Kicking a field goal on the two is like kissing your sister.
I hate ties. Both as a fan and as a stat cruncher. But I hate them a lot more as a stat cruncher.I could not have said it better myself. Nothing is more annoying when trying quantify a season and run simulations than coding for a tie. The Packers and Vikings tied 26-26 on Sunday after both teams kicked field goals in overtime. On the opening drive, Matt Flynn led the Packers to the Vikings 2-yard line before Mike McCarthy decided to kick a field goal, sending the game into the "chance to match down three" format. In the new overtime format, was this the optimal decision?
— Brian Burke (@Adv_NFL_Stats) November 24, 2013
A field goal from the two is a virtual certainty (an extra point essentially) and a made field goal gives the Packers a 79% chance to win according to Brian's new overtime win probability adjustments. If the Packers go for it on 4th-and-Goal from the 2, it converts league-wide at 55%. I would caution that the Packers offense is likely below average with Matt Flynn at the helm, although Eddie Lacy had been having a solid game. If the Packers convert and score a touchdown, the game is over, giving them a 100% chance to win. If they fail, the Vikings get the ball inside their own 2-yard line, but Minnesota would then only need a field goal to win the game.
From a team's own two, a league-average offense will score a touchdown 11.8% of the time and a field goal 6.6% of the time according to our Markov model. Using Brian's OT win probability model (which has similar conversion rates), that would put the Packers' chances to win at 76% after turning the ball over on downs. The expected win probability for the Packers is as follows:
E[WP FG] = 100% * 79.0% = 79.0% WP
E[WP Go-for-it] = 55% * 100% + 45% * 76% = 89.2% WP
That is a huge difference. The break-even point for when to go for it is if you believe you can convert at 13% or better (13% * 100% + 87% * 76% = 79.1%). The truth is, if the Packers were terrified of the Vikings being able to move the ball down the field (I don't see why they would be), kicking the field goal is not much different from failing on fourth down.
Edit: After digging a little deeper, I believe the conversion rate is actually closer to 45%, not 55% as listed in Brian's fourth down tool. That would put the E[WP] of going for it at 86.8%, still well above the 13% threshold.
The Packers should have gone for it; ideally they would have avoided the tie and eliminated a lot of future frustration in my statistical life.
Keith Goldner is the Chief Analyst at numberFire.com - The leading fantasy sports analytics platform - and creator of Drive-By Football. Follow him on twitter @drivebyfootball or check out numberFire on Facebook.