SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.

Also, our existing PFR blog rss feed will be redirected to the new site's feed.

Pro-Football-Reference.com » Sports Reference

For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.

PI Finds: Not Your Typical 6-1 Pats

Posted by Neil Paine on November 1, 2010

Stop me if you've heard this one before: At 6-1, the New England Patriots boast the NFL's top record despite a roster that looks less than overwhelming on paper. So does this mean 2010 will be another one of those patented Patriot seasons where they're somehow on top when the dust clears? Let's go to the PFR Play Index's Team Game Finder and break down the numbers...

Searching for New England Patriots Season Games matching the criteria, from 2000-2010, for games #1-7 of the season, we find this year to be the 4th time the Pats have started 6-1 or better in the Bill Belichick era:

Year W L T W-L% ▾
2007 7 0 0 1.000
2004 6 1 0 0.857
2010 6 1 0 0.857
2006 6 1 0 0.857
2009 5 2 0 0.714
2008 5 2 0 0.714
2003 5 2 0 0.714
2005 4 3 0 0.571
2001 3 4 0 0.429
2002 3 4 0 0.429
2000 2 5 0 0.286

From the W-L records, it certainly seems like New England has returned to their dynastic ways this year. Still, let's dig a little deeper by searching for Cumulative Season Games with the same finder tool, sorting by point differential through 7 games:

Year G W L T W-L% PF PA PD ▾
2007 7 7 0 0 1.000 279 120 159
2009 7 5 2 0 0.714 198 98 100
2006 7 6 1 0 0.857 167 87 80
2010 7 6 1 0 0.857 205 154 51
2004 7 6 1 0 0.857 168 124 44
2008 7 5 2 0 0.714 153 132 21
2003 7 5 2 0 0.714 145 126 19
2001 7 3 4 0 0.429 161 150 11
2002 7 3 4 0 0.429 168 158 10
2000 7 2 5 0 0.286 120 141 -21
2005 7 4 3 0 0.571 159 180 -21

While this year's version lags behind the 2006-07 Pats, they're still outpacing their Super Bowl-winning predecessors from 2004 in terms of point differential... Again, is this a sign that the Patriots can expect to keep winning well into the playoffs?

Well, let's take another look at Cumulative Season Games through the first 7 games of the season, this time sorting by total yardage margin:

Year G W L T W-L% Rush Pass Tot ▾
2007 7 7 0 0 1.000 285 821 1106
2009 7 5 2 0 0.714 39 803 842
2006 7 6 1 0 0.857 326 -26 300
2002 7 3 4 0 0.429 -357 636 279
2005 7 4 3 0 0.571 -312 338 26
2008 7 5 2 0 0.714 120 -116 4
2004 7 6 1 0 0.857 -107 69 -38
2003 7 5 2 0 0.714 135 -291 -156
2001 7 3 4 0 0.429 -208 -39 -247
2010 7 6 1 0 0.857 78 -454 -376
2000 7 2 5 0 0.286 -209 -169 -378

Wait, what?

New England's foes have outgained them by 376 yards this season, the second-worst margin of the Belichick era (a whopping two yards better than 2000, when the Patriots started a miserable 2-5). So how have they been able to go 6-1, when traditionally a yardage margin of -53.7 ypg earns you a .378 winning %? Let's look at the same finder search, sorted by turnover margin:

Year G W L T W-L% TO ▾
2009 7 5 2 0 0.714 9
2007 7 7 0 0 1.000 8
2003 7 5 2 0 0.714 7
2010 7 6 1 0 0.857 7
2006 7 6 1 0 0.857 6
2008 7 5 2 0 0.714 3
2000 7 2 5 0 0.286 2
2002 7 3 4 0 0.429 0
2001 7 3 4 0 0.429 -2
2004 7 6 1 0 0.857 -2
2005 7 4 3 0 0.571 -6

The Patriots are +7 in turnover margin so far this season, which helps cover up a lot of that ugly yardage deficiency. The other ingredients in their surprising performance? Special teams (2 kick return TDs, the NFL's 3rd-best net punting avg and kickoff avg), 3rd-down offense (the NFL's second-best 3D conversion %), an opportunistic scoring defense (2 INT return TDs), and a mere 2.2 opponent yds/carry allowed in the red zone.

The problem for the Patriots going forward is that most of these factors are notoriously influenced by randomness, and therefore tend not to be consistently repeatable in the future. Regarding turnover margin, the recovery of fumbles is almost totally based on luck, and the ability to avoid interceptions is only slightly less random. Return TDs aren't predictive events, either, and 3rd-down & red-zone performances tend to be highly volatile because of the sample sizes involved. No, kickoff and punting averages are not as prone to regression, but the rest of New England's success this year is based on strong performances in largely unpredictable statistical categories.

I'm not saying Belichick and the Patriots can't find a way to secure more reliable paths to victory in the future, but right now they are one of ten teams since the merger to win 5 or more of their first 7 games despite a total yardage differential of -300 or worse:

Tm Year G W L T W-L% Rush Pass Tot TO
GNB 1978 7 6 1 0 0.857 19 -340 -321 14
CIN 1990 7 5 2 0 0.714 81 -412 -331 1
SEA 1999 7 5 2 0 0.714 -186 -159 -345 11
NWE 2010 7 6 1 0 0.857 78 -454 -376 7
TAM 1995 7 5 2 0 0.714 -143 -255 -398 6
WAS 1976 7 5 2 0 0.714 -59 -385 -444 2
DET 2000 7 5 2 0 0.714 -203 -317 -520 11
DEN 1984 7 6 1 0 0.857 170 -759 -589 13
CHI 1971 7 5 2 0 0.714 -353 -277 -630 8
DEN 1992 7 5 2 0 0.714 -292 -513 -805 1

Those teams went a combined 32-44-1 the rest of the season, for a .422 winning %. If the Patriots suffer the same fate, they would be on pace to win 10 games and probably replicate last year's accomplishments (right down to the 1st-round playoff loss), which would come as a major disappointment to fans who looked at their 6-1 record on November 1 with visions of the Lombardi Trophy in their heads. But the simple truth is that these Pats just don't have the same sturdy statistical foundation as the Super Bowl-winning versions of 2003-04, and to think otherwise is to ignore the evidence.

This entry was posted on Monday, November 1st, 2010 at 1:42 pm and is filed under PI Finds, Play Index, Site Features. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.