Through four weeks the Cowboys have a -7 turnover differential, the second worst value in the league. Perhaps even more revealing than the simple number of turnovers are the consequences of those turnovers, specifically, the points scored off takeaways and turnovers, as not all turnovers are created equal. A pick six for example is much more damaging than a turnover that's followed by a three-and-out by the opposing team. And the Cowboys know this all too well.

Over four games, the Cowboys have allowed 38 points off turnovers (3 vs NYG, 10 vs SEA, 7 vs TB, 21 vs CHI) but only scored 7 points off a takeaway (vs. TB). The resulting turnover points differential of -34 is the second lowest value in the league and not a good place to be in. But before we dig deeper into into the Cowboys' numbers, I'd like to look at this year's leader and a couple of teams from last year.

This year, the 4-0 Atlanta Falcons lead all teams with a turnover points differential of +43. That's quite an achievement, especially since they scored 43 points off takeaways and did not allow any points off of their turnovers. However, consider that the Falcons' total points differential over four games is 48 points. Excluding points scored off takeaways/turnovers, their points differential is a decidedly pedestrian +5 points, which is something you're much more likely to see from a 2-2 team.

If you subscribe to the theory - like I do - that interceptions contain a large element of randomness and that fumble recoveries are almost entirely random, you'll understand that the Falcons are unlikely to be able to sustain this pace for the entire season, and it's quite likely they'll come back down to Earth pretty soon as the randomness begins favoring their opponents.

Two teams from last year illustrate this point: After Week 4 of 2011, the Lions and Bills led the rankings in scoring differential off turnovers. The Lions stormed to a 4-0 record behind a +47 scoring differential off turnovers and the Bills were at 3-1 with a +46 differential. But over the next twelve games, the Lions accumulated a -44 scoring differential off turnovers that gave them a 6-6 record over those 12 games, while the Bills imploded for a -67 differential and 6-10 record for the season.

The point here is that that if a given variable like scoring differential off turnovers is extreme on its first measurement (i.e. after four games), it will tend to be closer to the average on a second measurement (i.e. after 16 games). In statistics, this is called regression to the mean.

The 2011 Cowboys saw a dramatic reversal in their fortunes in this particular set of stats: After a 2-2 start, the Cowboys were ranked 27th with a -15 scoring differential off turnovers. Over the next twelve games, the team tallied a +38 differential, good for the 4th best value in the league over that period. It didn't matter much in the end because they still ended up 8-8, but it does show that significant swings can happen.

Here's the full table showing how teams rank in terms of takeaway and turnover points so far this year:

Points scored off takeaways/turnovers, week 4, 2012

At this point it's hard to say which teams at the top and bottom of the table will see a reversal in their fortunes over the coming weeks. But just like I believe some of the teams at the top of the table like the Falcons, Bears and 49ers will come back down to more normal levels, I also believe the Cowboys will find a few more balls bouncing their way than they have in the past four games.

I'll be watching this closely over the coming weeks, but on Sunday, I'm picking against the Falcons and for the Eagles, as I think both teams could see a significant swing in their turnover scores. We'll see how that goes.