NFL Overtime

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
So how many games is a sufficient sample size?
It depends on the question you're asking.

How much variation are you willing to tolerate and what confidence level do you want to have? You could ask the question like this: How many games would we need ("n") to have a 95 percent confidence level that an outcome of 48 to 52 percent is a true reflection of the actual odds?

My stats are a bit rusty, and I'm not going to show my work but it looks to me as though n in this case would be somewhere north of 450 and south of 1000.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
It depends on the question you're asking.

How much variation are you willing to tolerate and what confidence level do you want to have? You could ask the question like this: How many games would we need ("n") to have a 95 percent confidence level that an outcome of 48 to 52 percent is a true reflection of the actual odds?

My stats are a bit rusty, and I'm not going to show my work but it looks to me as though n in this case would be somewhere north of 450 and south of 1000.
So we need 30 years' worth of data at a minimum (3 years produced 45 overtime games). Uh yeah, don't think we're going to wait that long for the answer.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
So we need 30 years' worth of data at a minimum (3 years produced 45 overtime games). Uh yeah, don't think we're going to wait that long for the answer.
That's the son of a bitch with football stats. Rarely is the sample size big enough to draw conclusions. But that's why Bellhorn's approach is more valid than yours - the samples it can speak to are much larger.

The other factor to consider is that the chances of winning or losing based on the coin flip are not static from game-to-game or team-to-team. The Patriots could not stop the Chiefs in the second half to save their lives (even giving up a FG drive with < 30 seconds remaining), but the Chiefs have a terrible defense. For that matchup, the coin flip was much more critical than it would be between two defense-minded squads.
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
So we need 30 years' worth of data at a minimum (3 years produced 45 overtime games). Uh yeah, don't think we're going to wait that long for the answer.
I don't understand what you mean by "answer."

Here is a fact: Even if the true odds of the first team winning under current rules were actually 58 percent, 21-20-4 would not be a statistically unexpected distribution over 45 results. I think that may even be true as high as 62 or 63 percent but as I said my stats are a bit rusty.

It is what it is. So, if one were looking for a way to answer someone who thought the current rules are not "fair" or who wanted to articulate a different rule, the current sample doesn't really help that much, statistically speaking. That doesn't mean that the current system isn't defensible or even the best possible system. It just means that if you want to try to use statistics to help answer that question, you're out of luck. Sample is too small. It happens.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
That's the son of a bitch with football stats. Rarely is the sample size big enough to draw conclusions. But that's why Bellhorn's approach is more valid than yours - the samples it can speak to are much larger.

The other factor to consider is that the chances of winning or losing based on the coin flip are not static from game-to-game or team-to-team. The Patriots could not stop the Chiefs in the second half to save their lives (even giving up a FG drive with < 30 seconds remaining), but the Chiefs have a terrible defense. For that matchup, the coin flip was much more critical than it would be between two defense-minded squads.
I don't understand what you mean by "answer."

Here is a fact: Even if the true odds of the first team winning under current rules were actually 58 percent, 21-20-4 would not be a statistically unexpected distribution over 45 results. I think that may even be true as high as 62 or 63 percent but as I said my stats are a bit rusty.

It is what it is. So, if one were looking for a way to answer someone who thought the current rules are not "fair" or who wanted to articulate a different rule, the current sample doesn't really help that much, statistically speaking. That doesn't mean that the current system isn't defensible or even the best possible system. It just means that if you want to try to use statistics to help answer that question, you're out of luck. Sample is too small. It happens.
That's all well and good, but the league isn't going to wait years and years and years more for enough data to constitute a sufficient sample size to know with 95% accuracy whether the results over the last 3 years tell us if the current system is truly fair.

And because we are not using fair coins or die, but it's humans - with emotions and momentum and injuries and fatigue and varying performance from game to game and moment to moment - we will never really know what system is most "fair" using probabilities.

I mean, what if the current system is most "fair" if good offenses are getting the ball first against bad defenses? What does that REALLY tell us? What if that same system would be more "fair" to the team that plays defense first if it's the Cardinals getting the ball first going up against the Bears' defense? That's likely going to be a quick stop for the Bears, and excellent field position with Chicago only needing a FG to win.

But maybe in any one particular Ari-Chi matchup, the Bears are so banged up and Arizona is on fire (having scored their last 3 possessions in regulation) and so in that specific game you can throw the probabilities out the window?

Honestly, how are we *really* to know what the most "fair" system is using probabilities?
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
That's all well and good, but the league isn't going to wait years and years and years more for enough data to constitute a sufficient sample size to know with 95% accuracy whether the results over the last 3 years tell us if the current system is truly fair.
I agree. And, I'm not sure you and I are disagreeing.

Statistics is a discipline that uses commonly accepted methods to avoid wrongfully equating correlation with causation. If you want to support an argument for causation with the imprimatur of statistics you must play by its rules. If you can’t, you need other evidence.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
I agree. And, I'm not sure you and I are disagreeing.

Statistics is a discipline that uses commonly accepted methods to avoid wrongfully equating correlation with causation. If you want to support an argument for causation with the imprimatur of statistics you must play by its rules. If you can’t, you need other evidence.
Well if we really can't use probabilities because it's a human game, not a game of true probabilistic outcomes like fair coins being flipped (for example, how do we know the actual probabilities of the Patriots scoring an opening TD in overtime after they've scored on two straight possessions in regulation against the Chiefs, but with Gronk out due to injury and the backup right tackle in the game?), and we can't use anything less than 30 years' worth of data (being too insufficient a sample), how, exactly, are we to determine what's the most "fair" way of doing overtime?

I don't see how my thought process on the matter is any worse than anyone else's here. But as Bellhorn said....I really don't have a great grasp of the nuances of probability.
 

InstaFace

The Ultimate One
SoSH Member
Sep 27, 2016
21,759
Pittsburgh, PA
So how many games is a sufficient sample size?
Statisticians will tell you that it depends on the distribution type, degrees of freedom, variance, and what your threshold is for certainty (p=0.05?). It being a matter of heuristics, there is no right answer - merely a set of conventions that usually apply when you've got a normal distribution and are talking about the probability of the observed mean being within X of the true mean.

With binary inputs and outcomes (start with/without the ball, win/lose) for a given rule set, it's been too long for me to know what test would be applied to assess the null hypothesis, but regardless I know you will always have room for argument over the answer to this question.

edit: an important point that DDB alludes to in post 54, but doesn't squarely articulate, is that the outer bounds of the true probability, i.e. the range within which we can say the true odds reside with 95% certainty, narrows steadily and exponentially as sample size increases. So if that range is 40-60% at n=45, it might get down to 45-55% by n=100, and so on. So whatever range you'd be comfortable calling it "fair", as you increase your N you'll get closer to that. And even at our present n=45, there is some acceptable range we can be confident (p<0.05) that it is within.
 
Last edited:

bankshot1

Member
SoSH Member
Feb 12, 2003
24,652
where I was last at
It depends on the question you're asking.

How much variation are you willing to tolerate and what confidence level do you want to have? You could ask the question like this: How many games would we need ("n") to have a 95 percent confidence level that an outcome of 48 to 52 percent is a true reflection of the actual odds?

My stats are a bit rusty, and I'm not going to show my work but it looks to me as though n in this case would be somewhere north of 450 and south of 1000.
Its been years since Stats 101 and 102, but I've no idea how you've determined that on average 750 OT games are a sufficient universe to determine whether first possession represents an advantage greater than/less than 48- 52% with a confidence level of 95%. While we have far too many variables that can not be easily accounted for, or modeled, on its face that # appears far too large.
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
Well if we really can't use probabilities because it's a human game, not a game of true probabilistic outcomes like fair coins being flipped (for example, how do we know the actual probabilities of the Patriots scoring an opening TD in overtime after they've scored on two straight possessions in regulation against the Chiefs, but with Gronk out due to injury and the backup right tackle in the game?), and we can't use anything less than 30 years' worth of data (being too insufficient a sample), how, exactly, are we to determine what's the most "fair" way of doing overtime?

I don't see how my thought process on the matter is any worse than anyone else's here. But as Bellhorn said....I really don't have a great grasp of the nuances of probability.
I think statistics can be used on a more micro level if the league is interested in doing so. There's probably enough data to start to get answers to questions about average points per drive, average points per starting position, average points per period of time, and you maybe even could break it down based on the amount of time left in the game.

Just because the sample size for game results alone isn't all that predictive or statistically significant, that doesn't mean you throw the baby out with the bathwater.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
I think statistics can be used on a more micro level if the league is interested in doing so. There's probably enough data to start to get answers to questions about average points per drive, average points per starting position, average points per period of time, and you maybe even could break it down based on the amount of time left in the game.

Just because the sample size for game results alone isn't all that predictive or statistically significant, that doesn't mean you throw the baby out with the bathwater.
Maybe you could do that. But then, each year is different, and what the stats were for the previous three seasons may not look much like what they will for the current year (witness the jump in NFL scoring from 21.7 to 23.3 ppg this year, for example). So again, I'm not throwing it out, but in every case where we cite the "average", it needs to be understood that this is *not* fair-coin flipping. These are human beings playing a game with a million variables and extenuating circumstances. Knowing the real probabilities seems...impossible. We can make guesses - educated guesses - but that's about it, right?
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
The problem with any per drive stats is that they are not based on the OT format, they are based on drives during regulation. There is some validity in modeling the probably based on % of TD, % of FG, etc., and those might even hold for the first drive of OT, but they won’t hold for a second drive if the team knows it has to score, has all four downs to play with, whatever.

I think we are all missing the point anyway. Clearly “fair” does not mean 50-50 outcomes. “Fair” means both teams get a chance - it is an optics thing, not a probability thing. I think the best proposal I’ve seen is to just play the full 10 minute OT, possibly as a continuation of regulation, and see where you’re at. Then for the playoffs you could make it 15 minutes, and if still tied, game continues from that point like a quarter change and next score wins.
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
Its been years since Stats 101 and 102, but I've no idea how you've determined that on average 750 OT games are a sufficient universe to determine whether first possession represents an advantage greater than/less than 48- 52% with a confidence level of 95%. While we have far too many variables that can not be easily accounted for, or modeled, on its face that # appears far too large.
I was mostly guessing based on binomial distributions.

Doing a quick standard deviation calculation for binomial distributions where p of of each result is .5, you get a standard deviation of about 11 where n=500. That's close enough for government work for me. It means that outliers (1.96 standard deviations from the mean of 250) would be roughly >230 or <270. So, where n=500, a record of 250-250 should reflect true odds of +/- 3 percent (47-53). So, 48-52 would be n a bit north of 500. Again, back of the envelope and I'd be happy to be corrected by a stat expert.

Again, my key calculation is that the SD where n=500 and p0 is .5 and p1 is .5 is 11 (square root of 500x.25). If I have that wrong, then the whole thing is wrong.
 
Last edited:

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
The problem with any per drive stats is that they are not based on the OT format, they are based on drives during regulation. There is some validity in modeling the probably based on % of TD, % of FG, etc., and those might even hold for the first drive of OT, but they won’t hold for a second drive if the team knows it has to score, has all four downs to play with, whatever.

I think we are all missing the point anyway. Clearly “fair” does not mean 50-50 outcomes. “Fair” means both teams get a chance - it is an optics thing, not a probability thing. I think the best proposal I’ve seen is to just play the full 10 minute OT, possibly as a continuation of regulation, and see where you’re at. Then for the playoffs you could make it 15 minutes, and if still tied, game continues from that point like a quarter change and next score wins.
If ultimately you’re going to take it to the point where first score wins, why not just do that from the beginning of OT? It’s not like they haven’t had 60 minutes to settle things.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
If ultimately you’re going to take it to the point where first score wins, why not just do that from the beginning of OT? It’s not like they haven’t had 60 minutes to settle things.
Optics. It would mean both teams got the ball at least once so they had a chance.
 

Bellhorn

Lumiere
SoSH Member
Aug 22, 2006
2,328
Brighton, MA
1. How many games represent a sufficient enough sample size?
DDB gave you a good answer on this one. Here is a good primer on how to do this calculation: https://en.wikipedia.org/wiki/Checking_whether_a_coin_is_fair

Unfortunately, you continue to miss the point in your response to DDB, by presuming that "the answer" must necessarily consist of observing real-world data. I have already shown you that we know from a priori considerations that winning the coin flip still confers at least a net advantage of 5% win probability.

2. What setup is the most "fair" statistically speaking?
I think I speak for most skeptics of the current system when I say that fairness, per se, is not the issue - rather, it is the fact that there is a win probability swing associated with a completely exogenous event. See the blog post that I linked earlier for more on that. Extending the fourth quarter, or even performing the OT flip at half-time, are good ways to resolve this objection, even if they are not strictly "fair".
 

bankshot1

Member
SoSH Member
Feb 12, 2003
24,652
where I was last at
DDB-thanks.

It is optics and viewed through a lens blurred by tears of those annoyed by the Pats winning.

FWIW first move advantage in chess (playing white) is estimated at ~52-55%.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
I think I speak for most skeptics of the current system when I say that fairness, per se, is not the issue - rather, it is the fact that there is a win probability swing associated with a completely exogenous event. See the blog post that I linked earlier for more on that. Extending the fourth quarter, or even performing the OT flip at half-time, are good ways to resolve this objection, even if they are not strictly "fair".
Simple solution to this seems to be that whoever loses the opening toss should get choice in OT. Generally speaking teams defer the opening toss because of the chance of a double score in and out of the half - they perceive a potential advantage. So if one team gets a slight advantage earlier in the game, it could make sense that the other team gets the advantage later (minimal as both are).

Anecdotally, I’d like to see data across the NFL on:
Winning % when winning the toss
Winning % when deferring the toss
Winning % when receiving second half kickoff

Edit: only thing I found so far was in 2015 teams were winning at a 54.9% clip when deferring. Obviously that is subject to SSS too. Plus confirmation bias, good teams might defer more, who knows. But if that stat is true and valid (again it might not be), then the opening coin toss of the game might have just as much of an effect on who wins, yet no one seems to have any problem with it. That’s curious.
 

Morgan's Magic Snowplow

Member
SoSH Member
Jul 2, 2006
22,345
Philadelphia
I think I speak for most skeptics of the current system when I say that fairness, per se, is not the issue - rather, it is the fact that there is a win probability swing associated with a completely exogenous event. See the blog post that I linked earlier for more on that. Extending the fourth quarter, or even performing the OT flip at half-time, are good ways to resolve this objection, even if they are not strictly "fair".
Football involves tons of luck. Why is luck stemming from completely exogenous factors (a coin toss) so much different than luck stemming from endogenous but plausibly random factors (a fumble recovery)? If there was evidence for the first possession conferring a massive advantage, that would be one thing. But win probability swings larger than 5-10% can easily happen in football for other largely-luck-related reasons (fumble recovery, a tipped ball happens to fly in one direction or another).
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
DDB gave you a good answer on this one. Here is a good primer on how to do this calculation: https://en.wikipedia.org/wiki/Checking_whether_a_coin_is_fair

Unfortunately, you continue to miss the point in your response to DDB, by presuming that "the answer" must necessarily consist of observing real-world data. I have already shown you that we know from a priori considerations that winning the coin flip still confers at least a net advantage of 5% win probability.


I think I speak for most skeptics of the current system when I say that fairness, per se, is not the issue - rather, it is the fact that there is a win probability swing associated with a completely exogenous event. See the blog post that I linked earlier for more on that. Extending the fourth quarter, or even performing the OT flip at half-time, are good ways to resolve this objection, even if they are not strictly "fair".
I certainly don't know how to do the calculation, but I honestly suspect that no one knows how to do the calculation for what the real-world probability is of one team winning the game by getting the ball first in overtime. We aren't flipping fair coins, as has been pointed out numerous times. There are so many factors that go into it. If you're the Patriots, would you rather have the ball first in overtime if:

1. Your offense is on a roll and playing a crappy defense in Foxboro?
2. Your defense has allowed just 3 points to a terrible offense and there's a 30 mph wind blowing in one direction?
3. Your offense is playing well but you've suffered injuries to Gronk and Edelman, and the opposing starting QB went down on the last play of regulation?

I mean, these are all things that factor in. Where you're playing, who you're playing, the health of your team, which team has the momentum, the quality of the offenses and defenses involved, etc. None of which is anything remotely like a random toss of a fair coin.

So honestly, how in the world can you calculate the probability of the Patriots scoring a winning TD on the opening drive of overtime? As others have pointed out, looking to regulation drive results won't do it, because overtime drives are a different animal. Health is a factor. The likelihood of the Pats scoring a TD if Gronk and Edelman are out is vastly different than if they're in but, say, Watt and Clowney are out for the Texans' D (who they may happen to be playing that day).

How do you calculate the probability of this? You're going by estimations based on regulation data, which may have very little to do with realities in overtime.

I totally totally totally get that you know more about this stuff than I do. So I'm not trying to be a dink. I'm just questioning the veracity of your process.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
Both teams have already had 60 minutes' worth of a chance.
And yet they already felt compelled to change the OT rule.

This is kind of like the CFP. You’ll never get a perfect solution. Right now the 5th team bitches. If you go to 8 teams the 9th will bitch. Every year in March a handful of teams bitch about not being in the field of 64. And on and on. But if they were to play a 15 minute full quarter OT in the playoffs, with neither team having scored, then a team wins by scoring on their first possession of the second OT quarter, no one will be able to say both teams didn’t have a fair shot in OT.

Optics.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
And yet they already felt compelled to change the OT rule.

This is kind of like the CFP. You’ll never get a perfect solution. Right now the 5th team bitches. If you go to 8 teams the 9th will bitch. Every year in March a handful of teams bitch about not being in the field of 64. And on and on. But if they were to play a 15 minute full quarter OT in the playoffs, with neither team having scored, then a team wins by scoring on their first possession of the second OT quarter, no one will be able to say both teams didn’t have a fair shot in OT.

Optics.
I totally agree with you. I think the current system is just fine. It's plenty fair enough. And - yes I know SSS but I don't really care - over the last 3 years the real-world outcomes are essentially 50-50 (tiny edge to the team with the ball first, but not because they score a TD first, which means that much more often than not the second team has a chance to win), so I see no compelling reason whatsoever to change it from what it is now.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
Agreed, I see no reason to change either. But if they were to change, I’ve outlined how I’d prefer to see it done.
 
Doesn't everyone who has two (or more) children know how to solve this? It's basically the same problem as dividing a cake fairly, no?

No kick off in overtime. Team A chooses the ball starting position. Team B then chooses which team starts with the ball. Then you play with the current rules.

If Team A really wants the ball, they can pretty much have it - just elect to start on the 1yd line - we can agree Team B would generally be better off giving up the ball then surely?

If Team B really wants the ball, they just choose to take the ball, wherever Team A places it.

In practice, you should end up with teams tending towards the equilibrium point where Team B is taking the ball half the time/defending half the time, and winning half the time/losing half the time (and where teams doing stupid stuff put themselves at a disadvantage).

Advantages;
- Simple enough that kids can understand it
- No need to calculate probabilities to try to make the rule fair - you leave all that up to the teams
- No need to change the rules as the game evolves - if teams see that drives are more likely to end in TDs they should be prepared to take the ball closer to their own line
- Doesn't meaningfully mess with the rules or make the game longer
- Any rule that allows teams to shoot themselves in the foot by doing stupid stuff gives BB a better chance to win.

What's not to like?
 

BigJimEd

Member
SoSH Member
Jan 4, 2002
4,432
I'll pass. NFL doesn't need gimmicks.
The complaints are about both teams not being the ball. Any proposal to change the rules needs to start there.
 

snowmanny

Member
SoSH Member
Dec 8, 2005
15,668
Yeah. The Patriots win the toss and call 6 yard line.

Just play ten minutes. If it’s a tie after ten minutes it’s a regular season tie. In the playoffs sudden death begins with the second OT and the team that lost the OT coin toss gets to pick offense or defense to start.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
Yeah. The Patriots win the toss and call 6 yard line.

Just play ten minutes. If it’s a tie after ten minutes it’s a regular season tie. In the playoffs sudden death begins with the second OT and the team that lost the OT coin toss gets to pick offense or defense to start.
Well in that case if the Pats call 6 yard line, and you decide to kick off, you can't complain about not getting the ball. It was your choice to kick off.

Why not just continue the 2nd OT with the current game state? Why re-start another game?
 

Al Zarilla

Member
SoSH Member
Dec 8, 2005
58,871
San Andreas Fault
Interesting new proposal by nobody :) today but I’ll always vote for a protocol that the majority of people will understand and remember and not be gimicky (I know, college). I bet even in Kansas City before the OT protocol was explained by the ref at the coin toss or by Nantz, people at the game or at home we’re asking “how does this work, is it sudden death, a fifth period, or what?” People would not remember this new one. Overtime is too seldom of a happening.
 

snowmanny

Member
SoSH Member
Dec 8, 2005
15,668
Why not just continue the 2nd OT with the current game state? Why re-start another game?
You could, but there is some desire to get the game done I think for TV, injuries, not exhausting the winner for the next playoff game if it goes 3 OTs or something. So you play ten minutes of normal football -that's enough to pretty much guarantee that each team touches the ball - then you shift to sudden death. Now it's possible the team that gets the ball first ("Team 1") in the first OT goes on one of those 7 1/2 minute drives like the Pats rolled out against the Chargers and Chiefs in the playoffs and Team 2 has to respond with a two minute drive; that seems like a bit of an advantage for Team 1. Or Team 1 ends up with two possessions to Team 2's one. It's never 100% fair. But if Team 2 survives all that the advantage shifts to Team 2 in the second OT and they have a chance to end it with a single score.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,058
Hingham, MA
You could, but there is some desire to get the game done I think for TV, injuries, not exhausting the winner for the next playoff game if it goes 3 OTs or something. So you play ten minutes of normal football -that's enough to pretty much guarantee that each team touches the ball - then you shift to sudden death. Now it's possible the team that gets the ball first ("Team 1") in the first OT goes on one of those 7 1/2 minute drives like the Pats rolled out against the Chargers and Chiefs in the playoffs and Team 2 has to respond with a two minute drive; that seems like a bit of an advantage for Team 1. Or Team 1 ends up with two possessions to Team 2's one. It's never 100% fair. But if Team 2 survives all that the advantage shifts to Team 2 in the second OT and they have a chance to end it with a single score.
No I completely agree with going to sudden death in OT2. I am just wondering why you have another kickoff instead of just continuing like it is from Q1 to Q2.
 

bowiac

Caveat: I know nothing about what I speak
Lifetime Member
SoSH Member
Dec 18, 2003
12,945
New York, NY
I'll pass. NFL doesn't need gimmicks.
The complaints are about both teams not being the ball. Any proposal to change the rules needs to start there.
The complaint isn't about both teams not having the ball; it's about the advantage being conveyed by the coinflip.

I love this idea. It is a slightly simpler version of my previous favorite idea which is that both teams submit simultaneous "bids" for where they want the ball. Whoever submits the less field favorable position gets the ball. For example, if the Pats say they'll take it at the 10 yard line, and the Chiefs at the 8 yard line, it's Chiefs ball. I like this idea by NobodyInteresting even more, since it preserves the same core dynamics of forcing teams to make a choice, but is even simpler.
 

bowiac

Caveat: I know nothing about what I speak
Lifetime Member
SoSH Member
Dec 18, 2003
12,945
New York, NY
No I completely agree with going to sudden death in OT2. I am just wondering why you have another kickoff instead of just continuing like it is from Q1 to Q2.
This would work as well, but then you lose some of the "end-of-game" dynamics at the end the 4th quarter. It sort of depends how you feel about 65 yard field goal attempts, hail marys, and prevent defenses vs. ordinary football.
 

snowmanny

Member
SoSH Member
Dec 8, 2005
15,668
No I completely agree with going to sudden death in OT2. I am just wondering why you have another kickoff instead of just continuing like it is from Q1 to Q2.
I liked the idea of a second opportunity to create a scenario where time is running out and a team has to manage the clock. That's the best part of football to me. But there's merit to what you're saying. Just play football and at a certain point (10 minutes, 12 minutes, whatever) the game either ends or shifts to sudden death if the game is tied.
 

BaseballJones

ivanvamp
SoSH Member
Oct 1, 2015
24,375
Doesn't everyone who has two (or more) children know how to solve this? It's basically the same problem as dividing a cake fairly, no?

No kick off in overtime. Team A chooses the ball starting position. Team B then chooses which team starts with the ball. Then you play with the current rules.

If Team A really wants the ball, they can pretty much have it - just elect to start on the 1yd line - we can agree Team B would generally be better off giving up the ball then surely?

If Team B really wants the ball, they just choose to take the ball, wherever Team A places it.

In practice, you should end up with teams tending towards the equilibrium point where Team B is taking the ball half the time/defending half the time, and winning half the time/losing half the time (and where teams doing stupid stuff put themselves at a disadvantage).

Advantages;
- Simple enough that kids can understand it
- No need to calculate probabilities to try to make the rule fair - you leave all that up to the teams
- No need to change the rules as the game evolves - if teams see that drives are more likely to end in TDs they should be prepared to take the ball closer to their own line
- Doesn't meaningfully mess with the rules or make the game longer
- Any rule that allows teams to shoot themselves in the foot by doing stupid stuff gives BB a better chance to win.

What's not to like?
This is a pretty cool idea. Yeah it’s a bit of a gimmick but it involves STRATEGY, and the more strategy, the better. There’s no luck involved there. It becomes a Vizzini vs. Man in Black battle of wits.
 

BostonWolverine

New Member
Dec 6, 2017
109
Ann Arbor, MI
Play it like a beer pong rebuttal. If a touchdown is scored in the first possession the other team gets one play from their own 25 to score and then has to hit a 40 yard xp to tie it (no two point conversion allowed). From that point on, it is sudden death.

Both teams get to touch the ball and the rebuttal should have a sufficiently improbable to not significantly alter the statistics. Plus, it could produce some must see tv. Though, if the pats were ever to lose on another miami miracle, I'd cry a bit.
 
Last edited:

uk_sox_fan

Member
SoSH Member
Nov 11, 2006
1,273
London, England
Advantages;
- Simple enough that kids can understand it
- No need to calculate probabilities to try to make the rule fair - you leave all that up to the teams
- No need to change the rules as the game evolves - if teams see that drives are more likely to end in TDs they should be prepared to take the ball closer to their own line
- Doesn't meaningfully mess with the rules or make the game longer
- Any rule that allows teams to shoot themselves in the foot by doing stupid stuff gives BB a better chance to win.
One more advantage is that you have one fewer instance of the injury-causing kickoff return - one that other proposed rule changes is trying to eliminate.

Give the team that won the opening toss the requirement to state the starting yard line (i.e. Team A) and the opening toss loser the right to choose whether to take the ball or defend (Team B). Rules about scoring a TD ends it and FG gives other team a chance stay as they are now but the argument that one team never had a chance to touch the ball now doesn't hold water because either they were Team B and could have chosen to or were Team A and could have chosen a harder starting position to make the other team defer.

Sorted.
 

sheamonu

Member
SoSH Member
Nov 11, 2004
1,342
Dublin, Ireland
I think the overriding concern for the league (actually, the networks) will be whether a regular season OT solution significantly extends game time. OT in the regular season makes exec's extremely nervous because it bleeds game times into the kickoffs for competing broadcasts. Playoffs less so, because times don't butt up against each other as much. So, i think the sample size you are looking for is not the outcome of a given contest (of course it is fairer to give both teams a chance to influence the OT outcome beyond a random coin toss, whether via guaranteed possession or ball placement) the determinative measure will be which alternative, at least during the regular season, results in reaching a result that is both fairer and quicker. That should be something that is measureable through a computer simulation, i would imagine.
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,280
AZ
I think the overriding concern for the league (actually, the networks) will be whether a regular season OT solution significantly extends game time. OT in the regular season makes exec's extremely nervous because it bleeds game times into the kickoffs for competing broadcasts. Playoffs less so, because times don't butt up against each other as much. So, i think the sample size you are looking for is not the outcome of a given contest (of course it is fairer to give both teams a chance to influence the OT outcome beyond a random coin toss, whether via guaranteed possession or ball placement) the determinative measure will be which alternative, at least during the regular season, results in reaching a result that is both fairer and quicker. That should be something that is measureable through a computer simulation, i would imagine.
I wouldn't mind at all if the NFL went to ties in the regular season. They already have increased the likelihood of ties with the 10 minute overtime and first possession not winning the game unless a TD. With 16 games a tie is still a significant thing. The chance of a tie also would possibly increase drama as teams that perceived a tie was or wasn't good enough played out close games. I know US sports don't really do well with ties, but with current overtime rules it seems as though the NFL has already crossed that bridge and brought tie games into the zone of possibility. Having a few more games end in ties would be ok.
 

sheamonu

Member
SoSH Member
Nov 11, 2004
1,342
Dublin, Ireland
I wouldn't mind at all if the NFL went to ties in the regular season. They already have increased the likelihood of ties with the 10 minute overtime and first possession not winning the game unless a TD. With 16 games a tie is still a significant thing. The chance of a tie also would possibly increase drama as teams that perceived a tie was or wasn't good enough played out close games. I know US sports don't really do well with ties, but with current overtime rules it seems as though the NFL has already crossed that bridge and brought tie games into the zone of possibility. Having a few more games end in ties would be ok.
Logically I agree. Once you accept the possibility of a tie at all overtime is a bit silly. If a football game is 60 minutes long and team A beats team B within those parameters why should team C be able to come along a week later and get 70 minutes to accomplish the same task? If all those teams are battling for a division title C's win was unequal. (This would not be true if ties weren't possible at all, but once they exist giving a team extra time to avoid that outcome creates inequalities). But, illogically I understand that ties suck, so we have overtime. Making that choice doesn't threaten civilization as we know it, but I have an inherent bias against deciding a game using rules significantly different from those that resemble the game proper. That's why penalty shootouts, while entertaining, are unsatisfying - the World Cup, the planet's biggest single sporting event ends up decided by a free throw contest. That's why I'd prefer not to use ball placement procedures or "touchdowns end games while field goals don't" rules. The purist in me says give each team equal possession rights under standard rules until one team wins.
 

trs

Member
SoSH Member
Aug 19, 2010
525
Madrid
I wouldn't mind at all if the NFL went to ties in the regular season. They already have increased the likelihood of ties with the 10 minute overtime and first possession not winning the game unless a TD. With 16 games a tie is still a significant thing. The chance of a tie also would possibly increase drama as teams that perceived a tie was or wasn't good enough played out close games. I know US sports don't really do well with ties, but with current overtime rules it seems as though the NFL has already crossed that bridge and brought tie games into the zone of possibility. Having a few more games end in ties would be ok.
I also don't mind if the NFL went to ties in the regular season. It makes sense for player safety, reducing game lengths, and is the most "fair" and "pure" of any overtime scenario. I also get that US viewers are not accustomed to watching competitions that end in ties, but I also wonder whether we would see fewer ties if teams understood there was no overtime to play for.

Football is a timed sport, the final score should be what the score as at the end of the agreed length of the game. Unless there is no way to account for ties (like in a single game elimination playoff), why not have a tie be a possible result?
 

joe dokes

Member
SoSH Member
Jul 18, 2005
30,240
Optics. It would mean both teams got the ball at least once so they had a chance.
I think this is what it's all about. The results and probabilities are what they are, but the league seems driven only by not wanting a game (or at least a playoff game) to end without both teams having a guaranteed possession (as opposed to just a chance at one by playing successful defense).
I disagree simply because defense and ST is as much of the game as offense.
 
I wonder if having no OT in regular season would make it an easier sell (from a player safety/wear-and-tear perspective) to have full overtime periods in the playoffs. I think the biggest obstacle to the latter is television, and the risk of overlapping games, but certainly playing for 10 or 15 full minutes would be fairer than what we have now. Or maybe you could have a 15-minute OT period which continues to completion unless specific criteria are met - e.g., these could be:

--Both teams have had at least one full possession and one team leads by at least 6 or 7 points (so that if the first team scores a TD, the second would have to score at least a FG to stay alive).
--One team leads by 8 points (so a team can win the game in one possession with a two-point conversion).
--One team leads by 9 points.

(These would function like a sort of "mercy rule", I suppose.)