Busting the Turnover Myth

[Oldfan]

Following the loss to the Rams, the turnover myth was revived in this forum.

The myth: The turnover ratio is the most telling football stat (for some mysterious reason).

Mike Tanier at footballoutsiders.com wrote a facetious article to illustrate the principle in statistics that correlation does not imply causation. He first showed a high correlation between QB kneeldowns and wins and then concluded with tongue in cheek that it would be a winning strategy for QBs to take a knee more often.

The dreaded turnover mantra has been chanted by some famous coaches, including our own Joe Gibbs. They give us the high statistical correlation to wins as though the data explains itself. It doesn't. It's just a correlation that requires further study to determine significance.

Most turnovers are forced by the defense. They are caused, just as touchdowns are caused by the offense. Obviously, the offense should try to prevent them, just as the defense should try to prevent touchdowns. Over the course of a season, the teams with better coaches and players will have a positive turnover differential.

Let's try to take the mystery out of this. How much is the turnover worth?

According to “The Hidden Game of Football” (pp. 143-144), the average turnover is worth four points. That's equal to the difference between a field goal and a touchdown. So, when a team in the red zone fails to score and settles for a field goal, that's roughly equivalent to a turnover. Big plays (40+ yards) are also worth about four points...the equivalent of a turnover.

If we accept the extra point as automatic, touchdowns are worth seven points. So the value of touchdowns is 75% more than turnovers, yet we don't hear experts talking about winning the touchdown battle as the key to the game even though the correlation to wins is higher than the turnover ratio.

When fans overvalue turnovers, there's no real harm done. When coaches do it, it can lead to over-aggressiveness on defense and conservatism on offense. Coaches will prefer game manager QBs and will undervalue the gunslingers. QBs are trained to avoid mistakes which results in many becoming tentative, leading to fewer interceptions but more sacks and fumbles, and lower production.

The average turnover is worth about four points, roughly the same as a big play or a failed attempt to score from the red zone.

[LD0506]

Ok, so are you saying it is more productive in the long run to just throw it out there and go for the TD and not worry if it gets picked, since statistically the successful TDs will more than offset any INTs? As in 5 TDs more than makes up for 8 INTs?

[DiscoBob]

So in the Rams game they won the turnover battle 3-1

So according to this article, that resulted (in points) in 12 - 4

So without turnovers the score would have been 13 - 7 Redskins win....

It seems like they made a difference huh?

[Rudechain]

But if you turn the ball over, you lost a chance to score. So in my book, the turnover statistic is important.

[TheDoyler23]

When you turn it over 3 times in/near FG range and one of those goes for 7 the other way, then turnovers absolutely matter. That's a 16 point difference.

Maybe we don't hit all three fg's, maybe we score a TD or two, but the turnovers took their toll. Especially since the Rams couldn't move the ball.

[Oldfan]

Ok, so are you saying it is more productive in the long run to just throw it out there and go for the TD and not worry if it gets picked, since statistically the successful TDs will more than offset any INTs? As in 5 TDs more than makes up for 8 INTs?

Risk and reward should be decided on something more intelligent than on a turnover phobia due to a low math aptitude. That's what I'm saying.

[nightbird]

I've never read "the hidden game of football" but this doesn't make any sense.

Whatever algorithm or equation they are using is wrong.

For one thing, not all turnovers are alike. An interception on a hail mary just before halftime is meaningless.

A fumble in the red zone that gets returned for a touchdown is a swing of 10 or 14 points.

There is no "average turnover." It's not the same as a touchdown, which is always 7, or a field goal which is always 3.

[Oldfan]

So in the Rams game they won the turnover battle 3-1

So according to this article, that resulted (in points) in 12 - 4

So without turnovers the score would have been 13 - 7 Redskins win....

It seems like they made a difference huh?

Of course, but why not include the big plays and the failed attempts in the red zone in your calculations? Why isolate turnovers as being so special?

[Oldfan]

But if you turn the ball over, you lost a chance to score. So in my book, the turnover statistic is important.

Sure it is. Its value is four points roughly.

[TheGreek1973]

So in the Rams game they won the turnover battle 3-1

So according to this article, that resulted (in points) in 12 - 4

So without turnovers the score would have been 13 - 7 Redskins win....

It seems like they made a difference huh?

Actually you also need to erase our first td since it came from a turnover. 7-0. IN ADDITION our 3 turnovers, where they occured we would have had at least 9 more points thru FGs if we didn't score. Thus 9+7= 16. In otherword at minimum we should have won the game 16-12. Anyway you look at it turnovers gave this game to the RAMS.

[rebornempowered]

Was this article co-written by Brett Favre and Tony Romo sits to pee?

Just curious.

[Oldfan]

I've never read "the hidden game of football" but this doesn't make any sense.

Whatever algorithm or equation they are using is wrong.

For one thing, not all turnovers are alike. An interception on a hail mary just before halftime is meaningless.

A fumble in the red zone that gets returned for a touchdown is a swing of 10 or 14 points.

There is no "average turnover." It's not the same as a touchdown, which is always 7, or a field goal which is always 3.

Imagine a football field. Then, imagine a Bell Curve over it. The value is about six points at the one-yard line and lowers incrementally until its lowest point at the 50. Interceptions have more value than fumbles.

The word "average" doesn't mean that the value is always the same.

[Oldfan]

So in the Rams game they won the turnover battle 3-1

So according to this article, that resulted (in points) in 12 - 4

So without turnovers the score would have been 13 - 7 Redskins win....

It seems like they made a difference huh?

Yes, so did the touchdowns, FGs, big plays and failed red zone attempts. Why talk only about turnovers as being the difference?

[LD0506]

Risk and reward should be decided on something more intelligent than on a turnover phobia due to a low math aptitude. That's what I'm saying.

Don't be snotty, just wanted a clarification.

This is going to be another one of those threads where you take a stat in isolation and try to prove something earthshaking again, isn't it? Like the aerodynamic proofs that show a bumblebee simply cannot fly.

More TDs + more TOs ≠ fewer TDs but no TOs

[IbleedBnG83]

There are intangibles that can't be accounted for...things like momentum IMO make a huge difference.

In the end, the team that wins the turnover battle typically wins.

[Lombardi's_kid_brother]

I would break this down further.

I think fumbles and interceptions need to be viewed differently since - as Football Outsiders has proved - recovering a fumble is pure luck. In reality, a defense recovering a fumble is a 50/50 proposition and has pretty much nothing to do with skill.

Stripping the ball is a skill. Holding onto the ball is a skill. Pouncing on the ball as it is bouncing all over the place is not a skill. There is no correlation whatsoever between the percentage of fumbles recovered by a team in one year and the percentage they recover in the next year. The odds of recovery are based solely on the type of play involved, not the teams or any of their players.

Fans like to insist that specific coaches can teach their teams to recover more fumbles by swarming to the ball. Chicago's Lovie Smith, in particular, is supposed to have this ability. However, since Smith took over the Bears, their rate of fumble recovery on defense went from a league-best 76 percent to a league-worst 33 percent in 2005, then back to 67 percent in 2006. Last year, they recovered 57 percent of fumbles, close to the league average.

Fumble recovery is equally erratic on offense. In 2006, the Detroit Lions fumbled 21 times on offense and recovered just four of those fumbles. Last year, the Lions fumbled 29 times on offense--but actually had fewer turnovers because they recovered 16 of those fumbles.

Fumble recovery is a major reason why the general public overestimates or underestimates certain teams. Fumbles are huge, turning-point plays that dramatically impact wins and losses in the past, while fumble recovery percentage says absolutely nothing about a team's chances of winning games in the future. With this in mind, Football Outsiders stats treat all fumbles as equal, penalizing them based on the likelihood of each type of fumble (run, pass, sack, etc.) being recovered by the defense.

Other plays that qualify as "non-predictive events" include blocked kicks and touchdowns during turnover returns. These plays are not "lucky," per se, but they have no value whatsoever for predicting future performance.

http://www.footballoutsiders.com/info/FO-basics

In the game against the Rams, the Redskins lost 75 percent of their fumbles. If that number is 50 percent, they probably win.

I don't think Football Outsiders has done this yet, but I'm farily confident in thinking that interceptions are worth more points than fumbles since EVERY interception results in a turnover.

But I am thinking that recovered fumbles are more catastrophic than interceptions due to the filed position.

And finally that fumbles are far more randomized than interceptions.

I mean, anyone who thinks that Jason Campbell's interception-less streak is in any way sustainable is probably fooling themselves.

[Lombardi's_kid_brother]

There are intangibles that can't be accounted for...things like momentum IMO make a huge difference.

Yea...no such thing.

I think Oldfan is on to something, but I think he is viewing this too specifically.

[Oldfan]

Was this article co-written by Brett Favre and Tony Romo sits to pee?

Just curious.

<smile>

It does suggest that some coaches undervalue gunslingers. Think Joe Gibbs would have put up with a young Favre or a young Tiki Barber's fumbling?

[zCommander]

All I know is no one is going to win the battle of turnovers in this thread against OF.

[IbleedBnG83]

Yea...no such thing.

I think Oldfan is on to something' date=' but I think he is viewing this too specifically.[/quote']

No such thing? You are saying that team momentum on either side of the ball doesn't exist? I'm guessing you have never played a sport then?

[IbleedBnG83]

I find it funny that some random person decided to quantify these stats with numbers. What makes them right? Anyone can sit and analyze these things and try to quantify them as much as possible, but the reality is they can't be.

The team that wins the turnover battle typically wins the game...thats it. Not much too it. Perhaps some put too much stock into, sure. But there is no reason to further investigate this concept.

[Lombardi's_kid_brother]

No such thing? You are saying that team momentum on either side of the ball doesn't exist? I'm guessing you have never played a sport then?

Yes. There is no such thing as momentum. There is no such thing as clutch. There is no such thing as a jinx. There is such a thing as "luck" but it is randomized.

When idiot announcers say things like, "Boy, the Giants have all the momentum now" what they are really saying is, "Boy, a random combination of luck, poor play by the opponent, and good play by the Giants has put them in a position to succeed in this game where otherwise they may not have succeeded."

But John Madden would take 25 minutes to spit out that sentence.

[Lombardi's_kid_brother]

I find it funny that some random person decided to quantify these stats with numbers. What makes them right? Anyone can sit and analyze these things and try to quantify them as much as possible, but the reality is they can't be.

The team that wins the turnover battle typically wins the game...thats it. Not much too it. Perhaps some put too much stock into, sure. But there is no reason to further investigate this concept.

I can tell you how much an arm is worth based on actuarial tables. (Depending on age, $106,000 seems to be a good number).

I can certainly tell you the value of a fumble.

Math works.

[Oldfan]

Don't be snotty, just wanted a clarification.

If you don't want snotty, don't twist my words to make me sound stupid, as in...

Ok, so are you saying it is more productive in the long run to just throw it out there and go for the TD and not worry if it gets picked...

This is going to be another one of those threads where you take a stat in isolation and try to prove something earthshaking again, isn't it?

If this statement didn't come from a private part of your anatomy, you should be able to link a thread where I did as you suggest previously. So,give it a try. Support your assertion.

[nightbird]

Imagine a football field. Then, imagine a Bell Curve over it. The value is about six points at the one-yard line and lowers incrementally until its lowest point at the 50. Interceptions have more value than fumbles.

The word "average" doesn't mean that the value is always the same.

I know that, but I just think it's silly to quantify the "average turnover" as being worth 4 points.

Again, not having read the full report, I'm guessing there's a mathematical flaw.

But the idea that a turnover is just the same as a 40-yard pass lends credence to the idea that you should just chuck it out there all day. If your receiver catches it, great. If it's picked off, big deal. They're both "worth 4 points" so you may as well force it deep.