Washington Post: Was Cunfiff's 62-Yard FG Attempt Wise?

This week's article at the Post looks at whether the numbers favored the Redskins' 62-yd FG attempt or going for a conversion after Josh Morgan's insanely stupid unsportsmanlike penalty call.

After every game, each NFL coaching staff evaluates the performance of players on each and every play. Coaches also grade themselves on play selection and other decisions, such as challenges, clock management and fourth down decisions. The Redskins staff should take a hard look at the final decision of Sunday’s game to attempt a 62-yard field goal rather than try to convert a 4th and 16...

...There isn’t the luxury of time and a computer full of stats on the sideline during a game, so it’s hard to fault the Redskins’ coaches. The value of analyses like this one is not to prove there is always a right or wrong answer, but to recalibrate a coach’s intuition so that he’s armed with the best information next time.

Mike Shanahan disagrees with me, but he is wrong. Already the post has hundreds more comments than their typical post gets, so it seems it struck a nerve. The comments there have come a long way from a couple years ago, but a quick scan of them tells me most of them miss the point.

By the way, these posts now appear in the print version of the Sunday Post each week.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

14 Responses to “Washington Post: Was Cunfiff's 62-Yard FG Attempt Wise?”

  1. James says:

    Couple of questions:

    1. Does the conversion probability include a first down from penalties? It would greatly complicate the math as a penalty could be anything from encroachment (making it 4th and 11) to defensive holding (only 5 yards but new set of downs) to defensive pass interference all the way down to the 1.

    2. Are the EPA and WPA of the play counted separately from the penalty, so that Griffin and Morgan share credit for the play, but only Morgan is punished for the penalty?

    3. Congrats on being in print!

  2. Eric says:

    Brian is clearly correct here. There would have been more than a minute left on the clock. The Rams were moving the ball with ease and their rookie kicker has distance. A made 62-yarder from Cundiff - which is a longshot - would still most likely have resulted in a loss for the Redskins.

    The replacement refs were awful in this game and the Redskins should have been losing by double-digits anyway. Shanahan's last bad decision was just one in a long series of them. He was outcoached by Fisher for the other 59 minutes, too.

  3. Brian Burke says:

    The conversion rates are the average for all causes, including penalties. For 4th and 11, you'd need an in-play 15-yd personal foul or an automatic 1st down from a DPI or illegal contact.

  4. Anonymous says:

    still shocking to see so many angry comments

  5. Anonymous says:

    So Shanahan pissed away all of 0.03 wins? Sorry, but there are far biggers errors to pick on in the NFL if one wants to Monday morning QB.

  6. Ian Simcox says:

    Shanahan missing the obvious point that it’s not about whether Cundiff had a better chance of making the field goal than the offense did of converting 4 and 16.

    Even if Cundiff had a 19% chance of making the FG, the Redskins WP would be 9%, less than their chances had they gone for it. Even though the data is sparse on the likelihood of success, we can say pretty confidently that Cundiff’s chances were not 25% better than the offense’s, which is what they’d need to be to justify the attempt.

  7. Brian Burke says:

    Anonymous-An NFL head coach would stay up around the clock for weeks and spend a million dollars for an extra 3% edge. We know this because they do.

  8. Unknown says:

    One big problem with Shanahan's thinking here is that he keeps talking about the chances of converting a "4th and 16." Don't say that. Just say "what are our chances of moving the ball 16 yards or more?"

    I'll gamble on RGIII's ability to move the ball 16+ yards over a kicker's ability to nail a near-record-tying field goal. And even if you make the FG, you have less than a 50/50 chance of winning.

  9. Anonymous says:

    It is my conjecture, or guess if you prefer, his mistake was due more to the psychology than an understanding of game probabilities. Shanahan had made the decision to go for the field goal before the 15 yard unsportsmanlike penalty has called. He locked in and was committed to “kick a field goal and go for the tie on the road”. This is a coaching mental lock, repeated endlessly and made mental fact, if true or not. He knew this in his “gut” or whatever part of anatomy that coaches use to make quick “tough” decisions. Then the shock of the stupid penalty right in front of him caused enough cognitive noise to stop the rational process. Anger, rage, and incredulous disbelief of what he just saw, and what his staff experienced made it unlikely that a reasoned approach would be taken to the next play. The information that a field goal had just become long, so long that only a few have been kicked in the history of the league, did not make it into the decision tree. No probability or statistical analysis is required …it’s a really long kick guys the chances are known to be small. But the “right” decision had been previously made, “Go for the tie on the road.” So the kicking team was sent out and Shanahan and the rest of us watched the result.
    Shanahan should feel in good company. Military and commercial histories are littered will disastrous decisions that were based on experience and not immediate fact. Robert E. Lee’s decision to “seize the moment” at Gettysburg, and charge the center of the union line, was based on dozens of successes that were not supported by the facts of the day. Lee’s decision was far more disastrous for his men, than a loss of a game is for Shanahan’s. However to Lee’s credit he did admit to the mistake. Mike we are waiting.

  10. Unknown says:

    I agree with Anonymous' point as to Shanahan's thinking, but it begs the question of is kicking a 47 yd FG on 4th and inches with 1:12 left to tie the correct decision in the first place (my gut tells me every NFL coach kicks the FG and that they are all wrong to do so).

  11. Ian Simcox says:

    Unknown

    4th down calculator tells us that go-for-it on 4th and 1 with 1:12 left, down 3 is worth 0.33WP, while FG attempt is 0.28WP.

    So your gut is right, they should go for it, wind more time off the clock, possibly score the winning TD or kick a shorter FG and take 0.5WP into OT.

  12. Anonymous says:

    Was Josh Morgan insanely stupid or the ref? If Josh Morgan is Bayesian he believes that nothing is going to draw a flags from the retard refs, so he seems rational to me. No defense for the refs though.

  13. Unknown says:

    "Fourth and 16 situations like that are rare." the problem with breaking down rare situations into hard numbers is that there is significant margin for error. You haven't qualified your percentages with error bars so in the final analysis 10% looks better than 7%. What were the RedSkins' chances of making that 16 yard drive on that day against the Rams? I'm uncomfortable with the application of statistical techniques where there aren't many samples to go on. I feel that a statistical analysis of the situation perhaps should have come to the conclusion of insufficient data. A 3% edge is important in the NFL but was there really 3% here? I'm not convinced. Plot a graph of the probability of making that field goal as each of the 26 attempts is made and you'll see it jumps up dramatically after each successful FG. For example if we roll back time to before the last 6 long FG attempts had ended badly we'd be looking at 4/20 or 20% and the overall chance of winning up to 9.2%.

  14. Insaini says:

    Mr Burke,

    You calculate that the 4th down attempt would have improved the probability of winning by 3%. Surely there must be more than one way to calculate win probability, with errors built in to each method. Is it possible that 3% is within the margin of error? You are a statistics guy, and I am not, and I think that calculating error of probability predictions should be pretty complicated, so I am wondering what your thoughts on that might be.

    In other words when choosing between 2 different plays, one may need a threshold probability added in order to consider one play definitely better. Alternate decisions that do not meet some threshold of probability added may be considered a "wash." Would that be sound football logic?

    To be honest, even if the 2 options in this situation have similar win probability, I think that still puts conventional football logic on its head, since most coaches would probably choose to kick 99 times out of 100, even if they had a mediocre kicker.

    Regards,

    Naveen

Leave a Reply

Note: Only a member of this blog may post a comment.