Loyal Cougars

Stats: Going For 2 a Losing Decision…Every Time

At the end of the BYU/Utah game Saturday, BYU coach Kalani Sitake was faced with the crucial decision of whether to go for two to win in regulation or kick the PAT and go to overtime. He chose to go for two and the win. As we all know, Taysom Hill was stopped short of the goal line and BYU lost to Utah 20-19.

Debate erupted online about whether or not Coach Sitake’s decision to go for it was the right call. Many former players and media agreed with the call. Fans were split. Some relied on the old coaching adage: “when on the road, go for the win.”

The big question is: was it a “good call?” What were the odds of winning on a 2-pt conversion vs. winning in OT?

I dug through the last 5 years of FBS college football games, and found 26 instances where a coach was faced with this decision. His visiting team had just scored a TD in the last minute of the game, and he could elect to either go for two to win or one to go to OT. Of the 26 decisions, only 4 coaches decided to try for 2;  85% of the time, coaches chose to kick a PAT to force overtime.

Here are the winning percentages for each decision:

 Decision Winning Percentage
Go for 2 0% (0-4)
Kick PAT 73% (16-6)

Yes, you read that right. In the last 5 years, going for two has failed to yield a win EVERY TIME. When visiting coaches forced OT, they won at a shocking rate for a road team.

It gets worse. This last week, two teams tried to win the game with a 2-point conversion (BYU and North Dakota) and both failed. That makes 0-6.

And it gets worse. In the last 5 years, six home teams scored in the last minute and decided to try and win with a 2-pt conversion. Their record? 0-6. That makes 0-12 for this strategy.

What’s really odd is that BYU was involved in 3 of the 12 games. Twice BYU (with Taysom Hill at QB) tried for 2 and failed (at Utah in 2016 and at Boise in 2012) and once BYU won because the home team tried and failed (at San Jose State in 2015).

So, you decide. Was it a “good call?” You know my response.

 

Follow John Ahlander on Twitter: @NoneTheYser

*Source data: https://docs.google.com/spreadsheets/d/1nWBzNDBD7D42uLa7TjzXddpx-xZgmN7rhmMr2FyWnSE/edit?usp=sharing

18 Comments

  1. Kent

    September 13, 2016 at 9:59 am

    2-point conversions, over a much larger sample size than the one the article’s analysis relies on, succeed at a rate of over 40%. If there are factors particular to a specific game that make the odds of success in overtime low, then it makes sense to go for two. What sunk the game for BYU wasn’t a trend from a very few other games, but a botched play.

    • Jason Wilson

      September 13, 2016 at 5:47 pm

      Only one problem with the analysis, we don’t know anything about the teams trying for two or the teams kicking the extra point. On Saturday BYU was down their best offensive and defensive players and Utah had just rammed the ball down BYUs throat for about 11 straight minutes. Not to mention Utah had the better kicker. Good decision, wrong play call.

      • Batman

        September 13, 2016 at 8:49 pm

        I agree. Good decision. Good play call also. What happened is that the left side of the line blocked the wrong guys. One DL was mistakenly doubled and the end was able to go free. If blocked correctly BYU wins. Oh well.

      • William Munny

        September 14, 2016 at 12:16 pm

        Nah. BYU just scored in a hurry up offense. The game was close. The game was close. The games was really close. Utah had 6 freaking turnovers. I guess we are supposed to conclude that this trend would not carry over into OT. 2 BYU starters were out, but the replacements were fresher. The stats are the stats. It does not pay to go for 2. Poor call. All day long. BYU had just as good of a chance as UT to win in OT.

    • Kyle Kenny

      September 14, 2016 at 10:11 am

      Forgive me but i don’t understand your conclusion. Author says 0% and you say 40%. Both still sound like losing strategies to me. Put it differently, BYU had a 60% chance of winning going into overtime. Coach made a rookie choice and an even more rookie playcall. The end.

      • Kent

        September 14, 2016 at 10:25 am

        40+% is a figure based on all kinds of cases. Not all cases are average and have that probability of success. Some have more, some less. As I said, if there are factors particular to a specific game that make the odds of success in overtime low, then it makes sense to go for two. There were such factors in the game Saturday, as others have pointed out.

        It doesn’t follow from the 40+% figure that there was a 60% chance of BYU winning in OT. I don’t know where you got that figure. But in any case, it would represent an average case, and not all cases are average.

    • John Ahlander

      September 14, 2016 at 8:09 pm

      The research wasn’t about the success rate of 2-pt conversions. That’s an easy stat to find. It was about the success rate of game-winning 2-pt conversions. It’s like the difference between free throw percentage and free throw percentage with 2 seconds left to win the game. Same action, but totally different scenario.

      If game winning 2-point conversions were like normal 2-point conversions…they would be at 40%…and they aren’t.

      • Kentt

        September 14, 2016 at 9:14 pm

        As several have pointed out, your sample is too small to draw such a
        conclusion from. Your stats, if taken at face value, would imply no chance
        of success, which seems highly unlikely. Stats of exactly the same thing,
        apart from the game situation, based on a far larger sample, are likely at
        least as good a guide as such a small sample.

        Unlike with free throws, both sides in a football game are called upon to
        execute in a 2-point conversion. If the defense succumbs to nerves, the
        offense may walk in. There might be some advantage to the D over other
        2-point tries, but who knows how big it actually is.

        • John Ahlander

          September 14, 2016 at 11:05 pm

          First, it’s a rare situation, so the numbers will never be large no matter how many years you look at. Let’s say the occurrences hold per year. If we go back twenty years, there could only be 16 situations where the coach went for 2 to win. Would 16 be a large enough “sample” to convince people it is a bad idea? It’s all the data there is.

          The only reason the number is small is because so few coaches choose to go for two. 85% choose to kick the PAT, and when they do they win 73% of the time in OT. That has 23 data points, which won’t be enough for some people either.

          No matter how much people argue that there is not a lot of data, the fact remains. In the last 5 years, every coach that tried to win the game on a 2-point conversion at the end of regulation…lost. This last Saturday, that data got stronger.

          Of course, it won’t hold forever. Tomorrow, it may happen in a college game and the stat will be 1-13, instead of 0-12. I wish it was 1-11 right now. If it were, I probably never would have researched it. 🙂

      • Russ

        September 15, 2016 at 1:09 pm

        Hmmm here is a road team winning a game with a 2pt conversion
        Sooooooo that makes it better than 0%
        https://youtu.be/Y6X6bM71HEc

  2. Batman

    September 13, 2016 at 8:45 pm

    If the left side of the line had blocked the play correctly, BYU wins. It was the right call, there was just an execution failure.

    • John

      September 14, 2016 at 8:18 am

      Agreed. What would be interesting to see is BYU with Taysom at QB try the 2-point conversion against Utah 10 times and see what the success percentage is. I bet it would be over 50%.

  3. Spencer

    September 14, 2016 at 10:59 am

    “Yes, you read that right. In the last 5 years, going for two has failed to yield a win EVERY TIME.”
    What about TCU-WVU in 2012? Seems like a statistically insignificant sample, and a lack of basic research skills.

    • Kent

      September 14, 2016 at 11:28 am

      I wondered about that too. I think the TCU-WVU case was already in overtime. I suppose that’s why it wasn’t counted. But it’s the same kind of choice, going for 2 for the win rather than settling for a(nother) round of OT.

    • John Ahlander

      September 14, 2016 at 12:58 pm

      Interesting example. TCU actually elected to kick the PAT at the end of regulation (1:28 remaining) to force overtime, instead of trying to win on a 2-pt conversion. Sorry, I’ll go back to basic research now.

  4. polyesterman

    September 14, 2016 at 11:18 am

    This is a terrible conclusion made based on a poorly sized and chosen sample. The first comment is correct that all 2 point conversations should be selected. Then you need a statistically relevant sample period to measure from.

  5. Rob Lattin

    September 14, 2016 at 11:47 am

    BYU lost on the 2 pt attempt but KS won on every other front through out the game. I love what I saw. Heck, if KS told me to suit up and go to battle I would.

    • John Ahlander

      September 14, 2016 at 8:14 pm

      A really good point. Even though the 2-point conversion failed, and in my opinion was a bad decision, Coach Sitake won over a lot of people with the guts to go for it. My question is, if he would have played the odds, went to OT and won…would people like him more because he beat Utah for the first time in 6 years? I’d prefer a win to gutsy play calling.