Network News

X My Profile
View More Activity
On Twitter: RedskinsInsider and PostSports  |  Facebook  |  E-mail alerts: Redskins and Sports  |  RSS

Statistical analysis: Lucky turnovers, misleading tackle stats and other notes

Statistical analysis

The Colts took charge of Sunday night's game early in the first quarter.
There were a lot of ups and downs Sunday night, but if you wanted to point to one sequence of plays that changed the entire complexion of the game, it was the Donovan McNabb interception immediately followed by a 57-yard touchdown strike by Peyton Manning in the first quarter. The Redskins went from a solid drive into Indianapolis territory and a 60 percent chance of winning to being down by 7 points with a 29 percent chance of winning. The Redskins could never catch up. No other play, or sequence of plays, had nearly the impact that those two did.

The Redskins have been lucky with turnovers so far.
Turnovers were critical Sunday night, but in reality they are very random, meaning that generating and preventing turnovers is far less of a skill and far more driven by chance than most people believe. Teams with good turnover differentials halfway through the season are not much more likely to continue good fortune than teams with poor differentials. The good news is that the Redskins have a +5 differential so far this year, tied for third in the NFC. The bad news is that they can't count on it to continue.

Some stats mean different things for different positions.
LaRon Landry leads the league with 63 tackles, but despite his solid play, this is one stat a safety does not want. The more tackles a safety is making, the further down the field the opponent is getting.

Fletcher and McIntosh are leading the defense.
Individual tackles can be a very misleading statistic. One reason is that bad defenses tend to give up a lot of first downs and long drives, giving below-average defenders lots opportunities to rack up tackles. Plus, some positions, such as linebacker, are expected to have more tackles than others. To make better sense of tackles, I created a stat called Tackle Factor (TF). TF looks at the percentage of a team's tackles that each player makes, and then compares that to what proportion of tackles we should expect for a player at his position. For example, linebackers should typically account for 11 percent of a team's tackles and assists. London Fletcher has tallied 37 tackles and 23 assists this season, accounting for 18 percent of the Redskins' tackles and assists, making his TF 1.5. Put simply, Fletcher has about 60 percent more tackles than you'd expect for his position. Right behind Fletcher is Rocky McIntosh with a TF of 1.3. Landry actually leads the squad with a 1.8 TF, but that's mostly a result of failures elsewhere in the defense.

Speaking of failures, the Redskins' defensive line has been nearly silent in 2010.
As a unit, it has accounted for 13 percent of the team's tackles, second worst for 3-4 defenses and for the league as a whole. (The NFL average is 22 percent.) The line has notched only 6 tackles for a loss all season, compared to a league average of 11. The line is credited with 3 sacks and 4 QB hits, well below the league averages of 7 sacks and 12 hits. Obviously the transition to the 3-4 has had an effect, but regardless of scheme, the Redskins' defensive line is not getting the job done, neither stopping the run nor pressuring quarterbacks. The Redskins have an urgent need for a defensive lineman who can single-handedly impact a game. Unfortunately, there's no one on the roster who might be able...oh, wait, there is...at least for now.

Chicago is an enigmatic team.
They started off hot, mostly thanks to Jay Cutler's blazing pass efficiency, which was best in the league until the catastrophic meltdown against the Giants in Week 4. Cutler has been sacked 23 times this season, 5 more times than the next quarterback, which is amazing in itself. But he's played a game and half less than the other leading sack victims. A better way to look his sacks might be sack rate, the percentage of dropbacks that result in sacks. He's been dropped on a whopping 14 percent of all dropbacks. (The average is 6 percent.) Even without the 9 sacks suffered in the Giants game, Cutler would still sport a 12 percent sack rate, worse than any other current starter in the league. Making matters worse for Chicago, they have struggled to run effectively all season, ranking 24th in run success rate. The defense has been keeping Chicago competitive lately. It's among the better squads in the league, allowing only 5.5 net yards per pass attempt and 3.5 yards per carry, ranking 7th and 3rd in the league respectively.


Brian Burke is a former Navy pilot who has given up his F/A-18 for the less dangerous hobby of football analysis. He is the creator of Advanced NFL Stats, a Web site about football, statistics and game theory.

By Brian Burke  | October 20, 2010; 6:04 AM ET
Categories:  Statistical analysis  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Albert Haynesworth still with Redskins as trade deadline passes
Next: McNabb's influence, leadership are worth long-term investment

Comments

First off nerd boy the turnovers they are getting are forced by great hits not luck.

The majority of Landry's tackles are right at or near the line of scrimmage.

This blog has totaly sucked ever since JLC left, and it wasn't that great when he was here.

It's a good thing people like coming up here to b_tch and fight with one another otherwise this place would be dead.

You get no real news from this blog not one piece of inside information.

Posted by: Flounder21 | October 20, 2010 6:42 AM | Report abuse

Talke about a bunch of hypocrites the NFL brass is a joke player saftey my a@@.

Ask the NFL if they are so worried about player saftey why are the going to an 18 game schedule, it's called more money.

League is selling photo of James Harrison's illegal hit
Posted by Mike Florio on October 19, 2010 11:12 PM ET
The folks at KDKA-TV in Pittsburgh have uncovered an interesting angle regarding the hit that resulted in the imposition of a $75,000 fine against Steelers linebacker James Harrison.

The league currently is selling photos of the incident involving Harrison and Browns receiver Mohamed Massaquoi, at anywhere from $15.95 to $249.95.

Sorry, but the NFL can't have it both ways. If the league wants to legislate dirty hits out of the game, the NFL should do nothing to profit from those dirty hits.

For now, the NFL is.


Posted by: Flounder21 | October 20, 2010 6:46 AM | Report abuse


You get no real news from this blog not one piece of inside information.

Posted by: Flounder21 | October 20, 2010 6:42 AM |

Redskins Outsider, where you can get the latest AH thread that make some Redskin fans lose sight of reality.

of all the mistakes snyder has made in the 10 years he's owned the team, signing AH as a FA was not only his biggest mistake, it will set this team back financially and test the players and coaching staffs wills to no end as long as he is with the team.I'm adding AH to my list of ITA's as of today.

Posted by: hessone | October 20, 2010 6:55 AM | Report abuse

I'm with you Flound...this blog is a joke. No matter how much we complain about these stupid, offseason-oriented stat posts, they continue to throw up this bullsh*t. "Turnovers are lucky". Wow...next post: "The Sky is Blue"...

Posted by: brownwood26 | October 20, 2010 7:01 AM | Report abuse

And you're right about the TOs Flound...every single one of the Skins' takeaways in that game were forced. No fumbled snaps, no tipped passes, nothing quirky. Just putting a hat on the ball and stripping the ball and stuff aimed at taking the ball away. Chalking that up as "luck" is misguided at best. By that logic, it was luck that Keiland Williams caught that touchdown and that Torain trucked a guy on his way into the end zone...

Posted by: brownwood26 | October 20, 2010 7:05 AM | Report abuse

You get no real news from this blog not one piece of inside information.


News?

Who comes here for news?

I read this thing for jokes, puns, snarkiness, anxiety, fear, racist/homophobic commentary, crowing, braggadocio, and cluelessness.

And oh, there's the fellowship of other suffering redskins' fans.

Beyond that, I don't need no stinkin' news.

Because in this place, being ignorant is bliss.

Posted by: MistaMoe | October 20, 2010 7:35 AM | Report abuse

The good news is that the Redskins have a +5 differential so far this year, tied for third in the NFC. The bad news is that they can't count on it to continue.


Uh,why not?

Posted by: smittdiddy | October 20, 2010 7:38 AM | Report abuse

And why hasn't JReid or any other writer on this blog posted one thing about easily one of the most anticipated match-ups of this season for us:Our kick-off/punt coverage units vs. Devin Hester.

Posted by: smittdiddy | October 20, 2010 7:45 AM | Report abuse

"The good news is that the Redskins have a +5 differential so far this year, tied for third in the NFC. The bad news is that they can't count on it to continue. "


The 2009 New Orleans Saints send their regards...oh, and the 2000 Ravens also say 'hi'...

Posted by: brownwood26 | October 20, 2010 7:48 AM | Report abuse

The good news is that the Redskins have a +5 differential so far this year, tied for third in the NFC. The bad news is that they can't count on it to continue.


Uh,why not?

Posted by: smittdiddy | October 20, 2010 7:38 AM

I was wondering that too, but I figured it out myself. There's a slight chance (.00037%) that the team bus will be struck by a meteorite, killing the entire squad and coaches and canceling the rest of the season. If that happens, you don't even need to be a Navy pilot to realize that "they can't count on the streak to continue."

Man, this guy really has a bad way of expressing himself. If he actually is a statistician he would give the probabilities of it continuing. This after his amazing (pulled-it-out-my-rear) statement that we went from a 60% chance of winning to 29% in two plays.

Still prefer this guy to LaVar Arrington, Mike Wise, or Tracie Hogilton.

Posted by: beep-beep | October 20, 2010 7:49 AM | Report abuse

And why hasn't JReid or any other writer on this blog posted one thing about easily one of the most anticipated match-ups of this season for us:Our kick-off/punt coverage units vs. Devin Hester.

Posted by: smittdiddy | October 20, 2010 7:45 AM

You can't be serious with that question. He's writing a piece on Albert Haynesworth vs. Deven Hester.

Posted by: beep-beep | October 20, 2010 7:53 AM | Report abuse


Because in this place, being ignorant is bliss.

Posted by: MistaMoe | October 20, 2010 7:35 AM |

duh, what's that mean mrmoe ?

Posted by: hessone | October 20, 2010 8:03 AM | Report abuse

hess--Do you think that if the Redskins had traded Haynesworth to Houston for a ham sandwich that the Washington Post would have traded J. Reid to the Houston Chronicle for a jar of mustard?

Posted by: beep-beep | October 20, 2010 8:12 AM | Report abuse

And why hasn't JReid or any other writer on this blog posted one thing about easily one of the most anticipated match-ups of this season for us:Our kick-off/punt coverage units vs. Devin Hester.


I'm wondering the same thing, especially as I'm now sitting him Hester's home town, Riviera Beach, Florida.

(I've known Hester since he was in the 9th grade.)

Hester is the kind of player Brandon Banks might become if the skins can find clever ways to make his obvious speed and shirtiness apart of the offense.

I'd like to see Banks catch a few screens, run a reverse, or work in the slot out of a bunch formation.

Finally, I'd also like the skins to continue searching for players like Banks/Hester in the upcoming off season.

We've been a slow team for such a long time.

Posted by: MistaMoe | October 20, 2010 8:13 AM | Report abuse

McNabb contract beepage

Posted by: monk811 | October 20, 2010 8:16 AM | Report abuse

I thought we agreed that this NERD wouldn't post this crap so early in the morning!

Anyways...there's not suck thing as Luck, Tackle Factor(TF) or the Easter Bunny. Sorry.

Posted by: PlayAction | October 20, 2010 8:23 AM | Report abuse


hess--Do you think that if the Redskins had traded Haynesworth to Houston for a ham sandwich that the Washington Post would have traded J. Reid to the Houston Chronicle for a jar of mustard?

Posted by: beep-beep | October 20, 2010 8:12 AM |

beep, in comparison, the redskins were looking for a second round draft pick for AH and the WaPo was looking for the ham sandwich in the JReid deal. the deal fell through when the WaPo also demanded a bag of chips and a soda in the trade. sources close to the chronicle said JReid ate the food and drank the soda before the deal hit the table, or should I say before the food hit the table. when asked to comment on the deal JReid said, "you'll have to ask sources close to the deal".

Posted by: hessone | October 20, 2010 8:37 AM | Report abuse

i got a couple of sentences in and just stopped with this post. This stat guy is not a welcome addition. Drek.

Posted by: buke | October 20, 2010 9:42 AM | Report abuse

Good grief, can you knuckleheads give the dude a break? Thoughtfully consider what he is saying without using the word "nerd" in your response.

Sheesh.

Anyway, I think Brian makes a good point about the d-line---it STINKS right now.

Posted by: watchingfromasia | October 20, 2010 10:01 AM | Report abuse

I still don't know why the turnovers were "lucky." If anything, Colts were lucky we didn't have 3 picks and 3 fumbles. Two of those fumbles were caused by huge hits and the other just a great heads up play by Orakpo.

We've always had trouble forcing turnovers and even with Los dropping everything they're finally doing so, and we say they are "lucky...." I don't see any reason we can't continue to generate turnovers.

Posted by: rachel216 | October 20, 2010 10:34 AM | Report abuse

So LBs "should account for about 11 percent of tackles?" And the Dline league average is about "22 percent?" So, the secondary should account for, what, 67 percent??? Forgive me for the simplicity, but 3 linemen divided by 11 players equals 27 percent and 4 LBs equals 36 percent. And truly, parse Landry's tackles and its got to be close to 70 percent are within 3 yds of the line of scrimmage. Now I'd love to see a ratio of times that Doughty got blown up or blown coverage vs number of plays he DID make in that game! Jeez, just play an extra LB instead of him!

Posted by: campspringskid | October 20, 2010 12:36 PM | Report abuse

Hey Brian,

Two things.

First, I thought there was a statistical analysis not too long ago concluding that interceptions are related to quality of defense whereas fumbles are random. In other words, we can be more specific than your statement that turnovers are relatively random occurrences.

Second, you note that Landry's high TF is mostly a result of a poor defense. This can be tested. In general, is TF for strong safeties inversely related to their team defense's total yards allowed? If so, you could produce a weighting that would predict the TF of strong safeties based on the quality of the defense they play for. You could also calculate a stat that indexes average yard from scrimmage that a tackle is made. In other words, are Landry's tackles closer to the line of scrimmage than tackles for a typical strong safety, which would indicate that Landry indeed deserves credit? He's not just waiting for the lineman and linebackers to allow a ball-carrier to reach him.

Posted by: SkinsFaninCanada | October 20, 2010 1:27 PM | Report abuse

On both sides of the ball, there is a factor of luck. It's just not quantifyable. What creates turnovers? Bad judgement, confusion, stupidity, hard hits, poor execution, good play calls, and pressure among others. There are a lot of factors that are involved. While some of the turnover that the Skins have gotten have been timely, others have had no effect. The Skins got 2 TO's in the Packers game and converted 1 into a FG. The thing about turnovers is that they have to be factored in where on the field they occur. In the Colts game, that turnover was a minimum 10 point swing, I question the percentage of winning there because the play occured in the 1st quarter. Turnovers also have greater value as the game goes on. When there are only X amount of possessions in a game, a tunover late can have a more harsh affect. Take a look at the two turnovers in the Green Bay game. The first on the 2nd play of the game, the last on the last offensive play for the Packers. Suffice it to say, turnovers have a lot of factors built in which is why you can't really count on them. As they say with Investments, past performance is not necessarily indicative of how an investment performs.

Posted by: JWPinCHI | October 20, 2010 1:40 PM | Report abuse

I like your analysis from statistical view, enjoy reading it.

Posted by: Parinya | October 20, 2010 5:10 PM | Report abuse

Post a Comment

We encourage users to analyze, comment on and even challenge washingtonpost.com's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.




characters remaining

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company