One stat I would find interesting is percentage of yards possible vs actually gained. If the opponent gets 13 or 14 drives a game, yardage would be skewed. Assuming an average 20 yard line start, a 14 drive game could have a team gain over 400 yards and never cross mid field.
You make a good point but that doesn't seem to be the case here. Opposing offenses are getting to the Pats side of the field far more frequently than they should.
I find that the Pats defense is nearly impossible to evaluate at this point since:
1) They have kept teams relatively in check in the 1st half (10 pts or less). If they were a dumpster fire, at least one team of the 4 would have dropped 2 TDs on them.
2) Opposing offenses have not taken a snap (extra points excluded) with the lead. About half of the time they are snapping the ball down 2 scores.
The Pats defense has faced 7 2nd half drives with a tied score or up by one score:
MIA - TD after Brady pick. Hard to complain about a 9 yard drive
MIA - FG...long drive but a red zone stop
SD - Fumble after failed Pats 4th down near midfield. Huge stop
BUF - Punt, 3-n-out
BUF - Punt, 3-n-out
BUF - TD on long drive after Brady red zone pick. Long PI call a gut punch
BUF - FG to lose game. This was the killer drive on this list
OAK - None
So when the Pats have their backs against the wall in the 2nd half, it has been kind of a mixed bag. Take away the PI call against Buffalo and the picture is likely much different. Other than the last 2 drives in Buffalo, the other 5 are either excellent (two 3-n-out and a big turnover), not too bad (red zone stop for a FG) or understandable (short TD drive after a turnover).
These are the types of drives you want to evaluate since the context (2nd half, close game) is pretty clear and the variables (exchanging yards for time, playing prevent) are eliminated.