Jump to content
Washington Football Team Logo
Extremeskins
Forum Rules and Guidelines

Busting the Turnover Myth

Oldfan

By Oldfan
in The Stadium

Recommended Posts

Saqs

Saqs

Posted
Posted
Define "huge."

And why are they "huge?"

Huge - unusually great in size or amount or degree or especially extent or scope.

In this context it obviously relates to the impact they have on the outcome of the football game (who wins/loses).

A TO is the one thing that can totally kill a drive for the offense no matter where it occurs on the field, no matter what time of day it is, no matter what time the ref decided to wake up that morning. Major impacts range from time of possession, to field position, to negating a sure scoring drive, to wearing down a defense. There are other trivial impacts such as emotional swing, crowd factor, etc.

TOs are huge in football. No two ways about it IMO.

Link to comment
Share on other sites
Lombardi's_kid_brother

Lombardi's_kid_brother

Posted
Posted
Agreed.

Kinda curius though, given everything all of us know about sports is wrong ( ;) ) do you think the 4-point value has merit? Or more specifically, even if we could come up with an agreed objective average value of a turnover, would it correlate in any meaninful way to how any one given game played out?

I think the average turnover being worth 4 points is almost certainly correct. I've seen it enough places where it makes sense.

I think it probably needs to be broken down further to discuss. "Fumbles," "Fumbles lost," and "Interceptions." And I think that everyone needs to understand that randomness involved in "fumbles lost."

No team is "better" at recovering fumbles than another team. It's completely random. Teams can be better at creating fumbles.

Example. Team A creates 30 fumbles but recovers 9. Team B Creates 15 fumbles but recovers 14. Team A is probably a better defense than Team B. But, I met John Madden gives Team B more credit.

And those numbers you posted make perfect sense. Turnovers probably have the single biggest value of any one play in football. (Ignoring the difference between a three yard run from the 20 and a three-yard run from the, well, 3). And the average NFL game is decided by less than a touchdown. It makes sense that the biggest single difference maker would skew the data that much.

If I told you that team A hit 3 home runs a game while only allowing 1, you would be right in assuming that team A won a lot more games than it lost as Team 1 goes into every game with a 2-run advantage.

The issue with football stats - as I pointed out - is that they can be worth a hell of a lot more in isolation than over time. Example being the Kendall fumble. That was a 10 point swing in that game. Over the season, it is worth 4 points probably.

This does not happen in other sports.

The problem with football is that it is just a 16-game season. It is entirely possible that the Kendall fumble will cost the Skins a playoff berth or playoff seeding or something else important.

In baseball, an error is buried by 162 games worth of chances. In basketball, a turnover is buried by 82 games worth of chances. So, everything in the NFL is 10 times more important than everything in baseballe and five times more important than everything in basketball.

NFL stats are - I think - more valuable over a three year period than in one season - provided that the team remain relatively stable. I'm more interested in the Colts from 2004 to 2006 than the Colts in 2006.

Link to comment
Share on other sites
Lombardi's_kid_brother

Lombardi's_kid_brother

Posted
Posted
Huge - unusually great in size or amount or degree or especially extent or scope.

In this context it obviously relates to the impact they have on the outcome of the football game (who wins/loses).

A TO is the one thing that can totally kill a drive for the offense no matter where it occurs on the field, no matter what time of day it is, no matter what time the ref decided to wake up that morning. Major impacts range from time of possession, to field position, to negating a sure scoring drive, to wearing down a defense. There are other trivial impacts such as emotional swing, crowd factor, etc.

TOs are huge in football. No two ways about it IMO.

Can you quantify any of that to prove it?

Link to comment
Share on other sites
Oldfan

Oldfan

Posted
  • Author
Posted
Kinda curius though, given everything all of us know about sports is wrong ( ;) ) do you think the 4-point value has merit? Or more specifically, even if we could come up with an agreed objective average value of a turnover, would it correlate in any meaninful way to how any one given game played out?

No, the value of an average turnover does not relate to an individual game.

Link to comment
Share on other sites
Lombardi's_kid_brother

Lombardi's_kid_brother

Posted
Posted

Playoff stats are useless in all sports.

There is no chance for a regression to the mean in the playoffs. If Clinton Portis fumbles the ball twice in a game in October, it's a fluke. If he fumbles twice in a playoff game, it is a disaster of Biblical proportions.

Playoff stats are generally not predictive because the sample size is too small.

Link to comment
Share on other sites
Lombardi's_kid_brother

Lombardi's_kid_brother

Posted
Posted
No, the value of an average turnover does not relate to an individual game.

Which is the fundamental problem with advanced football stats, I think. The sample size in the NFL is so small that individual plays take on extraordinary value.

But there is a lot of value in them still.

If the Skins maintained their turnover margin from the previous 4 games, they were going to win 13 games this year. If they have the turnoved margin from Sunday, they are going to win 6 or 7 probably. Both were extreme examples.

Link to comment
Share on other sites
zoony

zoony

Posted
Posted

This is why nerds shouldn't watch football :silly:.

Comparing QB kneeldowns and 100 yard rushers, two things that often happen as a result of having a lead...

-vs-

Turnovers, something that in no way results from having a lead, or anything else really, is a bit silly to me. It's comparing a cause (turnovers) to an effect (qb kneeldowns and repeated running plays to run out the clock).

Or maybe I'm misunderstanding

How do your stats account for that emotional swing? That little voice in the back of the players heads that now says "hey, we can do this" or "oh no...here we go again".

This is a very good point too of course.

Link to comment
Share on other sites
Oldfan

Oldfan

Posted
  • Author
Posted
I think OldFan chose a poor title. It should be "Turnover Myth Explored." Because there are a lot of myths surrounding it.

I stated the one I was debunking:

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

Link to comment
Share on other sites
Gibbs Hog Heaven

Gibbs Hog Heaven

Posted
Posted

Very interesting read Oldfan, though ultimately, I'd don't share the sentiments. Statistics can be manipulated to suit most ANY argument.

Maybe's I'm looking at this too latterly, but for me, turnovers are HUGE in the outcome of football games. As we've seen yet again this season, in the modern NFL game, were teams are so evenly matched right across the board, the disparity between takeaways and giveaways are CRUCIAL. You takeaway the ball, your not only limiting an opponents scoring chances, but increasing your own. And vice versa when you turn it over yourself's. In tight games, as many now are, this can make ALL the difference.

Since 2002, when the NFL expanded to 32 teams and adopted its present eight-division format, two-thirds of the teams that have finished in the top 12 in turnover margin have made the playoffs. At the same time, only 16 of the 72 total teams who made the playoffs during that six-year stretch posted a negative turnover differential.

The most notable exception being the Giants last year, who finished a shocking minus-9 during the regular season yet still made the playoffs as a wild card. But not surprisingly, the Giants topped ALL playoff teams with a plus-5 turnover margin during that amazing four-game run to win it all.

Take that one step further, and the Super Bowl.

In the 42 Super Bowl's to date, teams winning the turnover battle have a 39- 3 record of success. The last one to lose the turnover battle in the show being the Steelers in '06. And you have to go back a further 26 years to 1980, and, incidentally, the Steelers again, to find the next team to lose the turnover battle in the big game and win the Lombardi.

Protect the football, and force the opposition to turn it over, you invariably win football games. Statistics tell us this, as if we didn't know. No amount of any "other" statistics would alter my view if I were a Coach. My :2cents:. If i miss understood the context of your post, i apologise.

Hail.

Link to comment
Share on other sites
Om

Om

Posted
Posted
I think OldFan chose a poor title. It should be "Turnover Myth Explored." Because there are a lot of myths surrounding it.

Agreed on the first sentence.

On the second, do you disagree with the NFL "myth" that suggests turnover differential is, generally, the single most impactful statistic in a given game? Because unless I misread it, that's the thesis of the discussion.

Link to comment
Share on other sites
Om

Om

Posted
Posted

http://www.sbgglobal.com/nfl-football/the-nfl-betting-turnover-factor

NFL betting sometimes involves looking at a lot of statistics. One statistic though is critically important and that is turnovers. The team with the fewest turnovers wins in the NFL about 78% of the time. They cover the Nfl betting spread about 75% of the time. All the other statistics in the world don’t measure up to this one factor. Turnovers are everything in football betting.
Link to comment
Share on other sites
OWUeagleMD

OWUeagleMD

Posted
Posted
No' date=' it's a bad point unless you think that 30 year old men capable of surviving the brutal world of the NFL are also emotional cripples.[/quote']

I tend to side against any appeal to emotion in a logical discussion, but I think you're selling the mental aspect of momentum a wee-bit short.

You don't have to be an emotional cripple to allow your mind to dwell on the negative at a time of lost momentum. I'd say it's nothing short of human to feel impacted by the trends that develop from your actions.

Link to comment
Share on other sites
Lombardi's_kid_brother

Lombardi's_kid_brother

Posted
Posted
Agreed on the first sentence.

On the second, do you disagree with the NFL "myth" that suggests turnover differential is, generally, the single most impactful statistic in a given game? Because unless I misread it, that's the thesis of the discussion.

I think it is impactful but for misunderstood reasons. And it is not "THE MOST IMPACTFUL PLAY."

I think the most impactful play in football is actually a kick return for a TD. It is always worth 7 points. And it is always a lightning strike. I've seen numbers where a punt return TD leads to a win something like 80 percent of the time.

And - as pointed out - a turnover is equally impactful as a 40-yard catch.

However, there are - generally speaking - a lot more turnovers than 40 yard catches.

So, a turnover is the most impactful play that actually happens on a regular basis. But - in terms of fumbles - it is almost completely random.

In conclusion, turnovers are important simply by volume. And a team does not control the turnover differential as much as you think; it is largely luck.

And things that are luck need to be discussed as luck.

The Skins didn't lose because Kendall fumble the ball. The Skins lost because of an incredibly fluky, unlucky play.

My guess is that Gibbs or Schottenheimer may have actually run a drill with the linemen this week to knock down deflected passes. My other guess is that Zorn will ignore it.

Link to comment
Share on other sites
Oldfan

Oldfan

Posted
  • Author
Posted
This is why nerds shouldn't watch football :silly:.

Comparing QB kneeldowns and 100 yard rushers, two things that often happen as a result of having a lead...

-vs-

Turnovers, something that in no way results from having a lead, or anything else really, is a bit silly to me. It's comparing a cause (turnovers) to an effect (qb kneeldowns and repeated running plays to run out the clock).

You missed the points, zoony. The kneeldown theory was Mike Tanier's way of illustrating the principle that correlation does not imply causation. In other words, the people offering the turnover stat as the most telling stat in football need to offer more than a strong correlation to make their case.

The 100 yard rushing stat was an analogy on a different point.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...