Chase, the only stat that really matters is turnovers. I'll take 15-25 for 100 yards and 0 ints anyday over 28-40 for 350 yards and 4 picks. Manning has time and again committed turnovers to end his team's season and Brady has not.

I'm not sure which one I'd take. For those performances to be equal, an INT would have to be equal to -47.5 yards.Regardless, I'll take 28-40 for 350 with 3 TDs and 2 INTs over 15-25 for 100 with 0 and 0. So for me, turnovers do not supercede everything.

OMG

Its not the yards that are important.... ITS THE POINTS THAT ARE CREATED AND LOST FROM THE TURN OVERS THAT WIN AND LOOSE GAMES.

I have a book called the Hidden Game of Football, written by Bob Carroll, Pete Palmer, and John Thorn. I highly recommend this book to anyone interested in sports statistics.

Anyway, to the subject of the value of turnovers. First, here's a note on the data they studied:

To come up with our answers, we've used the statistical footprints of literally thousands of pro football games. We've calculated more than five hundred NFL games on a play-by-play basis. That's more than two seasons' worth. For specific reference, we'll use all of the games played in the 1997 season.

On the subject of the cost of a turnover:

The most interesting information we can derive from this is an estimate of what a team loses on average each time it fumbles. By examining thousands of downs, we can see a "net yards expected" for each situation--how far a team might expect to move the ball down the field under normal circumstances. The details are a little involved, so we'll explain how we arrived at the figure at the end of this chapter. For now, we'll simply tell you that a fumble costs a team 50 yards (in round numbers).

Surprisingly, interceptions are slightly less expensive, although they are much more common. Our calculations indicate an interception will cost a team about 45 yards. Again, a full explanation will be found in the note at the end. For now we'll simply say that an interception normally happens farther down the field than a fumble.

Elsewhere in the book the 5 yard difference is explained as follows:

...the average return of an interception (about 14 yards including touchbacks) is less than the average length of the pass (about 19 yards). That's based on historical data.

I suppose this implies that the average return of a fumble ends up placing the ball at the original line of scrimmage, though I did not see them explicitly say this. They did break down fumbles for the 1997 season. There were 379 turnover fumbles, broken down as follows:40 on bobbled snaps

112 on running plays

103 by the passer on passing plays (sacks or flushed out of pocket)

58 by receivers after catching a pass

30 on punt returns

20 on kickoff returns

4 on pass interception returns

9 on returns of other fumbles

3 on botched kicks

This is all accounted for by their analysis, but this shows why fumbles are more costly. The majority of the 103 passer fumbles on passing plays were almost certainly behind the line of scrimmage, and the 40 bobbled snaps were at the line of scrimmage, give or take a yard. Of the 112 on running plays, at least some of them were probably on handoffs in the backfield. Heck, even some of the 58 fumbles after receptions may have been by backs behind the LOS. Meanwhile, as shown above, the typical interception occurs 19 yards downfield.

On the subject of equating yards to points:

In the NFL, it takes about 12 yards of offense to add a point on the scoreboard.

They have plenty of analysis that backs this up. Thus, a fumble can be said to "cost" about 4.2 points and an interception 3.75 points.This analysis says that 350 yards and 4 picks equates to 170 yards and 14.2 net points, which is greater than 100 yards and 0 picks, which equates to 8.3 points.