Regression analysis is the process of having a bunch of data points, say, the defensive production of a particular team for the last x years, and making a polynomial formula that fits all the data, so you can predict future trends. Volatility is basically how close to the average an individual instance is, over time. So if I have a 6 handicap, and it's made up of 100 consecutive games scoring 6, that's low volatility. If it ranges day-to-day between -2 and 14, that's high volatility. There, so now you don't need to put scare quotes around "regression". As for the media treating other team's defenses like they're etched in granite, but the Jets is written with gunpowder that's just been lit, you know that's how they do.

@@jessejordache1869 I've studied statistics but thanks for paraphrasing google's definition of it. You want to give me a primer on proof that the beta coefficient is an unbiased estimator, R squared, or lurking variables too? Or instead of that, can you maybe admit that we're not looking at random samples? I think I'll keep the quotation marks since people that do this operate in a black box and don't put out how they arrived to that point so it can be scrutinized. This isn't an econometrics class or a study of congestion pricing put out for peer review. This is someone from PFF saying "well they're due for a regression" and leaving it at that. Thanks for playing Jesse, see you in the comment section next time.

@@lightxgrenade I've got a math degree, but statistics isn't my thing. The purpose of my post was because I didn't know what regression analysis was for the longest time, and kind of wished I did. Your critique ain't wrong -- "due for a regression" isn't a thing. Plus advanced stats are in a lot of cases the modern version of sliding an index card to a baseball announcer, so that he can point out that the current pitcher had grounded out the most switch hitters at that particular stadium in history. I believe that defense is more volatile than offense, RELATIVELY, because having a franchise QB, or not having one, is a big stabilizing factor that defense doesn't have. But most of this stuff is a case of insufficient sample size.

@@jessejordache1869 The first week of my econometrics class, my professor showed us a quote by a famous statistician, George Box: All models are wrong, some are useful. The point was to not only concern yourself with if the model can depict a relationship reasonably accurately but if the methodology behind the model was appropriate. That always stuck with me. So many companies refuse to detail how they construct their models because they're in it for money and entertainment - that to me are inherent biases that kinda forces a fit for the data. Not to mention that human interpretation of a model's results can often be biased even if the model is fairly accurate. If you ever get a chance to read a paper on economic research note how cautious the writer is to make any concrete assumptions based on their findings. If you're interested in models and how blind faith in it can be dangerous, you might like looking into the Black-Scholes model. Partly developed by Fischer Black, one of the preeminent economists of the 20th century. It's a model for pricing options. By itself, it's a good model, but investors relied too heavily on it, not taking into account the limitations of the model and it played a significant role in the financial crisis of '08. It wasn't the main factor, but it was like adding fuel to an already out of control fire. And while I agree that in general, a defense can be more volatile year over year, I would argue that it's a lot less sensitive to one or two key players having a down year than an offense where if a QB or star receiver has a down year, it can alter the entire outlook for it because they're such a significant portion of an offense's success. If Quinnen Williams or Sauce Gardner got injured in game 1, would the defense drop off as badly as the offense did when Aaron Rodgers got injured?

Saleh develops the scheme and scouting for defensive players. Brich enforces and executes the scheme. I want our offense to have an identity that our defense has. Fast, smart, and nasty

@@joshuamantock4037 I'm with ya, I hope our offense is out to punish fools, and I think we will. That Braelon Allen pick showed our identity, run first. Dice em up with Breece and wreck em with Allen. I'm excited!