[2006alum]

Well we pull it out. There were some doubtful moments but we got it done.
Nice.

Florida State led 5-4 at the 7:33 mark in the first quarter.  That was its only lead of the game, and it was ahead for a grand total of 1% of the game.  Stanford took a 10 point lead midway through the second quarter, at which point ESPN’s win probability meter hit 90%.  By the start of the second half, Stanford was given a 98.4% win probability.  At the end of the third quarter, it was “down” to a 96.3%, and that was the low mark for the second half.  I wouldn’t characterize a game in which you have over a 95% chance of winning in the entire second half as “doubtful”.

BC

Tell that to Duke.

I literally made a screenshot of our 99.9% probability of winning with 2+ minutes to go and sent it to a friend asking why I was still nervous. She verified that we've seen worse things happen.

 
yes, duke was leading 40-30 at the beginning of the 4th quarter.

ESPN's probabilities have always seemed to sway too extremely to me. I don't know how anyone would say a team that is (a) on the road; (b) has won 1 Sunday road game all season; © has lost 8 of the last 10 games; and (d) has seen a 23 point lead dwindle to 8 points has a 97% chance of winning with 4:20 left. Maybe it really is the case that 97 out of 100 times that team wins, but I feel like experientially, those kinds of games feel very "doubtful" to me, too. And I suspect ESPN's algorithm is not takign into consideration (b) and © at all.

[2006alum]

Well we pull it out. There were some doubtful moments but we got it done.
Nice.

Florida State led 5-4 at the 7:33 mark in the first quarter.  That was its only lead of the game, and it was ahead for a grand total of 1% of the game.  Stanford took a 10 point lead midway through the second quarter, at which point ESPN’s win probability meter hit 90%.  By the start of the second half, Stanford was given a 98.4% win probability.  At the end of the third quarter, it was “down” to a 96.3%, and that was the low mark for the second half.  I wouldn’t characterize a game in which you have over a 95% chance of winning in the entire second half as “doubtful”.

BC

Tell that to Duke.

I literally made a screenshot of our 99.9% probability of winning with 2+ minutes to go and sent it to a friend asking why I was still nervous. She verified that we've seen worse things happen.

 
yes, duke was leading 40-30 at the beginning of the 4th quarter.

ESPN's probabilities have always seemed to sway too extremely to me. I don't know how anyone would say a team that is (a) on the road; (b) has won 1 Sunday road game all season; © has lost 8 of the last 10 games; and (d) has seen a 23 point lead dwindle to 8 points has a 97% chance of winning with 4:20 left. Maybe it really is the case that 97 out of 100 times that team wins, but I feel like experientially, those kinds of games feel very "doubtful" to me, too. And I suspect ESPN's algorithm is not takign into consideration (b) and © at all.