The numbers say it's the right move. You see it from time to time now, but in 5-10 years, every team will be doing it. The logic is as follows:
-2-pt conversions succeed 50% of the time. PATs are basically 100% (except for Bubba)
-Let's say you go for 2 after the first TD, and let's assume you hold the opponent scoreless the rest of the game
-You succeed 50% of the time on that first attempt. After you score your second TD, you kick a PAT and win. So, 50% of the time, you win the game.
-You fail 50% of the time on that first attempt. However, you succeed 50% of the time on that second attempt. So, 25% of the time (50% x 50%) you tie and go to OT, and the remaining 25% of the time you lose.
-OT is basically 50-50 since the format is so wacky and each team only gets 1 possession. So now, 12.5% of the time you win (50% x 25%) and 12.5% of the time you lose.
To recap:
-When you go for 2 on the first (and potentially second) attempt, you win 62.5% : 50% when you succeed on the first 2 pt attempt + 12.5% (50% x 25%) when you fail on the first attempt, succeed on the second, and go to OT
-When you kick the PAT, you win 50%: you go to OT 100% of the time, and win half the time
-You win 12.5% (or 1 out of 8 times) more when you go for 2 than when you kick the PAT
Throw in our terrible PAT kicking, and it was no doubt the right move.