Response to the stat-heads predicting 2012 niners will regress

Up on Grantland right now is an article which basically sums up all the reasons why most stat-heads think the niners will regress in 2012. (http://www.grantland.com/story/_/id/8237657/san-francisco-repeat-last-season-greatness) For those that buy into this sort of predictive analysis, here's a response to consider.

The author of the Grantland piece uses a few statistics to make his prediction that the niners will regress and finish around 8-8 this year. Here are some of them, summarized:
2011 49ers
Record: 13-3, up from a 6-10 record last year (+7 wins)
Turnover differential: +28 (#1 in NFL)
6-2 record in close games (games decided by 7 or less)
Pythagorean record: 12.3-3.7 (Pythagorean record: predicting a team record based on ratio of points scored vs. points allowed, so we outperformed our "expected record" by a game)

All three of those things fluctuate wildly from season to season, says the author. In other words, being good at them for one year probably doesn't portend continued success going forward.

But consider another example, "Team A".
Team A
Record: 13-3, up from 6-10 the prior year (+7 wins)
Turnover differential: +23 (#1 in NFL)
7-2 record in close games
Pythagorean record: 11.2-4.8 (+2 wins from expected record)

Team A is due for an even bigger regression, right? No hope of continued success. Except...Team A is the 1981 49ers, whose season marked the start of one of the great runs of dominance ever.

Food for thought when reading about the many niners doubters this offseason.

Love it!

Here is the reason why we wont regress

Originally posted by johnnyredneat:
Up on Grantland right now is an article which basically sums up all the reasons why most stat-heads think the niners will regress in 2012. (http://www.grantland.com/story/_/id/8237657/san-francisco-repeat-last-season-greatness) For those that buy into this sort of predictive analysis, here's a response to consider.

The author of the Grantland piece uses a few statistics to make his prediction that the niners will regress and finish around 8-8 this year. Here are some of them, summarized:
2011 49ers
Record: 13-3, up from a 6-10 record last year (+7 wins)
Turnover differential: +28 (#1 in NFL)
6-2 record in close games (games decided by 7 or less)
Pythagorean record: 12.3-3.7 (Pythagorean record: predicting a team record based on ratio of points scored vs. points allowed, so we outperformed our "expected record" by a game)

All three of those things fluctuate wildly from season to season, says the author. In other words, being good at them for one year probably doesn't portend continued success going forward.

But consider another example, "Team A".
Team A
Record: 13-3, up from 6-10 the prior year (+7 wins)
Turnover differential: +23 (#1 in NFL)
7-2 record in close games
Pythagorean record: 11.2-4.8 (+2 wins from expected record)

Team A is due for an even bigger regression, right? No hope of continued success. Except...Team A is the 1981 49ers, whose season marked the start of one of the great runs of dominance ever.

Food for thought when reading about the many niners doubters this offseason.

Well technically they did regress in 1982, finishing with a 3-6 record. I think our success this season will be based on winning the division and performing well in the playoffs, all of which can be done without going 13-3.

Why is it "stat-heads" that think the Niners will regress? Why isnt it "historians" that think the Niners will regress?

Look, I personally dont think we will regress, because I think we are THAT good. But based on history, teams that did something similar to what we did last year, did in fact regress. I just dont understand why OP and others feel the need to ridicule people who make the claim that "based on history, it is LIKELY that we regress this year."

I dont think its going to happen.....I hope it doesnt.....but I totally understand why one would think it will.

That guy looks like he belongs playing his x box and not writing sports articles. Idiot for not looking at the clear elite talent across the board on our team that extends up through the ranks to coaches and player personell.

We may regress in terms of our record. I predict an 11-5 record this year. But overall, I think we will be a better team than last year. And the last few SuperBowl winners have shown that all that matters is making it to the playoffs.

Then there is the whole injury situation. If we were to lose Alex, Justin, Vernon, or Patrick to a severe injury, we would definitely take a hit.

But that's why they play the games.

And who knows, we could end up being BETTER than last year too.

ya, were gonna regress because all 11 defensive starters are back, we have better value in our depth at every position than most teams have with their starters, our offense and defense will be in the same system for two years and they will have a full TC to install everything unlike last year, we fully upgraded our receiving corps and RB's and we were an overtime mistake away from the superbowl, do you think harbaughs gonna let them forget that? theyre gonna be hungrier than ever this year after tasting success. we look stronger any way you want to look at it at every postition. Y'all are f**ked nfl, the niners are coming for everyone this year and theres no place you can hide!!!

As someone who works in science, it's really difficult to get through a lot of these "stathead' articles without shaking my head. When you point to a historical average as precedent for why a team will "regress to the mean," it's absolutely pure lazy reasoning. If you're a "football analyst," then analyze those individual teams. For example, Peyton Manning's teams repeatedly win a larger than expected proportion of their close games. Why? Because Manning is f**king good.

Originally posted by dhp318:
As someone who works in science, it's really difficult to get through a lot of these "stathead' articles without shaking my head. When you point to a historical average as precedent for why a team will "regress to the mean," it's absolutely pure lazy reasoning. If you're a "football analyst," then analyze those individual teams. For example, Peyton Manning's teams repeatedly win a larger than expected proportion of their close games. Why? Because Manning is f**king good.

Originally posted by dhp318:
As someone who works in science, it's really difficult to get through a lot of these "stathead' articles without shaking my head. When you point to a historical average as precedent for why a team will "regress to the mean," it's absolutely pure lazy reasoning. If you're a "football analyst," then analyze those individual teams. For example, Peyton Manning's teams repeatedly win a larger than expected proportion of their close games. Why? Because Manning is f**king good.

Thank you!

If we could simulate this season an infinite amount of times, statistically we would regress from a 13-3 season. However, there is only one reality and most 13-3 teams dont return their entire starting defense and all but the worst player on offense .

Originally posted by johnnyredneat:
Up on Grantland right now is an article which basically sums up all the reasons why most stat-heads think the niners will regress in 2012. (http://www.grantland.com/story/_/id/8237657/san-francisco-repeat-last-season-greatness) For those that buy into this sort of predictive analysis, here's a response to consider.

The author of the Grantland piece uses a few statistics to make his prediction that the niners will regress and finish around 8-8 this year. Here are some of them, summarized:
2011 49ers
Record: 13-3, up from a 6-10 record last year (+7 wins)
Turnover differential: +28 (#1 in NFL)
6-2 record in close games (games decided by 7 or less)
Pythagorean record: 12.3-3.7 (Pythagorean record: predicting a team record based on ratio of points scored vs. points allowed, so we outperformed our "expected record" by a game)

All three of those things fluctuate wildly from season to season, says the author. In other words, being good at them for one year probably doesn't portend continued success going forward.

But consider another example, "Team A".
Team A
Record: 13-3, up from 6-10 the prior year (+7 wins)
Turnover differential: +23 (#1 in NFL)
7-2 record in close games
Pythagorean record: 11.2-4.8 (+2 wins from expected record)

Team A is due for an even bigger regression, right? No hope of continued success. Except...Team A is the 1981 49ers, whose season marked the start of one of the great runs of dominance ever.

Food for thought when reading about the many niners doubters this offseason.

No offense, but you're responding to a probability model with an anecdote. I'd like the 9ers to go 13-3 again too, but you lost before you started.

Anybody who knows anything about the 49ers can see they have improved on offense and on defense, and also due to the fact that this is the first actual offseason with the freaking coaching staff.

Anybody who knows nothing about the 49ers will just quote random statistics and say their schedule is tougher and that they only won because they had lots of turnovers last year and won't do that again.

The inherent problem with all these doom and gloom articles is they never account for potential offensive growth or effectiveness.

If the offense is more effective especially in the 3rd downs and Red Zone than:

A) There are likely to be less close games to win or lose because there will be less close games.

B) Turnover differential will be less important because more scoring gives you some breathing room to make some mistakes.

Alex Smith might throw for 10 INTs but what if he passes for 25+ TDs and 4,000 yards?

I guess the Packers are going to suck too because Rodgers only threw 6 INTs, Packers were 4-1 in close games, +24 turnover differential is unsustainable, and no way that defense stays healthy.

Who cares? Most admit that they think we'll make the playoffs and that's really what matters.