[BostonCard]

I think the probabilities reflect the pre-game assessment of the relative strength of the two teams (which is why the game never starts at 50/50) and the game situation (score, possession, down and distance).  Again, I would love to see a calibration graph of predicted probability versus actual outcomes.  In other words, if you take 1000 games with a team predicted to be 99.7% likely to win, do you see three (or so) losses?

That being said, a two touchdown lead with 2 min 15 seconds left in the game and first and ten on your own 25 (the situation at the start of the series where Love was injured) is about as close to the game being decided as you can really get.  An SC comeback in that situation would have likely required either a turnover, blocked punt, or punt return for a TD, followed by a successful onside kick, followed by another touchdown, and at least one 2-pt conversion or a win in overtime.  You never say never, but this was a much more dire situation than the Packers, who had 16 min 54 seconds and possession to overcome a 20-3 deficit, 11 min 52 seconds (with possession) to overcome a 20-10 deficit , and 2 min 39 seconds (with possession)to overcome a 23-17 deficit.  At the peak per ESPN  Chicago had a 98.2% chance of winning, which though high, does imply that you should see a comeback from that level once every 50 or so games, or about every three weeks in the NFL.

BC

[BostonCard]

I think the probabilities reflect the pre-game assessment of the relative strength of the two teams (which is why the game never starts at 50/50) and the game situation (score, possession, down and distance).  Again, I would love to see a calibration graph of predicted probability versus actual outcomes.  In other words, if you take 1000 games with a team predicted to be 99.7% likely to win, do you see three (or so) losses?

That being said, a two touchdown lead with 2 min 15 seconds left in the game and first and ten on your own 25 (the situation at the start of the series where Love was injured) is about as close to the game being decided as you can really get.  An SC comeback in that situation would have likely required either a turnover, blocked punt, or punt return for a TD, followed by a successful onside kick, followed by another touchdown, and at least one 2-pt conversion or a win in overtime.  You never say never, but this was a much more dire situation than the Packers, who had 16 min 54 seconds and possession to overcome a 20-3 deficit, 11 min 52 seconds (with possession) to overcome a 20-10 deficit , and 2 min 39 seconds (with possession)to overcome a 23-17 deficit.  At the peak per ESPN  Chicago had a 98.2% chance of winning, which though high, does imply that you should see a comeback from that level once every 50 or so games, or about every three weeks in the NFL.

BC