Estimated Win Differentials from Football Outsiders

Good post. I don't understand their charts very well either, but if what you are saying is correct let's all hope it works out. It would be nice like last season to at least have some meaningful games at the end of the year.

Kinda interesting to see what they predict. I remember seeing these guys predicting the Niners would get more than 6 wins this season. At the time, I thought they were smoking crack...........Now they look pretty spot on

Originally posted by nickbradley:
Football Outsiders puts together an analysis of estimated wins based on performance and opponent strength. Based on these numbers, we are the most underperforming team in the NFL, while the Arizona Cardinals are the biggest over-achievers.

so based on performance, including turnovers and penalties, strength of opponent, etc., we should be 3-3 and Arizona should be 1-4.

Statistically, both of us should revert back to the mean.

Also note that we have had the strongest schedule to date and have the weakest remaining schedule (based on opponent performance) in the NFL.

so statistically, we should rip off a lot of wins here, while Arizona should fall on their face.

http://www.footballoutsiders.com/stats/teameff

This is a really good post and only confirms what I've been thinking. Our performance isn't any worst than last year. If you look at where our wins came from, 5 out of 8 came from our division and I believe the other three where loser teams (Bears, Jax, Lions). So in reality, we just went through our regular mid season skid, since the beginning of the year was the toughest part of our schedule. How is our lost to the Saints/Eagles any diffrent from the Colts and Vikings of last year? The problem with us is we always play to the level of our opponent. The only thing that concerns me is we are a horrible road team and it's been noted the Niners travel more miles than any other team in the league this season.

[ Edited by Oakland-Niner on Oct 18, 2010 at 18:55:24 ]

Originally posted by FourNine49:
Who cares about predicting wins/losses, numbers, probabilities, etc?

Any given Sunday, you win or lose

OK coach singletary...

Originally posted by nickbradley:

Originally posted by FourNine49:
Who cares about predicting wins/losses, numbers, probabilities, etc?

Any given Sunday, you win or lose

OK coach singletary...

Lol, coach sing comes to the zone for advice.

Originally posted by Snider8706:

Originally posted by nickbradley:

Originally posted by FourNine49:
Who cares about predicting wins/losses, numbers, probabilities, etc?

Any given Sunday, you win or lose

OK coach singletary...

Lol, coach sing comes to the zone for advice.

no he doesn't, if he did he would start using Westbrook and Dixon more, bench Rachal for Snyder and Morgan for Zeigler. He would also stop being such a p***y once the team gets the slightest lead. Sorry "Coach" you can't sit on a 4 point lead.

Sweet, good info, thanks

Originally posted by SJniner7:

Originally posted by nickbradley:

Originally posted by WillistheWall:
Didn't they say before the season we'd get like 4 wins and be the worst team in the division?

That's based on projecting the prior year's performance and is not very accurate.

This is current year.

If I remember correctly, it was based on the fact that we had so many forced fumbles lasst year, and they said that was an anomaly. As far as I can think of, I do not know if we have recovered any of the forced fumbles this year...

It wasn't the forced fumbles. It was the fumble recoveries. Forcing fumbles is a skill, but recovering fumbles is random according to FO. So according to them, our defense was expected to regress this year (which we've seen) while our offense was not expected to improve dramatically.

And while you can argue with their stats all you want, they are more accurate than any other statistical measure out there.

Interesting stuff. I really hope we can find a winning streak starting...this Sunday.

10am road games have never been favorable for us. When was the last time we won a road game, let alone a 10am road game?

Too Bad Statistics dont throw the ball for us!

Originally posted by Oscar:

Originally posted by SJniner7:

Originally posted by nickbradley:

Originally posted by WillistheWall:
Didn't they say before the season we'd get like 4 wins and be the worst team in the division?

That's based on projecting the prior year's performance and is not very accurate.

This is current year.

If I remember correctly, it was based on the fact that we had so many forced fumbles lasst year, and they said that was an anomaly. As far as I can think of, I do not know if we have recovered any of the forced fumbles this year...

It wasn't the forced fumbles. It was the fumble recoveries. Forcing fumbles is a skill, but recovering fumbles is random according to FO. So according to them, our defense was expected to regress this year (which we've seen) while our offense was not expected to improve dramatically.

And while you can argue with their stats all you want, they are more accurate than any other statistical measure out there.

Yes, their in-year stats are the best out there, but their future year predictions have been terrible. I believe they've picked the Rams to be the breakout team for 3 years now.

They predicted the Saints to revert to the mean as well due to all the fumbles they recovered last year -- calling for only 9 wins by NO.

I like FO...I also like Pro-football focus and Advanced NFL stats. All of them have their own way of using statistical anaylsis, but they generally have the same opinion of the 49ers.

I remember reading FO's article on the 49ers defense regressing from last year because the amount of fumble recoveries and thinking that it was a load of crap, but they've been right for the most part.

It should also be noted that FO predicted the 49ers would finish last in the NFC West. Well, so far, they're right. Link

Quote:
From my conversation with Barnwell, here are four trouble spots as identified by Football Outsiders:

1. The offense has an identity crisis, with a head coach who likes to be able to pound the ball and a quarterback who operates best out of the spread. “It’s like they can’t decide what they want to do,’’ he said.

2. The defense is due for a regression to the mean. The 49ers forced 21 fumbles last season, which the Outsiders see as a statistical outlier unlikely to be duplicated. Ahmad Brooks, for example, forced four fumbles as a backup linebacker – a feat accomplished only twice previously in NFL history. The Outsiders don’t see it happening again.

3. The guys up front were dead last in Offensive Line Yards last year. While drafting Anthony Davis and Mike Iupati in the first round was a good move for the long term, rookie linemen are likely to have growing pains. That, plus an early lack of cohesion up front spell trouble.

4. Alex Smith’s numbers are deceptive. Smith established a career high with 18 TDs and an 81.5 quarterback rating, but Barnwell argues that he did it against “pretty much every weak defensive team.”

These guys called for the niners to be 6-11 before the season started and we all thought they were nuts. This was a good read

Originally posted by DesiDez:
I like FO...I also like Pro-football focus and Advanced NFL stats. All of them have their own way of using statistical anaylsis, but they generally have the same opinion of the 49ers.

I remember reading FO's article on the 49ers defense regressing from last year because the amount of fumble recoveries and thinking that it was a load of crap, but they've been right for the most part.

It should also be noted that FO predicted the 49ers would finish last in the NFC West. Well, so far, they're right. Link

Quote:
From my conversation with Barnwell, here are four trouble spots as identified by Football Outsiders:

1. The offense has an identity crisis, with a head coach who likes to be able to pound the ball and a quarterback who operates best out of the spread. “It’s like they can’t decide what they want to do,’’ he said.

2. The defense is due for a regression to the mean. The 49ers forced 21 fumbles last season, which the Outsiders see as a statistical outlier unlikely to be duplicated. Ahmad Brooks, for example, forced four fumbles as a backup linebacker – a feat accomplished only twice previously in NFL history. The Outsiders don’t see it happening again.

3. The guys up front were dead last in Offensive Line Yards last year. While drafting Anthony Davis and Mike Iupati in the first round was a good move for the long term, rookie linemen are likely to have growing pains. That, plus an early lack of cohesion up front spell trouble.

4. Alex Smith’s numbers are deceptive. Smith established a career high with 18 TDs and an 81.5 quarterback rating, but Barnwell argues that he did it against “pretty much every weak defensive team.”

Yes -- you can clearly see the deviation from the norm in defensive turnovers in 2009, where we had a 2006-esque performance on defense...about 1 out of every 32 opponent plays resulted in a turnover against the 49ers in 2009.

But as you'll also see, we also practiced excellent ball control on offense in 2009. If we practiced the same level of ball control this year, we should have 5.5 less turnovers by this point...that probably equates to 2 - 3 more wins (probably 2):

- 1/29 offensive turnovers/play in 2006 // 1/38 defensive turnovers/play in 2006

- 1/27 offensive turnovers/play in 2007 // 1/51 defensive turnovers/play in 2007

- 1/27 offensive turnovers/play in 2008 // 1/57 defensive turnovers/play in 2008

- 1/39 offensive turnovers/play in 2009// 1/32 defensive turnovers/play in 2009

- 1/24 offensive turnovers/play in 2010// 1/52 defensive turnovers/play in 2010

In 2009, we were 1/50 under Shaun hill and 1/36 under Smith. This year, we are committing more offensive turnovers than we did with Mike Martz! But at least under Martz we averaged an extra yard per pass attempt (that's a lot -- about 40 more passing yards per game through this point).

so what I'm saying is that if we can protect the ball on offense for the rest of the year, we probably finish with 8 wins.

Here's a chart for ya

Turnovers = L
None = W