The Dude Abides...Turnover Margin and Winning Percentage (Or How I Learned To Stop Worrying and Love Mike Nolan)
by TJ Johnson on Oct 30, 2009 7:43 PM MDT
Email
PrintAl Davis once said, "Don't worry about mistakes. Just win."
The problem with this statement is, that in the NFL, when you make a mistake, you rarely win.
In fact, you lose. A lot.
And the more mistakes you make, the more likely it is that you are going to receive a first-class, gold-plated, butt whoopin' of the sort that is reserved for pickpockets and teams quarterbacked by JaMarcus Russell.
In football, the biggest mistakes are not holding calls or missed assignments, although they loom large indeed. No, the biggest mistakes are turnovers. Give the ball up, you lose. Get your opponent to give the ball up, you win. In the age of QB ratings and BCS rankings, it's still amazing how turnovers are simply the most important single statistic that determines who stays and who goes.
Last week, I posted about how winning percentages correlate with the raw number of turnovers a team has in a game. That post is here. It essentially makes the case for 2 turnovers being the magic number for winning and losing. Give the ball up twice, your winning percentage drops below 50%. And in my weekly post on the Stats That Don't Lie, I regularly make the outrageous claim that if you win the turnover battle, you will win 80% of your games.
Several MHR members have wondered about Turnover Margin. In short, what is a team's winning percentage if they win/lose the turnover margin by a certain number? Is your winning percentage that much higher if you are even in the turnover battle than if you lose the turnover battle by one turnover? What if you win the turnover battle by two? Or three? Or more?
I think we can help answer these questions. For background, I once again looked at every game from the 2008 and 2009 seasons (right up until last week). This essentially represents a sample of 358 games. Not a bad sample size. When I have more time, I plan on getting 10 seasons worth of data. But unless you are sitting in the unemployment line known as the Black Hole, you simply don't have the time during the season to grab this amount of data, so the 358 games will have to suffice.
In summary, these are the results:
This chart illustrates the winning percentages of teams for each of the following turnover margins during this period:
- -4 or less turnover margin: 0%
- -3 turnover margin: 17.44%
- -2 turnover margin: 21.95%
- -1 turnover margin: 28.36%
- Even Turnover margin: 50% (of course)
- +1 turnover margin: 71.64%
- +2 turnover margin: 78.05%
- +3 turnover margin: 82.56%
- +4 or more turnover margin: 100%
A few things stand out when looking at this data. First, the teams that won the turnover margin by 4 or more had a great day. On the flip side, those that lost by 4 or more experienced a nightmare (see Bears, Chicago, 2009). It did not matter if the turnover margin was greater than 4, the results were the same. Second, winning the turnover margin by even 1 was extremely profitable, driving winning percentage to over 70%. Third, the winning percentage of teams with turnover margins of +1, +2, and +3 were not that different, ranging from 72% to 82%. Certainly while it's better to win the turnover margin by 3 rather than 1, it's not as big a difference maker as one would expect. Still, I'd rather win 8 out of 10 than 7 out of 10.
The last point deserves more attention, to be sure. Essentially, what matters more than the size of turnover margin is just making sure to win the battle in the first place. As we can see from the data, a tie in turnover margin was the equivalent of flipping a coin. But winning the battle improved winning percentage significantly.
Perhaps this explains why we've seen so many Super Bowl winners with attacking and ferocious defenses and offenses that focused on efficiency, field position, and taking care of the ball. +1 works. The formula works.
Which brings me to my main point: Mike Nolan is a genius.
He's a genius not because he adjusts well at halftime. He's not a genius because he knew Ron Fields was the 2nd coming of the Rock of Gibraltar. He's a genius because he let's loose the hounds. He attacks and attacks and doesn't stop until the final gun sounds.
The San Diego game was a perfect example. Despite the fact that Chargers played very well in the 1st Quarter, Nolan must have blitzed nearly 10 times. He blitzed from the corner, from the slot, up the middle, and off the edge. All in the 1st Quarter. And he didn't stop there. In the 2nd half, he blitzed some more. He blitzed from his base formation and he blitzed from both his "big" and small "nickel" packages. He blitzed from London. He blitzed from France. He even blitzed between Phillip's whining rants.
Why? Why does Mike Nolan blitz as often as drunk virgin Raider fans? He wants to win. While I'm sure he's never looked at stats like the ones above, he doesn't need to. His experience tells him that winning the turnover battle--even by one turnover--means the difference between winning and losing.
So he's not going out like a punk. If he's going out, he's going out like a rabid dog. He's going out trying to do everything in his power to get one...more...turnover.
Certainly, you remember the Broncos of the last three seasons? Remember the lack of blitzing? Remember rushing 3 and dropping 8? Remember being picked apart game after game after game? Remember the lack of turnovers? Remember the .500 record?
Of course you do.
Well those days are over, my friends. The Broncos are not going to simply rely on luck to generate turnovers anymore. Luck gets you nothing more than a coin flip's chance. No, Mike Nolan isn't just trying to stop you on 3rd down.
He wants the ball.
Go Broncos!!
This is a Fan-Created Comment on MileHighReport.com. The opinion here is not necessarily shared by the editorial staff of MHR
58 comments
|
23 recs |
Do you like this story?
Comments
Nice post dude...simple and spot on!
Those that cant coach, compete!
Failing to plan is planning to fail.
All I want is 53 Rod Smiths. Is that asking too much????
"Peyton Hillis didn’t rip the sleeves off his jersey, they flew off out of fear."
Calijoefornia.
This is a very complicated case
And, uh, lotta strands to keep in my head, man. Lotta strands in old Duder’s head. Luckily I’m adhering to a pretty strict, uh, drug regimen to keep my mind limber.
Rec’d
Hey SWG
You know it, man!
Thanks my friend.
Yeah, well, you know, that's just, like, your opinion, man.
Rec'd.
Love the title!
"All by their heads, he places crowns."
by Tempestuous Binary on Oct 30, 2009 7:58 PM MDT reply actions
I am glad that movie makes someone else laught too!
Yeah, well, you know, that's just, like, your opinion, man.
Dude
This isn’t just a movie to laugh at, it’s simply the best movie ever made, bar none. I watch it four times a year just to reconfirm my belief. and, btw, Lebowskibronco, you are far too industrious to qualify as the dude, known as the laziest man in all of Los Angeles.
If people aren't supposed to eat animals, why are they made out of meat??
SK was a genius and it didn't take 2001 to confirm this. I wish more people knew about the movie honsetly....
…fluoride, baby, fluoride…
Yeah, well, you know, that's just, like, your opinion, man.
I like the data, dude...
… but why don’t the percentages in the -3/+3 column sum to 100?
All of the others sum to 100 (-4/4, -2/2, -1/+1. 0/0)
Great post. I just joined the MHR community. Love this site.
by Irrationalnumb3r on Oct 30, 2009 8:42 PM MDT reply actions
That would make the most sense, donbok...
there is a remaining 4.54% to be accounted for and if the data are just from 2008 and 2009, that 4.54% could easily represent one fluky game that ended in a tie despite the three turn-over differential.
by Irrationalnumb3r on Oct 30, 2009 8:54 PM MDT up reply actions
heading out, will be back later...but yes, it's the tie game, great eye for detail
I forgot to mention this in the post. Irational, thanks for bringing that up..
Yeah, well, you know, that's just, like, your opinion, man.
The percentages still should have added up to 100%
if for the tie you gave each team credit for half a win rather than no win, which it appears you did. Evidently there were 22 games in the +3, -3 category. But it looks like you divided 3.5 and 17.5, respectively, by 22 to get your percentages, and those numbers add up only to 21. I suspect it should have been 18 wins and a tie and thus 84.09% for the +3 teams, although it might have been 4 wins and a tie for the -3 teams.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
Hey Spock,
I’ll take a look now. I didn’t give any team a win actually, but I think I included them in the overall numerator, and as such, there would have been an additional numerator for -3 and a numerator for +3. Let me go back and see if I simply forgot to include of on them.
It won’t change the basic points, but I want to get it right. I will be back with this later tonight.
Yeah, well, you know, that's just, like, your opinion, man.
Spock
It’s what I thought. I did not give them .5 for the wins. Thanks for pointing that out. I would have searched in vain for what was going wrong. I fixed the chart, it changes the numbers slightly. Thanks, my friend!
Yeah, well, you know, that's just, like, your opinion, man.
I'm a bit baffled now
Your original figures suggested there were 22 games in which the turnover differential was 3. Your new figures suggest it’s 43, with 35.5 won by the plus teams and 7.5 by the minus teams. But I can’t get your original percentages dividing by 43. However, I can get them if I divide 35.0 and 7.0 by 44, in which case the tie game, rather than being split 50-50 between the two teams, was inadvertently added to the divisor instead. Notice that 44 is exactly twice 22, which is why I thought the latter was the number of games with a turnover differential of 3.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
My man Spock...I just finished quoting you in my Spotlight post, so take a look
Here are my numbers (through 2008 and 2009, 7 weeks)
-4, +4: 29 (0,29)
-3, +3: 43 (7.5, 35.5)
-2, +2: 82 (18, 64)
-1, +1: 134 (38,96)
0,0: 142 (71,71)
I think this clears it up. At least it did when I went back and redid the spread.
Yeah, well, you know, that's just, like, your opinion, man.
Works for me.
It’s 43 like I thought it was. Thanks for the figures.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
Nice
I particularly like:
He’s a genius because he let’s loose the hounds. He attacks and attacks and doesn’t stop until the final gun sounds.
This has been missing from Denver for as long as I can remember.
MARK IT ZERO!!!!
Dude, this is a league game, this determines who enters the next round robin. Am I wrong? Am I wrong?
You mark that frame an 8, and you're entering a world of pain.
by waltersobchakbronco on Oct 30, 2009 8:43 PM MDT reply actions
Agreed
I hate it when any team goes into “prevent” mode. They are just playing not to lose rather than to win! I love that Nolan/McD have not once gone into prevent mode.
Actually, the four man line we tend to use in passing situations
is a very mild form of prevent defense, in that we sacrifice some uncertainty (on the other team’s part) in exchange for having all
non-rushers in their pass drops from the beginning of the play. In our base defense the three down linemen plus one other player normally rush, which means one (occasionally both) of the OLBs is dropping into coverage from the line of scrimmage, which means he doesn’t get into his zone quite as rapidly as he would if he were playing several yards off the line of scrimmage. But we still have a four-man (at least) rush, and the particular people we put on the line in this package are probably as a group slightly better pass rushers than the four who end up rushing from the base package. So your point still holds. It’s an aggressive, attacking defense with just a little extra margin of safety.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
Love the post Dude!
Excellent analysis of the importance of causing turnovers. Nolan indeed takes the approach of attack, dictate the action from our side of the ball. I still think the data on turnovers regardless of the margin is equally compelling. 2 turnovers means you greatly reduce your chance of winning. That’s good enough for me. Rec’d
Imagination is more important than knowledge. A. Einstein
Iowa State versus Nebraska
I know it isn’t the NFL, but the Cyclones really put the turnover theory to the test last Saturday. They won an historic victory in Lincoln, but were still barely able to do so despite being +8 in TO differential! Nebraska had a very real chance of still winning that game with a -7 or -8 differential.
by Irrationalnumb3r on Oct 30, 2009 9:00 PM MDT reply actions
Irrational
I have never looked at the college game. Would be an interesting thing to take a look at.
Yeah, well, you know, that's just, like, your opinion, man.
Great stuff as usual Dude
You can do the other 10 years on your smoke break. Lol.
Character may be manifested in the great moments but it is made in the small ones -- Philip Brooks
If you remember San Diego’s opening drive. Rivers was torching us behind our blitzing, they marched down the field. Most defensive coordinators would have abandoned the blitzkrieg… Not Nolan.
He simply changed up the look, delayed a man, ran some zone blitzes. But the pressure remained. By the fourth quarter, the pocket was collapsing on almost every play. Granted, Rivers threw some hot routes for big gains, But as DUDE points out in this post, Denver created the turnovers that won them the game.
Hey Rain,
About to do a Spotlight post on the 2nd half adjustments. You could have written it, because that is exactly what I noticed!
Yeah, well, you know, that's just, like, your opinion, man.
Oooooooooooooooohhh! Blending Coens with Kubrick!
Nice touch!
And great read as usual. Thanks Dude.
Conversation nonstarters: hoping McDaniels fails, comparing Bears to Broncos, Cutler to Orton, apples to oranges, and casual drinkers to Raiders fans.
MT..just be happy it wasn't Eyes Wide Shut!!! haha
Yeah, well, you know, that's just, like, your opinion, man.
Hey man, I appreciate your comment more than you know!
Yeah, well, you know, that's just, like, your opinion, man.
Fantastic!
Dude,
I enjoyed your previous post regarding the correlation of giving away the ball to winning percentage, but this is the one I really wanted to see because taking away the ball is equally important, I felt.
Obviously correlation does not imply causation, so we can’t necessarily say “because so-and-so lost the turnover margin by -1, they lost the game”, but it does at the minimum show how important this margin is. I think it’s safe to say, if you don’t give the ball up and you take it away at least once, you put yourself in a better position to win.
I too was surprised that the relative difference between +1 through +3 and -1 through -3 do not correlate as much with a greater or lesser winning percentage, respectively, as one might expect. That is fascinating. Just goes to show you that coughing the ball up in the red zone like our ex-friend in Chicago and coughing the ball up on your own goal line like our best friend in Oakland is not something you can easily say, “it’s ok we’ll get them next time”. It’s not easy forcing a turnover, and my guess is that would be the simplest way to “get back int the game” at that point, as opposed to convincing yourself that you’ll be efficient scoring on the next possessions.
I did have one question. 0 implies 50%, naturally, and shouldn’t +1/-1, +2/-2, and +3/-3 each sum to 100% as well since if one team has a +3, the other has a -3? The +3/-3 does not add to 100%, but the others do.
Also, if you are taking requests for your next analysis, I would love to see the correlation of number of defensive touchdowns with winning percentage. I think that is the next step for our defense to become truly dominant. DJ almost had one with the Renaldo sack-strip-fumble of Romo. I have to imagine that if our defense started scoring, with our offense, we’d never lose.
Keep up the great work, Dude!
Noleaf
I am going back through that right now. It’s the tie game I believe.
Yeah, well, you know, that's just, like, your opinion, man.
Noleaf
Also, I’ll do that research for you, no problem.
Yeah, well, you know, that's just, like, your opinion, man.
Does anyone recall how poor we were in the preseason generating turnovers?
I was a little bit concerned then, but not anymore. Nolan has got these guys playing very aggressively and I like it.
Imagination is more important than knowledge. A. Einstein
Great post, Dude!
I do appreciate the hard work you put in on this stuff. That is a vast amount of data to crunch!
A thought regarding Nolan:
I recall getting a bit nervous when Rivers and the ‘suspect’ SD OL seemed to have no trouble at all picking up our blitzes early in the Monday night game. That is, until I realized that Nolan was, in effect, probing the SD pass protection in an attempt to find areas of weakness.
It didn’t reap much in the way of dividends on that early drive, but I’d be willing to bet that it was a big part of the payoff later in the game when it really mattered. Look no further than the abundance of middle pressure in the second half. He found the weakness and exploited it.
Another thing that struck me was the fact that Nolan seems to rarely come out of the gates in his base defense. He seems to adhere to the philosophy that by showing the opposing QB and OL a varied pressure package immediately, he plants the idea in their heads that the pressure could be coming from anywhere at any time. More info to process for the QB and center is a BIG advantage to the defense.
Again, quality stuff, bro! Keep it coming!
- Jason
Horton is WIN - HORVIL TIKI
by jubei on Oct 31, 2009 8:59 AM MDT reply actions 1 recs
Nice insight and rec'd.
He does seem to be master at finding the weakness then attacking it over and over. Must be discouraging to face that.
Imagination is more important than knowledge. A. Einstein
Thanks, Ponderosa!
All I know is that I would not want to be an opposing QB!
One of the many things that strikes me about this defense is that they are SO good at both showing blitz and backing out, and masking the blitz when it’s coming. Lebowski mentioned the first defensive snap of the game for the Broncos against SD, and the defense coming out is some wacky formation or another. I still need to go back and check it out, but it emphasizes my point. You just don’t know where the pressure is coming from, or sometimes more importantly, WHEN it’s coming. And Nolan does a great job of not establishing his tendencies.
Thank you Nolan and McD for giving us back our defense!
- Jason
Horton is WIN - HORVIL TIKI
I noticed on the Ravens' board
that some of their fans are really nervous about our pass rush. I think that’s the aspect of our defense that scares teams and their fans the most. And as you so insightfully noted, Nolan did a nice job of probing their blocking scheme until he found the A-gap weakness. By bringing both ILBs up into the two A gaps, he forced the center to decide who to block, and whichever one he chose the other went free. (And the one he slid over to block probably dropped back into coverage.) That prevented Rivers from stepping up in the pocket, thus exposing him to the outside rush. Nolan figured Rivers didn’t have enough mobility to escape to the outside.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
Thanks, Jubei, I always appreciate getting input from you and I am glad we stole you from the DP. You were out of your element!!
This is a much better place for an intellectual like you!
Yeah, well, you know, that's just, like, your opinion, man.
who is second in the NFL in INT again?
Verbose in style, dispersion of thought, procrastination in life.
The guy formerly known as ZAPPA
Great post Lebowski and rec'd
We pretty much knew what we were getting with Mike Nolan. McDaniels is the genius for getting him so quickly before anyone did. My hat’s off to the both of them, very, very smart.
We have heard people say they worry about Coach Nolan going back to HC if the Broncos do a great job throughout the season. I’m one of those guys who believes Coach Nolan learned from his mistakes in San Fran and won’t ever try the reigns of head coach again. Take a look a Dick LeBeau of the Pittsburg Steelers, he tried the HC thing for a couple of years and went back to DC where he can profit from his teaching and coaching. It takes a special person to run the whole show and I believe McDaniels is one of thoose rare breeds. Nolan? Nope, DC, yep.
It should be noted that McDaniels
knew exactly what kind of defense he wanted to run, and he chose Nolan not only because he’s good but also because he was exactly the right man for that kind of defense. He also chose him because he felt they could work together effectively. McDaniels tells Nolan the opposing quarterback’s weaknesses and tendencies, and Nolan schemes ways of taking advantage of them.
"Surprised to see you, Captain, though pleased." — from Star Trek episode Space Seed.
Lebowski, I have a suggestion....
First:
You need to take all these posts, put them in a book, and then (between each Post) add what thoughts are going through that brilliant mind of yours.
Second:
When the book hits the ‘Best-Seller’ list and you’re rolling in money, send a bunch of that money to me for giving you the idea.
:0)
Guardian of the Gate to La La Land!
Gonsoulin, Taylor, Little, Wright, Gradishar, Atwater, Davis, and Sharpe...
Why are they not in the Hall...I just don't understand.
LOL
Character may be manifested in the great moments but it is made in the small ones -- Philip Brooks
Great Post
It answered a lot of my questions from an earlier post of yours.
Mike Shanahans philosophy on defense before he got fired seemed to be that a defense shouldn’t have to blitz. He wanted it to be able to play well and stop the opposing offense with good solid defensive play, no fancy stuff, no taking chances, just cover your assignment and if you do it well the offense won’t get first downs.
I thought that this was a good philosophy and that the defensive coordinaters he was getting to try to accomplish this just weren’t able to do it, either with the game plan, or the personel. It seemed like he wanted it this way because often you can get burned on a blitz by a slant our a well played post. There’s nobody left in the backfield to protect against the big play.
Looking back to our superbowl years with Robinson as our defensive coordinator I’m starting to think that maybe that’s not the best philosophy. Robinsons philosophy was to blitz a lot and take a lot of chances. Sure, we got burned once in a while, but we also got turnovers. At the time our offense was awsome and so they were able to overcome some of the mistakes our defense made. Shanahan got tired of getting burned and so fired Robinson to get a defensive coordinator in there that would be more conservative.
Looking back, I don’t think either one is right. McDaniels and Nolan seem to have found the perfect mix. They blitz a lot, but they keep the right guys back to prevent the 70 yard touchdown. Sure, sometimes they let the opposing offense get a 20 yarder or a 30 yarder out of a blitz, but more often than not they don’t and in the end, it works out. On the offensive side of the ball, the Broncos are more conservative. The play calling is a lot safer. There are fewer turnovers.
Seems like Shanahan had things backwards and McCaniels has the right approach. Offensive turnovers is what kills a team. By limiting that and playing agressive defense without letting up the big play, he’s found the perfect balance of offense and defense complimenting eachother.
GO BRONCOS!!!
G4
I think you should work up a post of your own with an expansion of these thoughts. You could do a feature of all of the defensive coordinators since Robinson and profile it with this angle. If you don’t want to do it, let me know. I like the idea… A LOT! But you seem to have a good handle on it!
Yeah, well, you know, that's just, like, your opinion, man.
I'll try
I’ll start doing some research on the coordinators and stats. What I said was based on gut feeling and observation. I could be totally wrong and find that the Bronco’s blitzed MORE, but somehow I don’t think so. Do you have some good website suggestions on where to get stats that can be easily downloaded into an excel spreadsheet? Compliing the data would be a lot faster that way.
Thanks, and
GO BRONCOS!!
p.s. Tough loss this weekend. Kinda brings you back down to earth getting wolloped like that.
G4
The first place I would start is to go through since 1997 and find out who all the defensive coordinators have been. You probably already know them, but this will help you confirm:
http://www.denverbroncos.com/resources/custom/PDF/RecordBook/2008/AllTimeCoaches.pdf
Now, since you are trying to explore correlation between aggressive blitzing defense and Broncos winning seasons, I would just create the following columns on a spread:
1) Year
2) Wins
3) Losses
4) Def Coordinator
5) Style (here you simply just have to google each coordinator’s name to find their style, for example here is something on Larry Coyers from HT: http://www.stampedeblue.com/2009/1/22/731801/what-tweaks-will-coyer-bri
)
You could say something like 3-4 heavy zone blitz or something short
6) Then you can also have a column for turnover differential
Using the categories, you can quickly surmise or postulate how our coordinators have affected both wins/losses and turnover differential in any given year. This should allow you to make a reasoned argument very easily.
The tough part is that there is no way around just googling the different coordinators and getting their style. You can get the wins/losses/turnovers/etc from nfl.com
Let me know if you need any help, my friend. I am here to help.
Yeah, well, you know, that's just, like, your opinion, man.
Thanks for this, Dude!
I was one of the masses clamoring for this. I must admit the curve is a little different than I expected, but only because the +1 – +3 numbers were so close. I was impressed to see that +1 immediately got you over 70%. Fascinating numbers.
Love it. Can you work up a graph of Broncos win % vs. the number of chili dogs I eat? I’d like a chili dog for lunch tomorrow. :)
BB, No Problem. I was hoping you would be the one MHR member that would like that graph the most
I plan on looking at TO vs down and distance,etc in awhile too, so look out for that.
Yeah, well, you know, that's just, like, your opinion, man.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Getting Social With MHR
Web Stuff
In Association With
Shutdown Corner - NFL - Yahoo! Sports
- Jim Irsay says he would welcome Peyton Manning back … at the right price
- Roddy White isn’t thrilled with Roger Goodell’s $20 million salary
- Could concussions actually kill football?
- J.T. Thomas arrested for possessing some marijuana
- Bucs hire Butch Davis as ‘special assistant,’ have no issue with his double-dipping
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
