Monday, February 2, 2015
it is time to pause and reflect. That was said by none other than Mark Twain. This, of course, does not mean the majority will always be wrong. However, I believe the (great) majority of people are wrong about Pete Carroll's decision to pass. I will try not to bore you too much with numbers, but the fact is that Pete Carroll made a very good decision that simply did not work out. The big key to passing on second down is that it gave him the option to run the clock down to 20 seconds or so before snapping the play. This is very important. He figured that either it would be a touchdown or incomplete. If it is incomplete, the Seahawks still can run two plays, if they fail on third down, they can call a timeout with a few seconds left, and run one more play. This is not to say they do not want to score too quickly, but if they score (hoping to on second down), there is only 20 seconds left for the Patriots to come down at tie (or win) the game. Conversely, if they run a play with 35 seconds left (which they could have), and score, the Patriots now have 30 plus seconds to tie or win the game, and that is a huge difference. There are computers that are equipped to handle these probabilities with certain general assumptions. I will try to tweak them in an unbiased manner. Assume the Seahawks have a 65% chance to score by running, and assume also they have a 55% chance of scoring if they pass. (History suggests it is closer than that, but I will give Lynch about 5 extra percentage points). What is the probability of an interception (or fumble) by Wilson if you pass? Mike Sando from ESPN, tweeted that before that pass, there were 65 passes from the 1 yard line this year, and none of them were intercepted. Still, of course, there is some chance of an interception. What is it? He did throw into traffic. However, the quicker the throw (from the time it is snapped), typically the less chance of an interception. It seems that interceptions thrown into traffic in the end zone are more likely to be a play that has had a few seconds to develop and are in the middle or back of the end zone. Usually on the type of play called yesterday, it is either a touchdown or when it is not, the defenders do not have the time to react quickly to get an interception. Wilson was only 6 yards away from receiver, and that close errant passes normally get some part of the receiver, whether it hits another part of his body, or goes off of his arms, and goes into the air, which are somewhat prone to an interception but still rarer than most think. Let’s factor in that Wilson is under 6 feet tall, so he will some passes tipped, which are of course vulnerable to pick. But it is still a rare occurrence. Let’s put it at 4%, which if anything, I believe is high. Finally, we need assumptions for the Patriots coming back to win the game with the amount of time left on the clock. Assume they get the ball on their 25 yard line with 20 seconds left and down by 3. According to advancednflstats.com, the average team will win 5% of the time. But the Patriots are not average, and despite Seattle’s defense, the probability is likely closer to 10%. Let’s call it 8%. If the Seahawks score with 35 seconds left, that probability goes up to about 18%. We can then use probability concepts for each scenario. Seattle scores on second down by playing Carroll’s way, they don’t score on second down playing Carroll’s way, they score on second down by running Lynch, etc. etc. A comparison of the two ways (I will spare you the math, though you can email me at firstname.lastname@example.org if you want to see it) yields that it is better to pass than run and the primary reason is that you can start the second down play with a little more than 20 seconds. Again, you could not do that by running Lynch and still be sure to run 3 total plays if need be. Then, the other question becomes HOWS they should have passed. Perhaps a fade pattern was safer. Perhaps have Wilson roll out. I do agree that this decreases the chance of an interception. And let’s assume that the chance of a TD is the same no matter how they choose to pass. The downside to another type of pass (I think) is that there is a better chance of a holding penalty due to the play taking longer to develop. I don’t know what is better. There was nothing wrong with the call. If Carroll erred, it was not by much. And I think he made the right call (and running the clock down to 20 seconds). It just didn’t work.
Posted by The Wizard at 5:41 PM