The problem with eyeball tests is that they're completely subjective. You see one thing, I see another. The value of statistics is that they're objective, everyone can see them, and they can even tell you things about teams and games you haven't eyeballed. But if the stats consistently disagree with your eyeballing, you'll discount them, either because they're bad or because you've got a biased view. The holy grail is a statistical system that correlates well with most peoples eyeballs. I've found the Football Outsiders system to be the best in that regard.
We also all have our prejudices, and they tend to affect what we see. You seem to always see problems with this year's wr's not stretching the field; NEM always sees bad offensive play calling. You can fairly accuse many of the rest of us as seeing too much good in the outcome of any Patriots game.
Statistics can sometimes alert us to our prejudices, if we only let them.
Kasmir, I agree with your comments on the FO stats. The other posts have been interesting as well. A fun discusion.
So here's the FO stats from the 2006 team (so far), compared to the 2004 and 2003 teams. Explanations of the methods are
here. All percentages are compared to league average, established over the most recent set of seasons, so they're comparable from year-to-year. Schedule strength is the average of all the total DVOA's of all the opponents from that year. Negative is bad for offense and ST and good for defense. Ranks in parens.
2006
------
Total DVOA: 22.3% (5th)
Off. DVOA: 6.3% (8th)
Def. DVOA: -11.4% (5th)
S.T. DVOA: 4.6% (3rd)
Schedule
Strength DVOA: -1.9% (22nd)
Variance: 21.0% (4th)
2004
------
Total DVOA: 35.7% (1st)
Off. DVOA: 24.2% (4th)
Def. DVOA: -11.3% (6th)
S.T. DVOA: 0.2% (16th)
Schedule
Strength DVOA: 4.9% (8th)
Variance: 17.3% (10th)
2003
------
Total DVOA: 22.8% (3rd)
Off. DVOA: 0.4% (14th)
Def. DVOA: -22.0% (2nd)
S.T. DVOA: 0.4% (17th)
Schedule
Strength DVOA: 0.6% (16th)
Variance: 15.0% (20th)
Here's the picture I see from these numbers (with a couple comments from my eyeballs):
2003 was a dominant defense (2nd to BAL) with a mid-pack offense (14th, go Antoine go!) that combined to get the job done. They played a mid-pack schedule (16th). They were reasonably consistent (20th in Variance, lower ranks, more consistent, higher ranks less consistent). The big thing is that this was the Judge Smails defense: "You'll get nothing and like it!" Law and Harrison were healthy, so the secondary was never better in the SB era.
2004 was a dominant team overall, with a MUCH better offense (24.2%, 4th to IND, KC, MIN) and a defense that slipped a bit. A tough schedule (8th) left no doubt about who the best team was, on a play-by-play basis. This Pats offense was a ton better than in 2003 (0.6%), almost twice as good as in 2005 (12.8%) and is almost 4 times better than this year so far (6.3%). We were spoiled by this offense, which had Branch and Givens at their peaks, as well as Dillon with an "Indian Summer" season in his career. The execution of the offense was such that it could do whatever it pleased. Brady and Weis were like kids in a candy store. As a whole, the team was a bit more up and down than 2003 (Variance was 10th), but by and large they could bring it whenever necessary. Defense had health problems and slipped numerically to half as good as 2003, but still was plenty enough to win, relative to the competition (ranked 6th in the league).
2006 has a team with a defense on par with 2004 (-11.4%) with a similar rank (5th). Although the offense is in the top 10 (6.3%, 8th), it is far off the pace of 2004 (24.2%). It's way ahead of 2003 (0.4%) and slightly behind 2005 (12.8%). The schedule strength is in the bottom half of the league (22nd). Much easier than in 2004 (8th) and slightly easier than 2003 (16th). The big thing about 2006: the team is VERY up and down (Variance of 21%, 4th in the league). And that's the problem I see with this current team. We just don't know which Pats team is gonna show up. In 2003, you could hang your hat on the defense. In 2004, on the offense. When the dominant squad in a given year had a bad game, most of the time the other squad bailed it out. My eyeballs tell me this year we've had more synchronized suckage and less synchronized poundage on both sides of the ball than in 2003-2004.
So, that's not to say the Pats can't make the SB this year, but so far they've been inconsistent and haven't conclusively shown that they can, which too me was the case in 2003-2004. Obviously, the playoffs are likely, and anything can happen there. And although I can see faults in this team, I'll just say that I'm totally enjoying the fact that our worst down year in this 6-year run is 9-7 and losing the playoffs on tiebreakers. Notice the "runs" by the other SB and playoff teams in the league over this stretch (e.g. IND, PIT, PHI, SEA, OAK, TB).
FYI, the Jags are first in variance, so that means anything could happen this Sunday.
