Note: You may have already read some of this in my contribution to Peter King's Monday Morning Quarterback column this morning. You can hear Dan Patrick and Peter King talk about the topic at this link (about 25 min into the 40-minute clip).
First, some facts.
Over the past decade, there were 158 OT games, including playoff games. There were 2 ties, and there was 1 game in which the coin flip winner chose to defend a side of the field rather than choosing to receive. (They lost.)
In 96 of the 158 OTs, or 61%, the coin flip winner won the game.
In 58 of the 158 OTs, or 37%, the coin flip winner won on their first possession while the loser never touched the ball. This includes 2 of the last 3 OT games in the playoffs.
Don’t be fooled by other numbers. In 2009 there happened to be only 13 OT games, and the coin flip winner won 7 (54%). In 6 of the 13 (46%), the loser never touched the ball. The sample size for any single year is too small for a reasonable estimate of the true numbers.
Also, don’t be tricked by people that say “only 61%.” If we agree 50% would be the fairest rate, you might think 61% isn’t very far from 50%. But that’s not the right way to look at it. As I wrote in my recent NY Times post, the correct comparison is 61% vs. 39%, the respective winning percentages of the coin flip winner and loser. That’s a big advantage--over 3:2. Another way to think of it is that the coin flip winner will win half again more often than the coin flip loser.
I've written previously that I favor a simple solution, which I first heard from economist David Romer: Return the kickoff line to the 35 for the OT kickoff. This is a modest change that will greatly increase touchbacks, forcing offenses to start at the 20. And in OT, a 1st down on the 20 yd-line (and not the 15 as previously thought) appears to be the break-even point between where the team on offense and the team on defense are equally likely to score next. I'd guess that teams may tend to play "tighter" in OT than in regulation, and this is why the break-even point becomes the 20.
The graph below plots OT Win Probability (WP) by 1st down field position. In OT, with the rare exception of the final meaningful possession in an eventual tie, time is not a significant factor.
The data is a little noisy, but the unscientific fat red crayon technique makes it fairly clear that the 0.50 WP line intersects at a team's own 20 or so. The 0.61 WP line, which represents the current expected winning percentage of the coin flip winner, intercepts the red line at about the 33-yard line, which is where the average return ends up (including penalties and TD returns).
Moving the kickoff line back to the 35, where it was when the current OT format was instituted, might go a long way toward equalizing the chances of the coin flip winner and loser. Unfortunately, that's only half the problem. Over one third of OT games result in one team losing the game without ever touching the ball. I think that's the bigger issue to many people.
Moving the kickoff line would reduce the likelihood that one team would never get the ball, but only slightly. I don't think we can solve that part of the problem without a rule requiring an even number of possessions. But such a solution causes a whole new problem.
However the specifics of the even-possession rule would work, the team with the second possession would have an even bigger advantage than the coin-flip winner has now. Knowing whether or not it needs a TD, FG, or can afford to punt, the second team can adjust its strategy accordingly, using its 4th downs to move the ball when necessary. The current college OT format has this problem, which is mitigated to some degree because teams alternate 'going first' on successive rounds. Even baseball's extra innings format has the same problem to a lesser extent, but it's accepted because it's no different than the advantage the home team enjoys in a tied 9th inning. It's simply considered part of home field advantage in baseball.
There is no perfect solution for NFL's overtime problem, but perhaps baseball's system illuminates the best path. Maybe the best we could hope for is to keep the current sudden death format, but award the first OT possession to the home team.
With this arrangement in the playoffs, it would be easier to accept what happened to a visiting team like the Vikings. We would say, "Yeah, it stinks their offense didn't get a chance. But hey, the Saints did earn the right to the first possession by winning home field during the season." Better that a team earns a break than has it bestowed by the flip of a coin.
I've recently been thinking that "home team gets the ball first" is a pretty simple, short-term fix to the OT problem. I'm not sure how someone can be in favor of the coin flip but not "home team gets ball first", although I'm sure someone would be. That sounds like as reasonable a tiebreaker as any, and it at least lets the visiting team know beforehand that they won't have the ball first in OT.
For a more intricate plate, Jason Lisk talked about the "Last Clear Chance" theory of OT; link in my profile name.
What bothers me is that you can win in OT with just a field goal. After the kickoff, that often involves driving the ball less than 40 yards. Instead of sudden death, how about the first team to score 6 points?
"If we agree 50% would be the fairest rate, you might think 61% isn’t very far from 50%."
You could argue that ideally home teams should win OT at the same rate that they do for complete games.
"Maybe the best we could hope for is to keep the current sudden death format, but award the first OT possession to the home team."
I'm pretty sure if you break out the stats for "home team wins coin toss" you'll see that the home team wins way more often in OT than they do overall. I heard Colin Cowherd on the radio recently state that when the visiting team gets the ball first that they win about 50% of the time. He was using that as justification for giving the visiting team the ball first in the current sudden death OT format.
Personally, I like the college football overtime; the main problem I have with it is that it inflates offensive stats (lots of potential scoring in OT with short fields). But it certainly is entertaining. Or how about they just play the full extra OT period with no sudden death, kind of like in basketball?
I have favored the even number of possessions. The first team with the ball will try harder to get a touchdown knowing that if ythe get a field goal the other team on their possession can still win with a TD.
Dave-Actually, home field advantage is strongest in the first quarter, then decreases throughout the game until it is undetectable in overtime.
I like the lead-and-possession or first-to-six-points proposals the most.
Sudden death to 6 points wins.
During the regular season you get one 10 minute overtime. If no one scores then we end in a tie.
During the playoffs you play until someone wins.
- Sean
Why not advance the kickoff line to the 45 yard line, or even to midfield? This gives the kicking team options.
A) Blast the ball into the end zone and guarantee the 20 yard line.
B) Work on a high kick to try to pin the other team inside the 20.
C) Try an onside kick. From midfield, giving the other team the ball at the 38 is not hideous (30-35% WP estimation). And if you get it, your WP is about 80%. It still looks negative overall, but it means someone who wants to try for the win has a chance.
What's amazing about all this is that I've heard literally dozens of ideas (nobody on this thread has mentioned my favorite--having the teams bid for field position), and every single one of them sounds better than the current system. Any chance the NFL is just too overwhelmed by good ideas to pick one?
No...probably not.
Adam--that's the first time I've heard that one, and I love it.
I love the idea of bidding for field position, but absolutely do not trust at least two thirds of NFL coaches to do it sensibly.
My preference is and probably always will be for the college OT format. (Or perhaps a modified college OT format, maybe starting a little farther out.) The reason has nothing to do with fairness. Life's not fair -- get used to it. I prefer the college format because it's a gajillion (technical term) times more exciting. I generally prefer NFL football to college football, but if an NFL game not involving the Browns goes to OT, I sometimes still turn it off if something else is going on. If a college game goes to OT, I'm going to do everything in my power to watch every snap, regardless of who's playing. To my mind this is the only argument worth having. Nobody I know will turn off a college OT game, period.
Of course coaches wouldn't do it sensibly--that's half the fun. Many of them already fail to handle fourth down decisions, timeouts, challenges, and whatever else sensibly, and presumably would continue to do so in overtime. To me, the ideal overtime system is one that doesn't introduce any potentially game-changing factors that weren't present in regulation, and the relative strategic abiliity of the two coaches is certainly a factor that was already part of the game for the first 60 minutes.
And I couldn't agree more with your second paragraph (well, replace Browns with Falcons and I'm 100% on board, but that's a pretty trivial distinction). I also prefer the NFL to college, and I also have turned off NFL games in overtime because I just don't care to see a game-winning 30-yard field goal.
(Incidentally, the start of my sister's wedding was delayed about 20 minutes because several people with key roles in the ceremony were watching an overtime game involving the local college team. It's hard to imagine that happening with NFL overtime, outside of the playoffs.)
I'd prefer for them to keep it simple, and as much like regulation as possible. So: get rid of sudden death. Regular season games get no overtime (they just tie), and postseason games play a full period of football (possibly a 10 minute period instead of 15) and whoever is winning at the end wins; if it's tied they play another period. That gets rid of the "no possession" problem, and the "just kick a fg and win" problem, although the coin toss still matters (possibly they could also move the kickoff back to the 35).
This is a minor quibble, but I basically disagree that the structure of baseball gives the home team any distinct advantage. Yes, the home team doesn't need to bother hitting in the ninth a large percentage of the time, but that is only because they scored more in earlier innings.
There are certain things that a team can do when they know they need only one run to win (sacrifice bunts, suicide squeezes, etc). This would seem to give the home team a big advantage. On the other hand though, there are things that defenses can do (ie moving the infield up, outfield up) when they only care about stopping one run. I believe the defensive adjustments are actually more effective in changing the game when it goes from as-many-runs-as-you-can to only-one-run counts.
Baseball has the lowest home field advantages of any sport. Basketball has the highest, despite having no structural differences and no differences in the playing surface or dimensions. I think baseball's HFA derives solely from the psychological factors common to all sports and not from the inning structure.
We've talked about the auction idea in previous threads. Its definitely my favorite. Its quick. Its fair. Its dramatic. I have no idea why anyone would be opposed.
The NFL will review all of these proposals and instead rule that the coinflip will be used from here on out to just randomly determine the winner. No more token competition to confuse things. The game winner will be declared when the coin hits the ground. Problem solved.
A tie in the SuperBowl is different though and will be broken using a fancy HD slo-mo coinflip on the jumbo screen. Tails will be Doritos and heads will be Bud Lite. The flip will be programmed to last for 10 minutes with one TV timeout so the fans can really get involved in the process of cheering for either heads or tails to determine the best team of the 2009 NFL season.
So Brian, if you break down the OT coin flip winners by home and away teams, you would find that each group won about 61% of the time?
That contradicts what I had heard, but I'll believe your stats if you share the numbers on that. From what I had heard on the radio, it sounded like when the away team wins the OT coin toss, they win about 50% of the time, but when the home team wins the toss, they win about 70% of the time. Maybe they were citing bad stats or using a small sample size.
Great article, and I wholly support both measures presented, although should it be home team "gets the choice" rather than "gets possession"? Moving the kickoff line to force more touchbacks may make some teams choose to defend. Either way, this is a much better system. Imagine you're the away team and you just scored a last second TD to go 1 point down. We'll see far more two-pt conversions.
The only query is the Superbowl. Do you give it to the team with the better record?
First, eliminate OT in regular season games. Sprinkling a dozen or so ties into win loss records will help reduce the use of the far worse end of season tie breaking procedures.
The changes it will force on coaching decisions will also be positive.
As previously pointed out in Pro Football Prospectus, any team that feels that losing the OT coin toss is unfair has only their kickoff kicker to blame.
In fact, having a top kickoff specialist may be one of the inefficiencies that the smart teams have been able to take advantage of, whereas the dumb teams haven't figured out yet. Once they do, the current system will look a lot more fair.
Anon, interesting point. So you're essentially saying that changing the OT rule would be like banning going for it on 4th down (or any other "smart" strategy) because the teams that do go for it have an unfair advantage.
I'm not sure 'get a kickoff specialist' is a logical answer. Even kickoff specialists rarely get touchbacks. Plus, deeper kicks are a double-edged sword. You need 2 yds of depth to net 1 yd of benefit on the return because the coverage team is that much further from the play. The returner has more time and space to accelerate and find a hole.
suggestions i like the best:
1. field position auction
2. move kickoff (or have one team start at the 20)
3. leave it
home field advantage is more recognizable in baseball because innings are generally short. but in the NFL, the team that gets the ball to start the second half also has has advantage of knowing the situation and having more possessions to do what they need to do.
that's why i don't really like the alternating possession idea. it's also why i don't really buy the "one team never touches the ball" argument: each team gets the ball 11-12 times during the game, and who's to say their number of possessions will be even then? we say it's okay if a team has one or two more possessions over the coure of the game; that's just good clock management. but if it happens in overtime, it's just luck. if coaches don't want to leave the game up to chance, how about they take control of it and have some balls when calling plays in regulation? i think if you play for the tie instead of the win, you deserve a 50% chance of losing.
or how about this: add an "in-case-of-overtime" clause to the beginning coin toss: if you choose to receive in the 1st half or defer to the 2nd, you not only kick off in the half you don't choose, but in OT as well. if you choose an end zone instead, you get the ball to start OT.
"However the specifics of the even-possession rule would work, the team with the second possession would have an even bigger advantage than the coin-flip winner has now."
IMHO, this is the best point in your article. Many people are clamoring for an each-team-gets-a-possession solution but most of them are ignoring the fact that this doesn't solve the problem of inherent advantage - it only shifts the advantage from the team that gets the ball first to the team that gets it second. Knowing how you must play is a huge advantage.
It might help some to think of a gambling analogy. When I lived in Vegas I was at first confused at how the house could win so consistently in blackjack. After all, the dealer plays the exact same game as the people seated at the table. But the key is that the dealer plays LAST. The players are forced to play with imperfect information. If they knew that the dealer would bust, they would all stand pat or if they knew that the dealer had 20 they would all hit until they had 21 (or busted).
"if coaches don't want to leave the game up to chance, how about they take control of it and have some balls when calling plays in regulation? i think if you play for the tie instead of the win, you deserve a 50% chance of losing."
This argument insinuates that, with the correct play-calls and the appropriate strategy, a tie (in regulation) can be avoided. However, even if there were two "perfect" coaches who always made the best, most aggressive, most tie-avoiding decisions possible, there would still be games that end regulation with a tie score.
Please explain to me why a system in which the coin flip winner wins the game without the other team touching the ball 37% the time is derided as "determining the game by a coin flip". It seems to me that a game being "determined by a coin flip" would see the winner win at least 50% of the time without the other team touching the ball.
Also, that 61% statistic seems misleading to me. You say that 58 times a team won without letting the defense have the ball, while 96 times the coin flip winner won overall. So that means there were 38 games in which both teams had its offense on the field. Why are these games even revelant to this discussion? If the supposed disparity is ball possession, then the first time a coin flip winner punts (or turns the ball over) that disparity no longer exists. Both teams have had their chance.
Defense and special teams are part of the game. Maybe they are not the most exciting part, but they compose 2 thirds of the team. The idea that defenses and special teams are helpless in the face of an offense that wins ball possession seems a bit misguided.
Face it. Peter King was just mad he couldn't coast on his sycophantic Bret Favre columns for two weeks.
Anon-That's not the correct way to look at it. Zero % would be optimally fair, not 50%. You're in good company though. I saw Judy Battista make the same point on Sports Reporters on ESPN Sunday. Luckily, Mitch Albom (of all people) called out the mistake.
Plus, even if both teams do get one possession, and neither scores, the problem starts all over again. So counting up games in which both teams get at least one possession is not that meaningful. That's why coin flip winners end up with a 3:2 advantage even though they only score on their first possession 37% of the time.
If it helps, think of it from the opposite direction. Would you be in favor of changing MLB's extra innings format so that the game ends when the visiting team scores in the top of an extra inning? Like most people, I'd bet you'd say no. And your reasoning would be exactly the same reasoning that says the current NFL system is broken.
@Alchemist:
that's true. my point was simply that most coaches don't... they don't play to win (contrary to herm edwards' opinion), they play not to lose.
the point of OT is that after 60 minutes, two teams have proven themselves to be equals on that day. the reason for the coin toss & sudden death is to just get it over with and determine a winner, even if it's not inherently fair, because we apparently all think ties are for soccer, that they're boring. it's like a hockey shootout: just end it already.
i just don't see how a team can really complain when they have 12 chances on offense and 12 on defense, that it's "not fair" they didn't get the ball last. the more i think about it, the more i see the randomness of overtime as a GOOD thing, because it pressures coaches - or at least it should - into make more decisive choices in regulation.
to me it's the same as "settling" for a 50-yard field goal at the end of the game instead of gambling a little to get closer... the coach can then blame a loss on the kicker, instead of himself. likewise, a coach can make absurdly conservative calls in regulation to force overtime, and then blame the coin toss & the system if it doesn't go his way.
Brian - if I can add to the reply to Anon. You also have to consider field position. If a team receiving a kickoff in OT gets a touchdown, moves 30 yards to midfield then is forced to punt to team receiving the punt is more likely than not going to start inside their own 20 - so even though both sides get the ball the two starting yard line means the situations are not equal.
From the stats point of view, the team receiving the kickoff has a 61% (96/158) chance of winning. If the receiving team is forced to give up possession this switches and the team that kicked off now has a 62% chance (reverse of [96 - 58]/[158-58]). Almost looks fair, until you consider that in order to have a 60% chance of a win
- one team only had to win a coin toss
- the other team had to make a stop on defense
Yes, defense and special teams are part of the game, too. But the NFL has spent better than 30 years changing rules to encourage kick returns and open up the offense (kickoff from 30, legalized holding, no contact after 5 yards). Meanwhile, the OT setup hasn't changed.
I personally like the auction idea--not sure how workable it is, I just want to see how some coaches handle it. One of my favorite moments in any broadcast is when the announcers say "I don't know what Andy Reid was thinking there." Always high comedy.
The funniest thing about this whole debate is how nobody has realized how the network TV environment has changed. Forty years ago, football was basically filler. TV put it on the air because they didn't have anything else to put on the air on Sundays.
Their biggest desire was that the games not interfere with their valuable Sunday night programming. The famous Heidi game was the biggest manifestation of this. After Heidi, the networks realized that they needed to stay with football until the end but they still weren't entirely happy about it.
The NFL didn't approve overtime for regular season games until 1974. I'm pretty sure that they were keeping the networks in mind when they wrote this rule. I can imagine somebody calling Rozelle and saying "Whatever you do, make sure it ends quickly"
The TV environment is drastically different now. Football is a crown jewel and draws much higher ratings. The networks also pay ungodly amounts of money in order to show it. The logic should be completely reversed. The networks want longer and more interesting games. If they have to bump something else to show an overtime thriller, they really don't care.
The difference between the best and worst kickoff coverage teams is what, about 7 or 8 yards? You'd never have the sample size to prove it empirically, but I'd bet that the best can get the overtime odds pretty close to 50/50 after losing the toss just by pinning them down on the 21 or 22 yard line, especially if the break even point in overtime is the 20.
I'm guessing the smart teams (a) already know this; and (b) vote against changing the overtime rule because it would eliminate their advantage.
I personally don't find the current OT system to be that much of a problem. Many people complain that this format places too much significant on kickers and special teamers. But the fact of the matter is that your offense didn't get it done during the 60 minutes of regulation.
It could be likened to baseball. If after 9 innings, your offense didn't get it done and the game is still tied, then you'd go into extra innings having to rely heavily on your relievers. Doesn't the baseball format place just as much importance on relievers during extra innings as the football format places their importance on kickers and special teamers?
Relievers are just as important to a baseball team as kickers and special teamers are to a football team.
OT should just be like transitioning from 1st-2nd and 3rd-4th quarters. No new flip, whoever has the ball has it at the field position they have.
I think the biggest problem with the current overtime system is the ability for one team to win the coin toss, drive 40 yards and kick a field goal to win. If a team wins the toss and is able to score a TD off opening possetion then I feel like the deserve to win the game.
An easy way to fix this problem would be to narrow the goal posts to something like goal posts in the Arena league. In addition to helping fix the overtime problem, this would make regulation more exciting as well. Field goals would not me as dependable as they currently are which would result in more teams going for it on 4th down. And I do believe that going for it on 4th is what most fans would like to see more of seeing how on any 4th and short where a coach elects to punt/kick a FG the crowd almost always boos.
The proposal that always seems to get short shrift in these discussions is the idea of a minimum number of points to win in OT. It was mentioned here three times and just brushed aside by everyone else. I still don't understand why this solution doesn't get more consideration in any discussions of OT.
It solves the major problems of the current system, and is consistent with OT in other sports (tennis, volleyball) which require a minimum lead to win in OT.
The major problem is the unfairness of the short drive+FG where the other team never gets the ball. This takes away that scenerio, while still allowing a team to win decisively with their first possession on a TD. That alone should be enough to alter the current 60/40 inequity.
Though, instead of 6 points, I would suggest the first team to score 4 or more wins sudden death (seems 2 safeties is a decisive statement), or the team that's ahead at the end of OT.
The reference to baseball miss the essential point of baseball extra innings--both teams receive 3 outs to spend. Nothing the opposing team does can keep a team from using all three outs. So I think all the baseball discussion does is show what attributes of a fair system might be--both teams start with equal opportunities and have to play both offense and defense.
Les
A OT format for the NFL has to be fair, satisfying, and relatively quick. The current sudden death format is generally quick but does not meet the other requirements. The various ideas offered about moving the kickoff spot, placing the ball on the 15 or bidding etc. may be fair, but do not seem satisfying. I don't think we would think a 100 yard touchdown run off the OT kickoff in the Super Bowl would be a satisfactory end to the game.
Meeting all 3 requirements isn't easy. The NFL could follow basketball, soccer, and other sports and play a pair of shortened halves--say two 10 minutes Stop play after 10 minutes, change ends and have a new kickoff. Seems to meet fair and satisfying but not quick (adds at least 20 + minutes). Perhaps the best two possibilities are one 15 minute full quarter or team that loses the coin toss gets one possession. Both these could fail to meet the quick requirement. So we could end up right where we are now--sudden death with winner of coin toss having big advantage.
Or the NFL could do what some soccer leagues do for some tournaments--replay the game on another day (now that would be fun).
Les
I wonder if the NFL's refusal to add a shortened overtime quarter (like NBA) or giving possessions to both teams (like NCAA football) is due to the players union's unwillingness to drag the game on longer, hence make them more susceptible to injury.
Does anyone know about this? Does the current OT format hold because NFL players themselves just do not want to play more?
In my opinion, there IS a perfect solution - treat fourth quarter to OT like first quarter to second or third to fourth. If the game is tied, just keep playing. First team to score wins.
This gives an enormous advantage to the team with the ball, of course, but they've earned that advantage by virtue of the flow of the game.
The only argument I've heard against this is that it takes out the drama of the last-second drive in a tied game. I have three distinct counter-arguments:
1) Most late-game dramatic drives aren't with the score tied; they're with the offensive team down by between 1 and 7 points. Those games will be just as dramatic under this OT rule.
2) There's actually a lot of drama introduced by this rule. A team that's down seven and scores late in a game is much more likely to go for two, because they don't have any chance of winning a coin toss for the ball. A team down three late in the game might go for the TD in stead of settling for a field goal that gives the ball back. These all-or nothing 4th down and conversion plays will add a lot of drama.
3) If a team that gets the ball late in a tied game is able to mount a clock-killing drive, taking the game close to or into OT and then winning, that's just good football. We reward that sort of skill all game long, and having a few endgame scenarios that reward the ability to grind the clock is not such a terrible thing.
@Tarr:
Why even have halftime then?
My beef with your suggestion is that the team down one TD late shouldn't have to worry about the "fifth quarter." The game is 60 minutes long, and if each team were to score on every possession and one team scores with 10 seconds left to tie it, the fact remains it was an EVEN game and should continue into overtime as such. With this rule you'd be almost forcing that team into going for 2 to win outright, and while it might be exciting, it's no more fair than a coin-toss.
Ryan, you have halftime to balance the expected number of possessions for each team. This remains true even in an overtime game, if you continue through. Due to uneven possessions, there's essentially no correlation between starting with the ball and having the ball after two quarters.
In fact, by having a new coin toss, you are just creating an unbalanced number of expected possessions. The problem you seem to think I'm causing is, in fact, a problem I'm _solving_.
To clarify, by "uneven possessions", I mean uneven time elapsed per possession. The point is that it's impossible to project who will have the ball at the end of the second and fourth quarters until you get close to the end.
The equivalently fair approach to what I'm proposing would be to have TWO kickoffs in OT; e.g. to have two 5 minute periods, each starting with a kickoff by one of the teams. This would also balance expected possessions per team, but at the expense of drawing the games out much further.
I did some digging into the subject matter. Here's a debate from NFL.com a couple of years ago.
http://www.nfl.com/news/story?id=09000d5d80f5307a&template=without-video-with-comments&confirm=true
My suspicion earlier was correct. The OT system was not implemented to appease fans like you or me. The OT system was created to minimize the injury risk of the players.
Quote from Terrell Davis, "As a former player, I love the rule as it stands now because I don't wanna be on that football field any longer than I need to be on that field."
I think that sums it up. Most NFL players would probably feel the same way. Given how short NFL players' careers are relative to other major sports, I think Terrell makes a valid point.
I know people hate ties, but instead of artificially declaring a winner, wouldn't it just make more sense to have a game that is tied after 6o mins end as a tie. I think this is the fairest result.
The best option is to simply ban field goals in overtime, forcing each side to score a touchdown. This will create excitement when a team reaches the red zone, knowing its either touchdown or turnover on downs(punts being meaningless). It also makes the coin flip more of a 50/50 proposition: you have 70-75 yards to stop the opposing offense, instead of just 45-55.
To win every match.This is the really reason i think.
Here's what you do:
Leave the OT rules exactly as they are, with the following exception. If a team kicks a field goal in overtime, we will add precisely one more possession to the game. The other team will get the ball one last time, and it's touchdown or bust. If the other team scores a touchdown they win, and if the other team fails to score a touchdown they lose. No matching field goals, no additional ties, nothing. Just one last drive for all the marbles.
If the first team gets the ball and scores a touchdown, they win, but at least they will have earned it, unlike a wimpy 40 yard field goal aided by bogus penalties that we saw in the Saints game.
Elad, that is the best solution, really, but it's unsatisfying to the general public. And you still need some alternative for the playoffs.
Again, just keep playing the game. No coin flip, no kickoff (unless a tying score happened as time expired, in which case you start with a kickoff by the team that scored). This is entirely fair and solves the problem without extending the length of the game or altering the relative importance of various aspects of the game.
Biran:
Do the studies in the ovarytimes and teh increased of the improved FG %s from the 60% to the 80% consider what the was of the impact of the changes of the football laws from a touchback of a missed FG to returning the balls to the sppot of the kick?
I am also impressed of that a 38 year old man is so retired at such an age. In my country we have problem and the problem is sociAl securities to the very old peoples. (Those 40 and older.) You will go to Allah soon, yes?
Tarr, your solution might actually cause a bigger problem because it would lead to more OT games & headaches for the TV networks. Why would a team try a (hopefully) game-winning FG in regulation if there was any chance of the opposing team getting the ball one final time? Why not wait till OT when there's no chance of that.
As someone pointed out earlier, does anyone not think that ALL of these potential ideas (or the vast majority) are better than what we have:
Play a 10 or 15-minute 5th quarter.
First to 6 (or 4 points) wins
Eliminate field goals altogether
Change the kickoff position
Bid for field position
Keep playing like after 1st or 3rd quarters
All of these seem preferable to the equivalent of ending a baseball game when the visiting team scores a run in top half of the 10th. I believe that the players may not want to keep playing, but that doesn't make the system legit.
Last week at work we were talking about "requirements-based" acquisition. The Iraqi military gets funding from the US, and they want to buy stuff. We ask them what they need, and they send us to a web link to some new cool patrol boat or helicopter that they want.
We say, 'Whoa. How do you know this is what you need? How long does it need to be on station? How many troops does it need to carry?' In other words, start with your mission requirements, then find something that fulfills those needs. Don't fall in love with something and then rationalize why you need it.
It's no different with OT. We all are in love with our own ideas, but let's look at it from the NFL's point of view. What are their requirements?
-Decisive
-Must end as quickly as practical
-Fair and non-arbitrary
-Must be "true" football, i.e. incorporate all phases of the game
-Must not encourage 'playing for a tie' or 'playing for OT' (like the NHL)
-etc.
Now we examine all the alternatives with regard to how well they meet the requirements, and choose the best one.
I'm a fan of both teams get 1 possession and then sudden death. This offsets two different advantages. First the team receiving will still have first chance in sudden death. Offsetting this the other team knows this, and going second should try for TD (over FG) or perhaps 2 pts over 1. Also if the 1st team scored a TD 2nd team can use all 4 downs. The main point is, neither team has an excuse (unless going 2nd proves to be too strong of an advantage).
Or alternatively, but reaching similar goals would be determining who will receive the OT kickoff before OT. i.e. home team always receives. In which case the visitor has all game to account for this and its their own fault if they play for a regulation tie.
@Anon, why don't we look at my proposal (i.e. continue the game in OT without a coin flip or kickoff) methodically. If you have the ball and you are...
Down 8+: No change in strategy.
Down 7: More likely to go for 2. Reduces the frequency of OT.
Down 6-4: No change in strategy.
Down 3: More likely to attempt to get a TD. Reduces the frequency of OT.
Down 2-1: No change in strategy.
Down 0: No longer in a rush to get the FG off in time - play as if there is plenty of time left. Increases the frequency of OT.
So, of the three situations that are impacted, two reduce the frequency of OT, and one increases it. "Playing for OT", as Brian says, is actually discouraged in aggregate - we take away two situations where it happens currently and introduce only one.
(And it's unquestionably decisive, fair and non-arbitrary, "true football", and it's as quick or slightly quicker than what we have right now.)
Tarr- Aha, yes I didn't think of the "Down 3" scenario. I think I'd be in favor of your format for postseason games but would prefer ties for regular season.
Chris - I have one small problem with the "both teams get one possession" idea... what if the receiving team drives down and scores and then pulls off a successful onside kick? Or if they kick off normally and the return man fumbles the ball back? Is the game over at that point? Seems like an unsatisfying way to end it.
@Anon - I actually agree. If it were up to me, I'd go for ties in the regular season as well. However, given the sports culture in the USA, I think a drastic increase in the number of ties is a non-starter. As such, I think the "continue the game" concept is the best idea _that has any chance of getting implemented_.
One of the strongest arguments for this approach in my mind comes from thinking backwards from after it gets implemented. I think that if this were the actual system that had been in place for years, nobody would even think about changing it or tweaking it. The idea of taking out "continue playing the game", and putting in "have a coin flip, then a kickoff, but if the first team to receive the kickoff kicks a field goal, they must kick off to the other team", would seem ridiculous. By contrast, "continue the game" is a very simple and intuitive approach. If it were in place it would be dogma.
For the record, this is not originally my idea - I read about it in some comment thread years ago. Probably on Footballoutsiders, although I'm not certain. It immediately seemed like a good idea to me, but it took a while before I became a strong advocate/agitator for it. It really doesn't have a significant downside.
RE: Baseball comparisons.
Each team playing offense makes sense in baseball because while defense controls the ball it is unable to score runs.
RE: Field position.
Again, I'm nonplussed by the underlying assumption here that kickoff coverage and defense are helpless in the face of an offense that only requires a field goal. If special teams is so inherently unfair, then why kickoff at all? Why not merely spot the ball at the 20 yard line after a score?
RE: Statistics.
Isn't it erroneous to assign probability based on past frequency? I mean, with the millions of variables occuring at play in every game, it seems like a gross simplification to say that the probability of the coin flip winner being 61%. The actual probability is actually unknown, or is at least dependent on the team's offense, defense, and special teams, weather, what they had for breakfast... etc. Isn't claiming 61% probability based on the 198 games like this: I ask you to pick a card from a standard deck 200 times. You pick a Jack 50 times. The probability of the next card you pick being a Jack does not BECOME 25%. It remains 4/52, or 7.7%.
Also, let's be rational people here and avoid things like availability heuristics. You claim that games in which both teams get the ball and neither scores starts the problem over again. How, exactly. You're looking at a situation in which a team stopped another team from scoring, gained ball possession, failed to score, punted, and then allowed the team to score, and you're attributing the loss the the initial coin flip?
The problem does not "start all over again", since the original team is now securing ball possession through their own efforts and the inability of the opposing team. Different scenario.
RE: Moronic comment.
Your entire comment is absurd. Would you like to bet that the future winning % of the receiving team in OT will be closer to 62% than 50%? I'll bet you every dime you have. Give me a break about decks of cards.
Derek Jeter gets a hit 300 out of 1000 at bats. We're not supposed to forecast how often he'll get hits in the future? I guess that puts an end to all applied statistics. Here that guys, stop all the math. Turn out the lights. Shut down the server. Anonymous here has picked a Jack 50 times out of 200.
The 'availability heuristic' comment does not make sense. The problem very well starts all over again. The worst the receiving team can do is have an equal number of possessions, and the best the kicking team can do is have equal possessions. That's the problem. That's what starts all over again.
Only in the rarest of circumstances do football defenses ever score directly. If you want to rest your argument on that, feel free.
Gosh, I cringe thinking you might be some professor of epistemology somewhere, responsible for teaching others. But you sure sound like it. Why is it that the dimmest people are the ones so infatuated with their own intellect?
I love how an event that would only happen once in 121,718,908,847,851 tries shows just how unreliable all this stuff is.
If I may, I'd like to try to mine something meaningful out of that pile of... stuff.
Brian, you wrote, "The worst the receiving team can do is have an equal number of possessions, and the best the kicking team can do is have equal possessions. That's the problem. That's what starts all over again."
And I agree, of course. However, I think it's even more illustrative to think about this in the context of the entire game.
Each posession in a football game ends one of three ways:
1) A score, followed by a kickoff
2) A safety, followed by a free kick
3) A punt or turnover
4) End of half/game.
At the end of each of the first three events, the other team takes posession. The only exception is an onside kick, which is an earned extra posession, and I'll ignore that for our purposes.
So, if team A receives to start the first half, they will either have at 1 more posession than team B in the half, or an equal number. In the second half, it's reversed - either team B has one more, or it's equal. So, in total, the two teams will have either an equal number of posessions in regulation, or one team will have ONE more than the other.
The single-kickoff OT ruins this setup. In an OT game, it's entirely possible for one team to have TWO more posessions than the other team.
That's what the coin toss and the single kickoff causes, and jurry-rigged rules about requiring a second posession or a win by 4 or whatever don't address this fundamental issue.
You can solve this problem by having two kickoffs, or by having an abomination like college football does. Both of those solutions have an even number of "kickoffs". Of course, zero is also an even number, which is how the "keep playing" solution works.
The problem with keep playing is that it totatly changes the time dynamics of the 4th quarter. Time pressure at the end of each half is big part of football.
Keep playing basically means that the last team with the ball with time running out has the advantage of no time pressure--no need for hurry up. Just take your time and not worry about the clock.
Les
For playoff games, play another 15:00 period to the end.
For regular season, personally, I like ties, especially as they reduce the idiotic tiebreaker procedures.
But, in lieu of the current format how about a system of each team gets one possession, beginning with a kickoff, to do what they can? After each there is either a winner, or if it ends in tie either leave it that way, or the team that was most succesful (i.e. gained the most yards) or efficient (scored in the shortest amount of time) is awarded the win.
I don't know if this any better, probably not, than other recommendations, but some of the other recommendations inspired me, so in the interest of adding to the brainstorm, I've submitted it.
Les, that's only true in the relatively small fraction of games where the score is tied when the last team has the ball. Most of the dramatic scenarios in football involve one team being down by between 1 and 8 points as time ticks away. Those are unchanged - except the down 3 and down 7 scenarios, which have a new added element of drama.
Being able to run long drives irrespective of the clock is also a big part of football. Heck, being able to kill clock is already a part of lots of endgame scenarios, for the winning team. The 2 minute drill doesn't go away with this change, it just becomes less important in one relatively uncommon scenario.
And of course, overtime as it currently works already has essentially no time pressure. What this rule does is makes it so a game that's tied in the closing minutes starts to feel like an overtime game immediately.
If having a small portion of endgame scenarios involving relatively little clock pressure is a deal breaker for you, then so be it, but this seems like an odd hangup to have. It's a system that's fair, involves playing normal football the whole way, and on average produces slightly more dramatic moments (going for 2 when down 7, going for it on 4th down when down 3) and slightly shorter games (both of those scenarios avoid OT).
Tarr
I guess the problem I see with keep on playing is you are simply deciding who gets the ball in overtime. The game is over at the end of the 4th quarter. Now you are into overtime and you need to figure out who gets the ball and where. You want to give the ball to the team that had possession at the end of the game at the place they were when the game ended. The current method gives the ball to the winner of the coin toss at where ever they return the kick off to.
I don't see your method of fairer than a coin toss. Your method just ignores the fact that the game is now in overtime.
All of our discussion probably won't result in any changes. The owners, players, and TV want a quick end to the game and the current method usually gives them that. Plus all the comments about how it can be improved keep the NFL in our conversation after the games are over--a plus for the game.
Les
No, scenarios in which both teams get possession does not "start the problem all over again." I thought your issue was the arbitrary nature of the coin flip and its use in deciding who gets the first possession. When the receiving team gets possession for the second time, the ball was not arbitrarily awarded. It was earned. That's a difference. If you do not see this difference, then for some reason you do not think that defense is a real part of the game. You do not like to watch American football. Kind of like how Peter King doesn't like to watch football, he just likes writing about how rich he is and how neat-o his hero Brett Favre is.
Ok, it was awarded. But if the coin-flip winning team scores, they've had 2 possessions to score, and the coin flip loser had 1.
Pretty simple.
Les, you write,
"You want to give the ball to the team that had possession at the end of the game at the place they were when the game ended. The current method gives the ball to the winner of the coin toss at where ever they return the kick off to.
I don't see your method of fairer than a coin toss. Your method just ignores the fact that the game is now in overtime."
You don't see how it's fairer? OK, here's an incomplete list:
1) As Brian has been explaining, it avoids the problem of creating an unearned extra posession for one team. In the current system, an OT game can have 2 more possesions for one team than the other team over the course of the game - that can only happen in an OT game. "Keep playing" solves that problem.
2) It's completely non-arbitrary, so teams know in advance what the situation will be and can strategize for it.
3) It rewards the team that, by virtue of the flow of the game, was in the best position to score as time expired. Rewarding that is different than the way things work now, but it is anything but arbitrary, and it's perfectly fair.
Your objections ("ignoring that the game is in OT") are basically process objections as opposed to outcome objections. You're objecting to having an OT that doesn't work or feel like a distinct game, like the current one (sort of) does. I don't dispute that. If you like the current OT structure and want something that mimics it, this isn't for you.
But the outcome goals of OT (fair, non-arbitrary, decisive, quick, "real football", etc) are all satisfied by this approach.
^It's also a bit fairer in that the team with the ball starts OT on the down they were on - could be 2nd, 3rd, 4th... Under the current format one team gets the ball arbitrarily and gets a new set of downs.
The home team gets the ball first would also allow the teams to adjust their strategy in the 4th quater. wouldn't a visiting team be more likely to go for the win instead of the tie at the end of the game, if they knew they would not get the ball first in overtime?
I like Dave's idea the best. Same basic SD format with the flip, first TD wins the game at any time, if you kick a FG the other team gets a chance to answer with a TD to win, if they fail they lose.
Seems to me this is a good compromise betweem the "first team wins 61%" of the time and "equal possesetions gives the advantage to the second team" problems. While the team attempting to score a TD after a FG knows what they need to do to win, and thus has an advantage, at least they will always have to acheive a relatively difficult task in scoring a TD and not some chip shot FG on 3rd down from the 15.
And while this system may still favor the team winning the coin flip it won't favor them quite as much, it will leave fans feeling like the winner earned it and it will be way more exciting.
I'd also be open to modifying this slightly to give the home team the option to kick or receive in OT, and possibly having the team that made the FG kick off from the 20 or 25. This would add a lot of drama to those 4th down and 2 type choices while in FG range. Does the coach take the FG and make the other team score a TD knowing they will have good field position or does he go for it? How much more fun would that Vikings v. Saints game have been in a system like this?
Anon,
"The home team gets the ball first would also allow the teams to adjust their strategy in the 4th quater. wouldn't a visiting team be more likely to go for the win instead of the tie at the end of the game, if they knew they would not get the ball first in overtime?"
That's a good point. I've always liked baseball's system in the 9th inning because of that reason. The visitor plays agressively in the top half knowing just 1 run to tie or go ahead may not be enough. So, visitors in football would react similarly late in the 4th quarter and wouldn't be so content playing for a tie. And I really hate watching teams play for a tie.
Give us the stats on games that ended on first possession field goals. Stop lumping in touchdowns in that number!
Fight the fair police and play defense!
Hey all you who favor homefield OT possession should think about the fact that the visiting team at the end of the game might not kick a field goal or the extra point to tie it and send it into OT because of the other team being at home and definitely getting the ball. That would give the visiting team the edge in the last 2:00 of the 4th quarter knowing and choosing whether or not to go for it on 4th down or the 2-point conversion based on the skill of the home teams offense
How about a simple kick off. Each kicker kicks back and forth from the 30 then the 40 then the 50 and the 60 etc. First one to miss loses. No team just the kicker. It would be completely fair and would be over in a few minutes.
The only problem with the old sudden death overtime rules is the coin flip. The option of possession should be awarded based on yards gained. Since possession would be earned rather than random, there should be no sense of unfairness.