RayNAustin wrote:Irn-Bru wrote:I don't think it's the best measure; in fact, it has as many problems as judging defenses by yards allowed. (Yards allowed probably is the better measure if you only have a choice between those two, however.)
I'm having trouble getting my head around that ... so you're saying, you'd prefer your defense to give up less yards rather than fewer points? You cannot be serious?

Who said anything about that? I'm talking about ways to measure, statistically, the effectiveness of a defense over many games against various opponents in different conditions. To talk about yards allowed being a better metric than points allowed for those purposes doesn't mean I'd rather have a defense that gave up fewer yards but allowed lots of points. This is like football stats 101.
There are many reasons that a team might give up a lot of yards, as I indicated before ...
As I indicated in my last post,
I do not advocate using yards allowed as the best metric for ranking defenses. I did point you to two sites where you can see, dissect, agree/disagree with more advanced analysis of NE's defense. Both sites, which use different methods, have the Pats in the bottom third of the league.
I prefer wins over statistics,
That's your prerogative. But it relegates your arguments to the banter you get between drunk fans in a bar: lots of chest thumping or "look at the scoreboard" but no depth or really anything of interest going on. Anyone can say "give me a defense that holds 'em to less than 14 points, any day," but that's not a realistic goal in this sport and doesn't tell you anything about the way the game is actually played.
I tell you what, I found the ultimate stat connected with winning, and you'll probably love it: I call it the outscore stat. Teams that can figure out a way to outscore their opponents, no matter who, put themselves in the position to win. So you can scrap your points-allowed fixation with defensive stats. I've got the inside track on the statistic that matters, and it ain't yards or points allowed or any of that nerdy mumbo jumbo. Give me the team that knows how to go out and score more points than the other team, any day.
Any argument against this will have to show me how their stat better predicts wins than my outscore index. (I've yet to see someone who can do it.) Whatever other fancy schmancy stats you want to discuss — and I'm sure there are plenty of stats like points allowed per game that are interesting to talk about on a purely theoretical basis — I'll stick with the one that's proven the best predictor of NFL success, thank you very much.
After all, to quote a wise man I once heard, "I prefer wins over statistics."
Those complex statistical factors may make for interesting conversation and debate ... but at the end of the day ... it's any given Sunday, and personnel matchups and execution that determines the course of the game. Some teams simply match up better against other teams that would generally be considered more talented.
You make it sound like "any given Sunday" is the realm of the analysts who reject statistics. But that's not the case. People who look at advanced statistics don't think in terms of certainties — actually the whole point of using the statistics is because outcomes are uncertain. So this whole line of argument is misleading.
You mean, harm them all the way to the Super Bowl? Come on now, you pulling my leg here ... I feel a tug.
So you're one of those people who think the 2000 Ravens offense was top quality?