Go for 2?

Discussion in 'National Football League' started by ColoradoContrails, Feb 4, 2018.

?

You're down 1 at the end of the game. Do you kick the PAT r go for 2?

  1. Go for the tie and OT

    5 vote(s)
    50.0%
  2. Go for the win.

    5 vote(s)
    50.0%
  1. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    This article was in the NY Times and I thought it was interesting:

    https://www.nytimes.com/2018/02/02/...ight-region&WT.nav=opinion-c-col-right-region

    Sorry I can't cut and paste it here, but hopefully you can read it via the link. It points out how most NFL teams when faced with the choice to kick the PAT for the tie at the end of the game, or go for two to win almost 90% of the times over a 10 year period the teams chose to go for the tie. But only 40% of the time did that team ultimately win - IOW, 60% of the time the team that went for the tie lost in OT.

    It seems that coaches tend to view the situation myopically: they base their decision on the upcoming single play rather than looking at the overall picture.They based their decision on the potential to lose the game immediately, as opposed to how they might perform in OT.

    The other finding that's very interesting is that when people think they have a choice they'll usually choose to avoid the risk, but when they feel they have no choice they'll take the risk. The experimenters set up this scenario:

    They divided the test group into two groups. They were both told that their team had just scored in the final seconds. But Group A was told their team was down 1, while Group B was told they were down 2. Then they told both groups that the team would go for 2.

    Group A were asked if the 2 point conversion would work or fail, and they believed it would fail. In other words, they saw the risk as "optional" and therefore they didn't believe it would succeed.

    I'm of the belief that more teams should go for two in situations like this and this study confirms my belief. What do you guys think?
     
    NCJetsfan likes this.
  2. HomeoftheJets

    HomeoftheJets Well-Known Member

    Joined:
    Feb 7, 2016
    Messages:
    3,203
    Likes Received:
    4,194
    Richard Thaler is a Nobel prize winner, so I'm sure he addressed this in the actual paper. But as far as I know, there's no good reason, aside from small sample size, that the team kicking the PAT should lose 60% of games that go into OT. It should be 50-50, which would mean that it doesn't matter whether you kick or go for 2.
     
  3. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    Well it was 42 out of 47 times over a ten year period that the teams in that situation decided to go for the tie. I'd say it's a pretty decent sample size. Moreover, they tested the theory looking at the NBA. In 772 cases where a team was down by two with less than 24 seconds, 71.1% of the teams decided to go for the tie instead of shooting the three to win. The teams that shot the three won 17.3% vs. the two point teams who won 14.5% of the time.On top of all that, they assembled groups of fans and gave them the scenarios and they chose similarly. So there is a proven bias towards trying to be risk-averse, but in actuality, it often costs you the game.
     
  4. HomeoftheJets

    HomeoftheJets Well-Known Member

    Joined:
    Feb 7, 2016
    Messages:
    3,203
    Likes Received:
    4,194
    I'm not talking about teams kicking most of the time. I'm talking about the team going for the tie losing 60% of the time in OT. Since I can't think of a good reason it isn't 50%, 47 is a small sample, and 60% isn't that different from 50%, I don't think this paper shows that teams are making the wrong decision to kick. The conservative decision but not the wrong one.
     
    Ralebird likes this.
  5. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    It may seem like not much difference to you, but most coaches would opt for the higher percentage play. That's supported by the fact that 90% of them believed the odds favored going for the tie, but they were wrong 60% of the time.I think the sample sizes I provided back this up.
     
  6. statjeff22

    statjeff22 2008 Green Guy "Most Knowledgeable" Award Winner

    Joined:
    Jun 4, 2005
    Messages:
    20,234
    Likes Received:
    1,633
    There's no question that NFL coaches are much too conservative; they go for it on 4th down far less often than the numbers say they should, for example. The combination of a short season and the outsized importance of a few plays on the outcome of a game makes coaches not make untraditional calls that will be criticized if they don't work, even if they're the right calls to make.

    As to the specific point of whether this is enough evidence to say that going for the tie is associated with significantly worse performance in the overtime, there isn't. The question here is whether 25 losses out of 42 games, or 59.5% (the correct sample size is 42, not 47, since you're only looking at the games where they went for the tie and went into OT), is significantly different from .5. This is a standard binomial hypothesis testing problem, and the p-value for the exact test statistic testing that hypothesis is .280, not even remotely close to statistical significance at the usual levels. Put in another way, the implied confidence interval for the true probability of winning the game in OT if you tied at the end of the fourth quarter with an extra point is (0.433, 0.744), with 0.5 comfortably inside the interval. In other words, HomeoftheJets is correct that the sample size of 42 is too small to say that the observed 60% indicates that the true probability of them winning in OT is different from 50%. Note that if you accept that the expected rate of success of two-point conversions is 50% (as they claim in the article), this means that the performance of the teams that kicked the extra point, while worse than that of the expected performance of teams that went for the 2 point conversion, was not statistically significantly worse. I would also like to know the success rate of two-point conversions late in close games; it might be less than 50% because of the pressures involved.

    Another issue that is now true is that extra points no longer have a 98% success rate, but rather more like a 92% success rate. They're no longer super-gimmes, as we saw in the Super Bowl today. That would make the 2-point conversion more attractive.
     
    slimjasi likes this.
  7. JetBlue

    JetBlue Well-Known Member

    Joined:
    Nov 24, 2004
    Messages:
    8,725
    Likes Received:
    2,897
    In stats, each isolated incident, like a coin toss, has a 50/50 chance; the same would apply to OT.

    Sure, the recorded events may indicate that teams that have gone to OT after kicking the game tying point lose 40% of the time, just as heads may come up 60 times out of 100 when flip a coin, but statistically you still have a 50/50 chance at either heads or tails.

    Kick the point and take the 50/50 chance to win — the same chance you had when the game started.
     
  8. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    Well I'll admit I'm way out of my depth to try and argue based on a pure statistical basis. That said, I still maintain that almost no coach -in any sport - would be familiar with the odds as you describe them. Rather, they're going to base their decision on their personal experience, both first hand and from what they've observed other coaches do, and on that basis, if they were aware of the outcome of this study it might well change their thinking.
     
  9. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    But that's where the fallacy lies: in believing that you start the OT with a "blank slate", unaffected by what has happened already in the game. Consider that your team has just scored, their emotions are sky high and so is their adrenaline level. That is exactly the best time to strike, and not allow the other team to regroup or your team's energy to recede. Sure, there will be times - 40% according to the study - where you lose, but more often -60% of the time - you'll win.
     
  10. statjeff22

    statjeff22 2008 Green Guy "Most Knowledgeable" Award Winner

    Joined:
    Jun 4, 2005
    Messages:
    20,234
    Likes Received:
    1,633
    The problem is that coaches are VERY well aware that it's almost always a good idea to go for it on 4th down and short when you're between your own 45 and the opposition 30, yet very few are willing to do it. They are completely risk averse, for the very good reason that no one gets fired for punting the ball away when "common wisdom" says it's the right thing to do, but coaches do get fired for making so-called "crazy" decisions that don't work out. Doug Pederson is probably the most aggressive coach in the NFL, in large part because the Eagles rely on analytics more than any other team in the league. It won them 13 games this year, and it just won them the Super Bowl, too. It would be great if that would embolden other coaches, but I'm not counting on it.
     
    #10 statjeff22, Feb 4, 2018
    Last edited: Feb 4, 2018
    ColoradoContrails likes this.
  11. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    You're right...just because the statistics bear out that the risk is rewarded, coaches ARE risk-averse, (and we have one of the most risk-averse coaches in the NFL), and won't take the chance. But Pederson was rewarded for his calculated risks - a coach after my own heart.
     
  12. JetBlue

    JetBlue Well-Known Member

    Joined:
    Nov 24, 2004
    Messages:
    8,725
    Likes Received:
    2,897
    If you have emotions and energy after the score, wouldn’t that still be on your side, if that’s meaningful, going into OT?

    Did I read it wrong? 60% of the time you lose in OT. Basically you are weighing the 100% chance you lose if you miss the 2 pt attempt versus the 60% chance you lose if you kick the XP.
     
  13. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    I don't think so. With the time stoppage involved it gives the other team a chance to regroup and the adrenaline and heart rates of your team will have settled down. Of course I'm not saying this is what happens in every case, but the ODDS favor it. I think Pederson's approach to calculated risk taking bears this out, and again, something that Todd Bowles should learn from.
     
  14. JetBlue

    JetBlue Well-Known Member

    Joined:
    Nov 24, 2004
    Messages:
    8,725
    Likes Received:
    2,897
    Either way, I think the decision should weigh more on objective criteria — how good is the opponents D in short yardage or goaline D, how strong you are in those situations, and have you been successful in the game in comparable situations.
     
  15. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    Sure, those things have to be considered, but when all things are weighed, "Fortune favors the bold".

    And I don't know if this study, or any others, looked into the effect that NOT going for the win has on the psychology of each team. Does that send a "loser's message" to your team, or a "surrender message" to the other team? I believe it does, and it plays into the success/failure rate.
     
  16. slimjasi

    slimjasi Well-Known Member

    Joined:
    Sep 27, 2007
    Messages:
    7,069
    Likes Received:
    1,603
    Been saying this for years.

    It's trivial statistics.
     
  17. Ralebird

    Ralebird Well-Known Member

    Joined:
    Sep 25, 2012
    Messages:
    3,187
    Likes Received:
    1,655
    No one has mentioned home field advantage and it's hard to discount that as a key condition. Stats I could find show that in all NFL overtime games in the 2012-2015 seasons the home team won 58.9% of the games leaving the visitors with just 41.1% of games ended with a decision. So a choice of kicking the point, with a 94.3% success rate in 2016 then playing the OT would result in a home team win percentage of 55.5% and a visitor's of only 38.8%.

    All two point conversions were successful in 49.5% of the time in 2016. This indicates the home team would have a clear advantage in forcing the overtime while the visitors would statistically benefit by going for the two point play. Not included in this calculation is the kicker's conversion rate home vs. away or if there is a difference in two point conversions based on home or away but either would probably not be wide enough to outweigh the OT scenario.
     
    ColoradoContrails likes this.
  18. ColoradoContrails

    ColoradoContrails Well-Known Member

    Joined:
    Nov 28, 2016
    Messages:
    3,821
    Likes Received:
    3,659
    Good point. I don't know if that study accounted for HF advantage or not, but what you pointed out makes sense. Bottom line: I think coaches could be a lot more aggressive than they are, especially our HC.
     
  19. Sam Hammer

    Sam Hammer Well-Known Member

    Joined:
    Aug 7, 2015
    Messages:
    1,504
    Likes Received:
    1,282
    You have to go with the safer option in a big spot like that. Going for 2 when only need one to tie, is a college thing and honestly doesn't make sense to me. It tells the team you don't have confidence in them in over time.
     
  20. Ontherise

    Ontherise New Member

    Joined:
    Jan 25, 2011
    Messages:
    25
    Likes Received:
    1
    Any chance the team that is "just able to tie it at the end of the game" is actually the worst team and would be expected to have a less than 50% chance in OT?
     

Share This Page