tonight.

But, yeah, I'm probably an idiot.

Please do this because the link I showed above suggests it's only %47.

I'm short on time so someone else might have to double-check my math, but ...Again, the working thesis here is that your team is the 2011 Ravens:

As per the FO link above, the Ravens' D yielded .120 TD's per drive and 1.29 points - by calculation, this implies .150 FG's per drive. (Both figures led the AFC, BTW.)

Meanwhile, their offense scored .200 TD's and 1.90 points (thus .167 FG's) per drive - both just slightly above the 2011 NFL average.

And the 6 scenarios when we receive become:1') You score a TD

2a') You score a FG, opponent fails to score

2b') You score a FG, opponent scores a FG

2c') You score a FG, opponent scores a TD

3a') You fail to score, opponent fails to score

3b') You fail to score, opponent scores a FG or TD

If the odds of us winning in (2a) and (3a) were 42% without the ball, then they should be 58% in (2a') and (3a') with the ball. So the math works out like so:

1') .200 * 1 = 0.200

2a') .167 * .730 * 1 = 0.122

2b') .167 * .150 * .58 = 0.014

2c') .167 * .120 * 0 = 0

3a') .633 * .730 * .58 = 0.268

3b') .633 * .270 * 0 = 0

-----

Total odds of winning = 0.604

Pretty much what I thought. Even with that team - with a top-notch defense and a thoroughly average offense - you're 10% more likely to win by opting to receive rather than kick to start OT ... even though you're a favorite in both cases.

Make whatever assumptions you like about average starting field position, special-teams TD's, and so on ... there's just no way I can see in which you're going to overcome that sort of statistical chasm between these two options.

 

[CalBear]

1') You score a TD

2a') You score a FG, opponent fails to score

2b') You score a FG, opponent scores a FG

2c') You score a FG, opponent scores a TD

3a') You fail to score, opponent fails to score

3b') You fail to score, opponent scores a FG or TD

If the odds of us winning in (2a) and (3a) were 42% without the ball, then they should be 58% in (2a') and (3a') with the ball. So the math works out like so:

1') .200 * 1 = 0.200

2a') .167 * .730 * 1 = 0.122

2b') .167 * .150 * .58 = 0.014

2c') .167 * .120 * 0 = 0

3a') .633 * .730 * .58 = 0.268

3b') .633 * .270 * 0 = 0

-----

Total odds of winning = 0.604

Can you really look at these as dependent probabilities? I don't think so, given that we're looking at statistical likelihoods based on pooled results--this isn't a die roll. If you calculate them as independent, it comes out to .507.

 

[Ignoratio Elenchi]

Haven't read most of the thread, but I see a bunch of numbers so I'm looking forward to seeing what everyone's come up with.

For a slightly less numerical approach, let's make a couple of simplifying assumptions:

- Ignore the possibility of onside kicks, fumbled kick returns, etc. for the moment (they are definitely an intriguing wild card here). Just assume if you kick off, the other team starts with the ball, and if the other team kicks off, you start with the ball.

- Let T = probability of scoring a TD when you get the ball, F = probability of scoring a FG when you get the ball, N = probability of not scoring when you get the ball. Since these are the only three possible outcomes, T + F + N = 1. Assume these probabilities are the same for both teams.

If you receive the opening kickoff, scenarios in which you win are: scoring a TD on the opening drive, or scoring a FG and holding your opponent to no points on the second drive. So you have a T chance of winning on the opening drive, and you have an F*N chance of winning after two drives (i.e. your first possession and your opponent's first possession). So your probability of winning after two drives is T + FN. If the game is still tied after these two drives, then you go into a sudden-death overtime and you get the ball first. This has an additional win probability associated with it which we'll call SD+ (SD for sudden death, and the + indicating that you get the ball first).

If you don't receive the opening kickoff, scenarios in which you win are: holding your opponent to no score and then scoring a FG or a TD, or holding your opponent to a FG and then scoring a TD. So your probability of winning after two drives is NF + NT + FT. If the game is still tied after these two drives, then you go into a sudden-death overtime and you do not get the ball first. This has an additional win probability associated with it which we'll call SD- (SD for sudden death, and the - indicating that you do not get the ball first).

So

Win when receiving the kickoff to start overtime = T + FN + SD+

Win when kicking off to start overtime = FN + NT + FT + SD-

If we set up the inequality, subtract FN from both sides and redistribute the variables on the kicking off side, we get:

T + SD+ ??? T(F+N) + SD-

Since F+N is <= 1, and SD- < SD+, then clearly the left hand side is greater than the right hand side. It doesn't matter what values you choose for T, F, and N. The left hand side is always greater than the right hand side, which means receiving to start OT is better than kicking off to start OT.

I have a feeling I've done something wrong here. Even if not, there are some obvious objections to this analysis which I'll save for another post (though I'm sure someone will beat me to that), but it seemed like a simple enough place to start.

 

[flc735]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding.

being up by 3 is better than being down by 3 with 4 downs to tie and 3 downs to win. you even agreed to this. unless you changed your mind over night, you have no argument here.if team A scores 0. Team B is equal to them. they have an equal change to get a td or fg as team A did. if team A scores 0, team B DOES NOT have 4 downs. they would punt on 4th down. advantage: evenif team A scores 3. Team B gets their extra down to tie but still 3 downs to win with a TD. (they would kick a FG on 4th down if possible. they would not go for a TD on 4th down). you are right in that additional info will allow you to make a more informed decision. that is why these new rules are better than the old ones. HOWEVER, i would rather be team A, up by 3 than team B, down by 3 with the knowledge that they have to go for it on 4th until they are in FG range.this is my opinion. you already said this is your opinion as well.advantage: team A (according to both of us)if team A scores a TD, team A winsadvantage: team Ai already said all that but again, there is no scenario in which team B has the advantage.

 

[sporthenry]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding.

being up by 3 is better than being down by 3 with 4 downs to tie and 3 downs to win. you even agreed to this. unless you changed your mind over night, you have no argument here.if team A scores 0. Team B is equal to them. they have an equal change to get a td or fg as team A did. if team A scores 0, team B DOES NOT have 4 downs. they would punt on 4th down. advantage: evenif team A scores 3. Team B gets their extra down to tie but still 3 downs to win with a TD. (they would kick a FG on 4th down if possible. they would not go for a TD on 4th down). you are right in that additional info will allow you to make a more informed decision. that is why these new rules are better than the old ones. HOWEVER, i would rather be team A, up by 3 than team B, down by 3 with the knowledge that they have to go for it on 4th until they are in FG range.this is my opinion. you already said this is your opinion as well.advantage: team A (according to both of us)if team A scores a TD, team A winsadvantage: team Ai already said all that but again, there is no scenario in which team B has the advantage.

Well considering team A only has a 47% chance of winning when they start at 1st and 10 from their own 20, I'd say team B would have a little advantage or at best an equal chance? I'm not going to get into more circular logic with you and will just wait for Stats or someone to post a more statistical approach then what we have here.

 

[daveR]

I think we are ignoring a ton of human factors in this discussion, such as momentum, expectations of taking the ball, and just plain exhaustion, to name a few.

 

[Leroy Hoard]

I like that the team recieving gets it not only to start (and possibly finish) but maybe 2 of the first 3 possessions.

 

[rick6668]

Haven't read most of the thread, but I see a bunch of numbers so I'm looking forward to seeing what everyone's come up with.

For a slightly less numerical approach, let's make a couple of simplifying assumptions:

- Ignore the possibility of onside kicks, fumbled kick returns, etc. for the moment (they are definitely an intriguing wild card here). Just assume if you kick off, the other team starts with the ball, and if the other team kicks off, you start with the ball.

- Let T = probability of scoring a TD when you get the ball, F = probability of scoring a FG when you get the ball, N = probability of not scoring when you get the ball. Since these are the only three possible outcomes, T + F + N = 1. Assume these probabilities are the same for both teams.

If you receive the opening kickoff, scenarios in which you win are: scoring a TD on the opening drive, or scoring a FG and holding your opponent to no points on the second drive. So you have a T chance of winning on the opening drive, and you have an F*N chance of winning after two drives (i.e. your first possession and your opponent's first possession). So your probability of winning after two drives is T + FN. If the game is still tied after these two drives, then you go into a sudden-death overtime and you get the ball first. This has an additional win probability associated with it which we'll call SD+ (SD for sudden death, and the + indicating that you get the ball first).

If you don't receive the opening kickoff, scenarios in which you win are: holding your opponent to no score and then scoring a FG or a TD, or holding your opponent to a FG and then scoring a TD. So your probability of winning after two drives is NF + NT + FT. If the game is still tied after these two drives, then you go into a sudden-death overtime and you do not get the ball first. This has an additional win probability associated with it which we'll call SD- (SD for sudden death, and the - indicating that you do not get the ball first).

So

Win when receiving the kickoff to start overtime = T + FN + SD+

Win when kicking off to start overtime = FN + NT + FT + SD-

If we set up the inequality, subtract FN from both sides and redistribute the variables on the kicking off side, we get:

T + SD+ ??? T(F+N) + SD-

Since F+N is <= 1, and SD- < SD+, then clearly the left hand side is greater than the right hand side. It doesn't matter what values you choose for T, F, and N. The left hand side is always greater than the right hand side, which means receiving to start OT is better than kicking off to start OT.

I have a feeling I've done something wrong here. Even if not, there are some obvious objections to this analysis which I'll save for another post (though I'm sure someone will beat me to that), but it seemed like a simple enough place to start.

Where your logic is faulty is that the probability of scoring a FG when you get the ball = F is not the same for both teams depending on whether the Receiving team scores a FG or not. If they do, Team kicking in OT has a greater chance of a FG than the receiving team as they will have 4 downs to get into FG range.

In your example, the F in each equation is not necessarily equal.

 

[CalBear]

Where your logic is faulty is that the probability of scoring a FG when you get the ball = F is not the same for both teams depending on whether the Receiving team scores a FG or not. If they do, Team kicking in OT has a greater chance of a FG than the receiving team as they will have 4 downs to get into FG range.

Also, better average field position.

 

[Pipes]

I think we are ignoring a ton of human factors in this discussion, such as momentum, expectations of taking the ball, and just plain exhaustion, to name a few.

Also non human factors as well such as the weather. What if there are strong winds coming directly from one end of the stadium...choosing what direction to go could be huge in that scenario.

 

[flc735]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding.

being up by 3 is better than being down by 3 with 4 downs to tie and 3 downs to win. you even agreed to this. unless you changed your mind over night, you have no argument here.if team A scores 0. Team B is equal to them. they have an equal change to get a td or fg as team A did. if team A scores 0, team B DOES NOT have 4 downs. they would punt on 4th down. advantage: evenif team A scores 3. Team B gets their extra down to tie but still 3 downs to win with a TD. (they would kick a FG on 4th down if possible. they would not go for a TD on 4th down). you are right in that additional info will allow you to make a more informed decision. that is why these new rules are better than the old ones. HOWEVER, i would rather be team A, up by 3 than team B, down by 3 with the knowledge that they have to go for it on 4th until they are in FG range.this is my opinion. you already said this is your opinion as well.advantage: team A (according to both of us)if team A scores a TD, team A winsadvantage: team Ai already said all that but again, there is no scenario in which team B has the advantage.

Well considering team A only has a 47% chance of winning when they start at 1st and 10 from their own 20, I'd say team B would have a little advantage or at best an equal chance? I'm not going to get into more circular logic with you and will just wait for Stats or someone to post a more statistical approach then what we have here.

again, no. you don't seem to be able to grasp this."47%" to score on their first drive. that doesn't mean team B has a 53% chance of scoring. team B also has a 47% chance assuming all other factors are equal.for the 3rd time, there are scenarios where team B = team A but there is no scenario in which team B has the advantage.ps. i don't think you know what circular logic is.

 

[sporthenry]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding.

being up by 3 is better than being down by 3 with 4 downs to tie and 3 downs to win. you even agreed to this. unless you changed your mind over night, you have no argument here.if team A scores 0. Team B is equal to them. they have an equal change to get a td or fg as team A did. if team A scores 0, team B DOES NOT have 4 downs. they would punt on 4th down. advantage: evenif team A scores 3. Team B gets their extra down to tie but still 3 downs to win with a TD. (they would kick a FG on 4th down if possible. they would not go for a TD on 4th down). you are right in that additional info will allow you to make a more informed decision. that is why these new rules are better than the old ones. HOWEVER, i would rather be team A, up by 3 than team B, down by 3 with the knowledge that they have to go for it on 4th until they are in FG range.this is my opinion. you already said this is your opinion as well.advantage: team A (according to both of us)if team A scores a TD, team A winsadvantage: team Ai already said all that but again, there is no scenario in which team B has the advantage.

Well considering team A only has a 47% chance of winning when they start at 1st and 10 from their own 20, I'd say team B would have a little advantage or at best an equal chance? I'm not going to get into more circular logic with you and will just wait for Stats or someone to post a more statistical approach then what we have here.

again, no. you don't seem to be able to grasp this."47%" to score on their first drive. that doesn't mean team B has a 53% chance of scoring. team B also has a 47% chance assuming all other factors are equal.for the 3rd time, there are scenarios where team B = team A but there is no scenario in which team B has the advantage.ps. i don't think you know what circular logic is.

Yep, you are right it was more square thinking. And I guess I don't know how this website works. Plug in 15:00 left, 1st and 10 from your own 20 in a tie game in OT and it gives a 47% chance of winning. There is a chance of a tie but unless I'm missing something, and 47% chance to score on the first drive? Not sure what math you are using. http://wp.advancednflstats.com/winprobcalc1.php

 

[flc735]

the mistake here is people brushing off the chance that their opponent scores a td first. it's easy to say "i trust my defense to stop them" and rationalize the odds of that happening down to zero. if you are going to say that, why not say "i trust my offense to score a td" and be done with it all?it is still better to receive.as far as 4 downs for the second team go:it's better to be ahead than to have an extra down and trying to catch up. lets say team A kicks their FG. which team would you rather be? i'd choose team A+3 over team B + an extra down. also, if team A does not score, team B is not going to use all 4 downs on their drive.

The chance of them scoring a TD on a drive that starts at their own 20 is 15%. I'm not brushing it off, I'm just going with the other 85% of the time. As far as the 4 down scenario, your comparison is lacking. You have an 8% chance of kicking a FG on a drive starting at your own 20. So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8% (although can't find the stats exactly). This is still simplified and it'll take an extreme statistical analysis with more information based on drive statistics but it isn't as simple as people make and I'd say the statistics are very close to 50-50.

that's completely wrong.first of all, where are you getting this? what is is based off of? are you saying the chances that a team in OT make a FG starting with a kickoff (not the 20, btw. a kickoff return) are 15%, i don't have to look at anything to know thats wrong. i remember an old stat that said the receiving team in OT won on thier first drive 47% of the time (or something close to 50%, i forget exactly).that right there tells me the 15% number is wrong unless....you are talking about TDs only in which case...if it's from OT games this year, the sample size is too small.if it's from all OT games, sample size is fine but the number is skewed because in the past, teams would forgo the TD attempt in exchange for a 2nd and goal FG attempt from the 6, for example.if it's from all recent games, that doesn't factor in the situation. often, teams will start the game running the ball for the sole purpose of establishing the run, even though they know it does not give them the best chance to score on that drive.second of all, even if that's a legitimate stat, that does not mean the defending team has an 85% of winning. lets say the second team gets the ball back at their 20, they would also have a 15% chance of scoring.

So of course I'd like to be up 3 in OT, but those chances are rather slim and the chance to get a first down with 4 downs rather than 3 downs will be greater than 8%

i already addressed this. if team A kicks the FG, team B gets 4 downs. in this case, team A still has the advantage <---we both agree on that.if team A scores a td, they win. again, the advantage goes to team Aif team A does not score, team B is even with team A. as in, they have an equal chance of scoring as team A did on the previous drive. Why on earth would team B go for it on 4th down if it is still tied?The only way team B has an advantage is if they intercept early or 3 and out team A. the odds of that are just as slim as team A scoring a td on their first drive and therefore, you can cancel the int vs td chances out. and that leaves the FG or no FG advantage/no advantage to either that i explained above.this is fairly simple.

First, the difference between a kickoff and starting from the 20 is just semantics. This is a statistical debate last year over 40% of all kickoffs ended in touchbacks and the average starting field position was the 22.6 yard line. So yes, a kickoff is the same as starting at the 22.6 or about the 20 yard line in a statistical debate. I'm saying the chance of a drive starting on your 20 and ending in a TD are 15% and ending in a FG are 8% (for all drives anytime in the game). The stat about games in OT were that 30% of games were ended on the first drive and 60% of the time the team receiving won. So if you add up the FG and TD% you get to 23% which is decently close to 30% (although this would go more to the point that defenses are more tired towards the end of the game). Perhaps that is the flaw with drive stats that they don't take into consideration other things such as the time in the game, but those are the statistics available to us and all other conjecture is just that. It would be like saying batting average increases in the 14th inning so should that affect our decision to bunt? You don't see many baseball statistics on this and football is even further behind the statistical 8-ball than baseball. Another thing that the facts don't show in the situation is that often times, teams become very conservative once they are in FG range even in the current OT rules so this would decrease the chances of us losing on the first drive.As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding. Nobody is saying you have an 85% chance of winning if you stop them but the average field position after a kickoff is worse than after a stopped drive (especially if you turn them over), so you already have increased your chances right there. Perhaps it is just a matter of preference but either way, the stats are very close to 50-50 at worst.

As far as your second point, the benefit to deferring is that say the other team is going to score a FG regardless. If they have the ball first, you will have 4 downs whereas if you have the ball first you will only get 3 downs. Additional information gives you a better chance to make a more informed decision. And this point can be made likewise, if the other team has the ball first, they won't go for it on 4th and short which is actually has a very high probability of succeeding.

being up by 3 is better than being down by 3 with 4 downs to tie and 3 downs to win. you even agreed to this. unless you changed your mind over night, you have no argument here.if team A scores 0. Team B is equal to them. they have an equal change to get a td or fg as team A did. if team A scores 0, team B DOES NOT have 4 downs. they would punt on 4th down. advantage: evenif team A scores 3. Team B gets their extra down to tie but still 3 downs to win with a TD. (they would kick a FG on 4th down if possible. they would not go for a TD on 4th down). you are right in that additional info will allow you to make a more informed decision. that is why these new rules are better than the old ones. HOWEVER, i would rather be team A, up by 3 than team B, down by 3 with the knowledge that they have to go for it on 4th until they are in FG range.this is my opinion. you already said this is your opinion as well.advantage: team A (according to both of us)if team A scores a TD, team A winsadvantage: team Ai already said all that but again, there is no scenario in which team B has the advantage.

Well considering team A only has a 47% chance of winning when they start at 1st and 10 from their own 20, I'd say team B would have a little advantage or at best an equal chance? I'm not going to get into more circular logic with you and will just wait for Stats or someone to post a more statistical approach then what we have here.

again, no. you don't seem to be able to grasp this."47%" to score on their first drive. that doesn't mean team B has a 53% chance of scoring. team B also has a 47% chance assuming all other factors are equal.for the 3rd time, there are scenarios where team B = team A but there is no scenario in which team B has the advantage.ps. i don't think you know what circular logic is.

Yep, you are right it was more square thinking. And I guess I don't know how this website works. Plug in 15:00 left, 1st and 10 from your own 20 in a tie game in OT and it gives a 47% chance of winning. There is a chance of a tie but unless I'm missing something, and 47% chance to score on the first drive? Not sure what math you are using. http://wp.advancednflstats.com/winprobcalc1.php

i like that you are taking the word of some unknown calculator over us, who are trying to explain it to you. we don't even know how they come up with 47% or how they calculate any of their other numbers. it also saysTD Prob: 0.15FG Prob: 0.08meaning it is almost twice as easy for team A to score a TD than to kick a FG. that doesn't make a lot of sense.i am sorry you don't get it.

 

[CalBear]

i like that you are taking the word of some unknown calculator over us, who are trying to explain it to you. we don't even know how they come up with 47% or how they calculate any of their other numbers. it also saysTD Prob: 0.15FG Prob: 0.08meaning it is almost twice as easy for team A to score a TD than to kick a FG. that doesn't make a lot of sense.i am sorry you don't get it.

There's definitely serious wonkiness in the FootballOutsiders stats (not surprising because they often fail to understand statistics). From 2000-2012 in OT games, 121 were won by FG and only 26 by TD. PFR doesn't break it down by "first drive", but if you limit it to the first three minutes of OT, it's still 15-6 in favor of FGs over TDs. In fact, if you look at the first two minutes of OT in that time frame, a total of four were won on offensive TDs, while 1 was won by INT return and 2 were won by punt returns. In the first three minutes, there are 6 offensive TDs, 3 INT returns and 2 punt returns. So, in this very limited data set, the team which kicked off was almost as likely to score a TD as the team which received the kick.Any data set which assumes more probability of an offensive TD than an offensive FG on an OT drive is probably broken.

 

[sporthenry]

i like that you are taking the word of some unknown calculator over us, who are trying to explain it to you. we don't even know how they come up with 47% or how they calculate any of their other numbers. it also saysTD Prob: 0.15FG Prob: 0.08meaning it is almost twice as easy for team A to score a TD than to kick a FG. that doesn't make a lot of sense.i am sorry you don't get it.

There's definitely serious wonkiness in the FootballOutsiders stats (not surprising because they often fail to understand statistics). From 2000-2012 in OT games, 121 were won by FG and only 26 by TD. PFR doesn't break it down by "first drive", but if you limit it to the first three minutes of OT, it's still 15-6 in favor of FGs over TDs. In fact, if you look at the first two minutes of OT in that time frame, a total of four were won on offensive TDs, while 1 was won by INT return and 2 were won by punt returns. In the first three minutes, there are 6 offensive TDs, 3 INT returns and 2 punt returns. So, in this very limited data set, the team which kicked off was almost as likely to score a TD as the team which received the kick.Any data set which assumes more probability of an offensive TD than an offensive FG on an OT drive is probably broken.

Well you have to understand that drive stats don't take into consideration the time. This is a flaw but there aren't nearly enough OT games let along new OT games to have accurate stats. You are using 121 games which isn't big enough of a sample size especially considering they were under the new rules where a TD provided no added benefit. And just b/c you think the numbers seem weird is supposed to refute a reliable stat website? If you can refute this stats that would be one thing. This year there have been 765 XPA while there have been 554 FGs. Yes, it isn't quite 2 to 1 b/c rate that the FG% goes up is much higher than the rate TD goes up as field position gets closer to the opposing EZ. http://www.advancednflstats.com/2009/01/drive-results.htmlYes, one problem with the drive stats are they seem to hold the OT drives equal to that of a drive starting in the 1st quarter but there isn't enough stats to give us true OT stats.

 

[SaintsInDome2006]

The new OT rules can make a mockery of this whole issue.

HOU got the opening kickoff, got a FG. Now Jax has a chance to win but they tied.

Jax stupidly went for it on 4th down (4th & 10) from the HOU 47 when they clearly could have punted and held HOU deep. Even if HOU gets just 10-15 yards they could have won. There was 2:10 left when Johnson took the game to the house. I think that is the sequence everyone should be talking about:

- In a tie game in OT with 3 minutes or less left do you punt from the 50 and risk a tie (or hope to pin for a late score) or do you go for it on 4th and long?

Because if Jax had punted the game is very likely a tie in the end. Jax has a terrific young punter too, he could have pinned them. They just went for it because getting that upset win would have been the high point of their whole year.

I think the NFL has only served to prolong overtimes and lead to greater possibility of ties. No one is helped by accepting or deferring in this situation, no matter what you will get your chance.

The only way I see this being different is if it has been a very, very low scoring game with bad weather.

 

[CalBear]

Yes, one problem with the drive stats are they seem to hold the OT drives equal to that of a drive starting in the 1st quarter but there isn't enough stats to give us true OT stats.

A limited data set (though N=150 isn't bad) of actual OT results has infinitely more relevance than a large data set of non-OT results. Consider the scenario of first-and-goal on the 6 yard line; do you really think the results of that situation are anywhere close to equivalent in the first quarter and in OT?This we can say: With the old OT rules, games end with a FG at least 75% of the time, so if you are looking at a data set that has TDs scored twice as often as FGs, you're looking at the wrong data set. And it is 100% obvious why this is the case; in OT (with the old rules) you have incentive to kick a FG when you would otherwise continue trying for a TD.With the new rules, the team with the first possession will be more likely to keep going for a TD than they would have under the old rules (for example, they won't kick a FG on first and goal from the 6), though perhaps they'll still be more conservative once in field-goal range than they would be in a normal-game situation. On the other hand, after a FG on the first possession, the team with the second possession will be infinitely more likely to go for it on fourth down, which increases the chances of a successful drive (either FG or TD). You can't directly apply a data set of normal-game-situation outcomes to situations which operate under different rules.

 

[CalBear]

- In a tie game in OT with 3 minutes or less left do you punt from the 50 and risk a tie (or hope to pin for a late score) or do you go for it on 4th and long?

Because if Jax had punted the game is very likely a tie in the end. Jax has a terrific young punter too, he could have pinned them.

(Go Bears!)Going for it on fourth and 10 in that situation is insane. Your win probability is probably higher if you punt; if you pin them inside the 5, you could still win on a safety, or if you hold them to 3-and-out you'll get one more possession.

 

[flc735]

i like that you are taking the word of some unknown calculator over us, who are trying to explain it to you. we don't even know how they come up with 47% or how they calculate any of their other numbers. it also says

TD Prob: 0.15

FG Prob: 0.08

meaning it is almost twice as easy for team A to score a TD than to kick a FG. that doesn't make a lot of sense.

i am sorry you don't get it.

There's definitely serious wonkiness in the FootballOutsiders stats (not surprising because they often fail to understand statistics). From 2000-2012 in OT games, 121 were won by FG and only 26 by TD. PFR doesn't break it down by "first drive", but if you limit it to the first three minutes of OT, it's still 15-6 in favor of FGs over TDs. In fact, if you look at the first two minutes of OT in that time frame, a total of four were won on offensive TDs, while 1 was won by INT return and 2 were won by punt returns. In the first three minutes, there are 6 offensive TDs, 3 INT returns and 2 punt returns. So, in this very limited data set, the team which kicked off was almost as likely to score a TD as the team which received the kick.

Any data set which assumes more probability of an offensive TD than an offensive FG on an OT drive is probably broken.

Well you have to understand that drive stats don't take into consideration the time. This is a flaw but there aren't nearly enough OT games let along new OT games to have accurate stats. You are using 121 games which isn't big enough of a sample size especially considering they were under the new rules where a TD provided no added benefit. And just b/c you think the numbers seem weird is supposed to refute a reliable stat website? If you can refute this stats that would be one thing. This year there have been 765 XPA while there have been 554 FGs. Yes, it isn't quite 2 to 1 b/c rate that the FG% goes up is much higher than the rate TD goes up as field position gets closer to the opposing EZ.

http://www.advancednflstats.com/2009/01/drive-results.html

Yes, one problem with the drive stats are they seem to hold the OT drives equal to that of a drive starting in the 1st quarter but there isn't enough stats to give us true OT stats.

Well you have to understand that drive stats don't take into consideration the time. This is a flaw but there aren't nearly enough OT games let along new OT games to have accurate stats. You are using 121 games which isn't big enough of a sample size especially considering they were under the new rules where a TD provided no added benefit.

we are only talking about the 1st two or 1st four drives. time is irrelevant.

And just b/c you think the numbers seem weird is supposed to refute a reliable stat website? If you can refute this stats that would be one thing. This year there have been 765 XPA while there have been 554 FGs. Yes, it isn't quite 2 to 1 b/c rate that the FG% goes up is much higher than the rate TD goes up as field position gets closer to the opposing EZ.

the numbers are illogical and none of us, including yourself understand where they came from. from that, we can conclude the "47%" is unreliable. this site is a reputable FOOTBALL website, not a reputable mathematics website.

http://www.advancednflstats.com/2009/01/drive-results.html

Yes, one problem with the drive stats are they seem to hold the OT drives equal to that of a drive starting in the 1st quarter but there isn't enough stats to give us true OT stats.

if that is what you think, everything you said based on that site is worthless.

regardless of all that, i originally explained why team A has the advantage in two different ways. both of which do not require any specific percentages so that we could avoid this argument. you even agreed with me. worst of all, you don't know anything about this. all you are doing is pointing to a mystery calculator and claiming that we are wrong. go back and read. try to wrap your head around what we are saying. we call that learning.

 

[CalBear]

Yes, one problem with the drive stats are they seem to hold the OT drives equal to that of a drive starting in the 1st quarter but there isn't enough stats to give us true OT stats.

Again: 150 games of actual OT situations is worth a lot more than 1000 games of non-OT situations. Your position is like saying that we should ignore the stats on fourth-and-5 situations, because we don't have enough of them, and use first-and-10 situational stats instead, because we have more. It's more important that the stats be relevant to the situation than that they be voluminous.

 

[Ignoratio Elenchi]

Where your logic is faulty is that the probability of scoring a FG when you get the ball = F is not the same for both teams depending on whether the Receiving team scores a FG or not. If they do, Team kicking in OT has a greater chance of a FG than the receiving team as they will have 4 downs to get into FG range.In your example, the F in each equation is not necessarily equal.

Excellent point. I knew the simplified assumptions were faulty, my original post was simply setting up a non-numerical framework to build off of.So we want to say that not all F's are equal. I agree. However, to the extent that the second team has a higher probability of kicking a FG if the first team already kicked a field goal, that seems like it might actually favor the first team, because the scenarios in which both teams are tied 3-3 after each has had a possession favors the first team (since they are then in sudden death and they get the ball first). Of course, changing the values of F also implies that they'd have different values for N, which would also impact the equations. I'd have to give it some more thought but by immediate gut feeling is that this probably wouldn't ultimately change my conclusion that it's better to get the ball first.

Also, better average field position.

Is it? Honest question, I don't know. Can you explain why? I'm not sure I see why we'd assume that the second team to get the ball has better average field position than the first team.

 

[mbuehner]

Is there any data on kick off returns that can be filtered by OT? I think that average return might be highly relevant here, and I suspect the average OT return is better than in the rest of the game.

 

[CalBear]

Also, better average field position.

Is it? Honest question, I don't know. Can you explain why? I'm not sure I see why we'd assume that the second team to get the ball has better average field position than the first team.

I'm assuming so but I don't have stats. The second drive starts after:1) Punt2) Turnover3) Missed FGFor #2 and #3, starting field position is almost certainly going to better than average kickoff starting field position. For punts, it'll depend on how far the other team advanced and whether they put the punt inside the 20.

 

[CalBear]

Is there any data on kick off returns that can be filtered by OT? I think that average return might be highly relevant here, and I suspect the average OT return is better than in the rest of the game.

Opening-OT kickoffs, 2000-2012(This excludes on-sides kicks, of which I think there was one).

161 kickoffs, 39 touchbacks (24.2%), average return 24.68 yards.

The non-OT stats are:

16.1% touchbacks, average return 22.63 yards.

So the returns are a little longer, but touchbacks are somewhat more likely. It's probably a wash.

 

[Ignoratio Elenchi]

Also, better average field position.

Is it? Honest question, I don't know. Can you explain why? I'm not sure I see why we'd assume that the second team to get the ball has better average field position than the first team.

I'm assuming so but I don't have stats. The second drive starts after:1) Punt2) Turnover3) Missed FGFor #2 and #3, starting field position is almost certainly going to better than average kickoff starting field position. For punts, it'll depend on how far the other team advanced and whether they put the punt inside the 20.

What about 4) Made FG? Not that that would impact this question anyway, since it just results in a kickoff anyway. I think you're probably right, though, that the 2nd team will have better average field position. What might swing it the other way is if punts are much more likely outcomes than turnovers and missed FGs (which I'd guess is probably the case) and if average starting field position following punts is worse than kickoffs by enough that the sheer volume of punts would drive down the average. Still, not sure this would change my conclusion. Is the field position differential enough of an advantage that it outweighs the potential benefits of getting the ball first?

 

[CalBear]

Still, not sure this would change my conclusion. Is the field position differential enough of an advantage that it outweighs the potential benefits of getting the ball first?

It's a factor, probably not a compelling one in isolation.