The Dude Abides: The Stats That Don’t Lie, Week 16

"Are you here to tell me what a bad statoogoolizer I am?"

--Derek Zoolander

In the NFL, I'm pretty sure there's more to life than being really, really ridiculously good-looking.  Week 16 proved this if nothing else.   Two teams were caught staring at their reflections for too long (New Orleans and Minnesota). Another team was tragically lost to the playoffs in a freak-gasoline-fight accident (Giants).    And the Oakland Raiders were back to normal, showing us all why they are still reserving spots in the Derek Zoolander Center-For-Kids-Who-Can't-Read-Good-And-Wanna-Learn-To-Do-Other-Stuff-Good-Too.  In other words, don't expect any playoff help from the Silver-and-Black next week.

For your Denver Broncos, their penalty issues have been a bad wardrobe problem for weeks, and it finally got them on the worst-dressed list.

Welcome to another edition of the Stats That Don't Lie, your weekly catwalk of stats.  These are the stats that realize a male model's life is a precious, precious commodity. They are your statistical hair gel in a world of bad hair.  As always they are Turnovers, Field Position, Time of Possession, and  3rd-Down Efficiency.

This week, as you've probably surmised, we're going to put the spotlight on penalties--and the movie Zoolander, perhaps the finest movie ever made (actually, the only one) about the cutthroat world of high-fashion-male modeling.

Penalties--Vain, Stupid, and Incredibly Self-Centered

A lot of Bronco fans have been saying that the officiating cost the Broncos the game against the Eagles. This is quite a subjective statement, for it assumes three things:

1) Most of the calls were bad or completely wrong

2) Most of the calls were one-sided against the Broncos

3) The calls could not be overcome and were decisive in the outcome

As a stats guy, I have a hard time with these kinds of subjective determinations, so for the purposes of my discussion about penalties, I will assume that all of the penalties in the game really were penalties (including those that worked in Denver's favor).  This will help us bring a tad more objectivity as we look at how penalties affected the game (and the Broncos' season).

First, however, let's look at how the Broncos have trended lately with regard to penalties in general.  For the season, the Broncos average 5.9 penalties and 50.1 penalty yards per game.  This puts them in the middle of the pack statistically.  Yet during the last 3 weeks, these numbers have dramatically increased to 7.7 penalties per game (4th worst) and 77.3 penalty yards per game (2nd worst).  To make things even more grim during this 3-week stretch, on average, they have been allowing opponents 3 first downs via penalty per game.  This puts them at worst in the league in this category.  They allowed 4 such first downs in the game against the Eagles.

Penalties & Points - I have an after-funeral party to attend

Penalties, over the long run, are moderately correlated to team wins, but the correlation is negative.    This makes sense intuitively, of course.   The more penalties you have, the harder winning becomes.  In the off season, I plan to run some correlation coefficients on 10 years of penalty data, but for now, we'll trust Brian Burke over at Advanced NFL Stats, who has run some penalty correlations for the NFL seasons of 2003-2006.  The correlation between team wins and penalties is somewhere in the range of minus-0.18 to minus-0.31.  Thus, penalties are on par with sacks in accounting for the variance in team wins. So they do have real value. 

Another way of looking at penalties is to run a multi-variable regression against a team's point differential using a variety of independent variable s, including penalty-yards differential (my opponent's penalty yards subtracted from my team's penalty yards).  This is exactly what Wayne Winston did in his excellent book, Mathletics, for the 2003-2006 seasons.  His finding?  That each yard of penalty differential is worth about 0.06 points.

We can now apply this to the game against the Eagles.  Denver had 95 penalty yards in the game; the Eagles had 59. That's a difference of 36 yards, or 2.16 points (36 x 0.06).   Does this mean that the penalties actually caused the Eagles to score 2 more points than the Broncos?  Certainly not.   We have to be careful in asserting this.  However, it would suggest that the difference in penalties in this game was certainly a contributing factor in the margin of victory.  

So for all of you that want to scream at the refs, you've got your ammunition.  They lost by 3 points; the penalties might have cost them a little over 2 points.  Again, this assumes that some of the penalties really weren't penalties after all.  

My main reason for bringing this to light, however is to point out how sloppy play (as defined by penalty yards) is really hurting the Broncos--specifically in the last three weeks.  Penalties mean something...if the differential between your penalty yards and that of your opponent's are great enough.

Over the entire year, though, penalties have tended to even out.  Here are the penalty yards for-and-against in the 15 games Denver has played, along with their respective potential points:

  Broncos Opponent Differential - Yards Potential Points
CIN 39 27 12 0.72
CLE 50 43 7 0.42
OAK 30 53 -23 -1.38
DAL 81 70 11 0.66
NE 20 46 -26 -1.56
SD 48 44 4 0.24
BAL 35 94 -59 -3.54
PIT 25 28 -3 -0.18
WAS 45 19 26 1.56
SD 65 15 50 3
NYG 61 45 16 0.96
KC 20 25 -5 -0.3
IND 65 20 45 2.7
OAK 67 62 5 0.3
PHI 95 59 36 2.16
Total 746 650 96 5.76

The final column on the right is the actual estimated points that penalties have cost Denver over the course of the season.  So what has been the cost of all of the bone-headed mistakes?  A whopping 5.76 points, or two field goals.  

When we get down to it, one could really only make a legitimate claim that penalties influenced the outcome of three Broncos games: Washington, Indianapolis, and Philadelphia.   And because Washington and Indianpolis beat Denver by much more than these point margins,  the Philly game remains the only real suspect. So let's not get carried away and claim the referees have had it out for Denver from Week One.  I'll give you the Eagles game, perhaps, and only because it amuses me to watch people handle pitchforks with their bare hands.  But the entire season?  No. 

Just be glad you are not a fan of the Baltimore Ravens, because you might have a legitimate gripe about the referees. Their penalty-yards differential for the year is 356 yards against.  This translates to 20.76 points, or almost 3 full touchdowns. I counted no less than three Ravens games (Cincinnati, Minnesota, and Pittsburgh) in which the points could have had a real influence on the outcome.   This doesn't mean the referees hated Ray Lewis.  It could actually mean that the Ravens legitimately commit a lot more turnovers than their opponents.  Either way, penalties have been a factor in several of their games.

Before we leave the penalty issue, perhaps you've asked yourself if the Colts--due to Peyton Manning's golden-boy image--get all of the calls to go their way.  While we can't answer this for sure (beside the Christmas-card exchange program Manning has with the referee's union), we can certainly determine if penalties have helped influence all of that winning the Colts have experienced this year.   The penalty-yards differential for the Colts this year is 318, which translates into 19.08 points in their favor.

They are essentially the opposite of the Ravens.  I counted 4 games (Jacksonville, New England, and  Houston twice) in which their penalty-yards differential could have had a real outcome on the final score.  I'll leave it to you to decide whether Manning gets all of the calls or not.  You can imagine where Gary Kubiak stands on the subject, however.

Team Rankings

TEAM TOs Rank TO Margin Rank Field Position Rank TOP Rank 3rd Down Rank
GB 1 1 1.47 1 32.92 1 32:45:08 1 45.29% 5
SD 1 1 0.64 5 30.26 15 30:15:47 15 44.02% 6
MIN 1.2 3 0.4 9 32.58 2 32:35:08 2 45.91% 4
DAL 1.2 3 0.13 15 27.45 29 31:44:04 29 39.48% 13
NE 1.33 5 0.47 8 30.4 12 33:24:36 12 43.62% 7
DEN 1.33 5 0.53 6 29.68 18 30:00:12 18 34.45% 24
IND 1.47 7 0.27 11 28.29 25 28:00:56 25 49.38% 1
CIN 1.47 7 0.2 13 30.16 16 34:04:12 16 42.97% 9
ATL 1.47 7 0.27 11 30.66 11 29:31:36 11 41.3% 11
BAL 1.47 7 0.53 6 31.95 4 29:40:44 4 39.97% 12
PHI 1.47 7 1 3 30.83 9 28:30:24 9 37.42% 18
SF 1.47 7 1.07 2 29.2 20 29:20:16 20 29.46% 29
NO 1.6 13 0.87 4 31.43 6 31:45:00 6 46.41% 3
JAX 1.6 13 -0.13 17 28.37 24 31:08:56 24 42.63% 10
PIT 1.6 13 -0.33 22 29.01 21 32:54:28 21 39.07% 15
WAS 1.67 16 -0.73 28 28.27 26 29:01:56 26 37.87% 17
KC 1.67 16 0.07 16 29.48 19 28:24:36 19 26.4% 32
MIA 1.73 18 -0.47 25 28.45 23 32:17:40 23 47.3% 2
HOU 1.8 19 -0.13 17 30.02 17 31:20:28 17 39.24% 14
BUF 1.87 20 0.33 10 30.87 8 28:02:00 8 27.69% 31
NYG 1.93 21 -0.33 22 30.74 10 32:13:44 10 43.3% 8
TEN 1.93 21 -0.13 17 27.32 30 28:27:04 30 39% 16
NYJ 2 23 -0.13 17 31.64 5 31:25:36 5 36.27% 20
SEA 2 23 -0.6 26 27.87 27 27:30:12 27 32.31% 26
CLE 2 23 -0.8 30 31.32 7 28:38:20 7 31.21% 28
CAR 2.07 26 0.2 13 28.48 22 30:08:24 22 37.07% 19
OAK 2.07 26 -0.73 28 27.24 31 28:02:00 31 28.64% 30
TB 2.13 28 -0.31 21 30.26 14 28:57:26 14 32.25% 27
ARZ 2.2 29 -0.33 22 30.33 13 30:24:52 13 35.23% 22
STL 2.2 29 -1 31 27.87 28 29:19:48 28 34.35% 25
CHI 2.27 31 -0.67 27 32.05 3 28:34:28 3 35.57% 21
DET 2.6 32 -1.07 32 26.62 32 28:45:52 32 34.72% 23

Last week I wrote a thesis about the Broncos' woes on third downs (complete with data), but if you want the summary points, here they are:

1) Denver continues to act like they are break-dance fighting on 3rd down, hovering between 34%-35%.

2) Next year, they will be over 40% on third downs as they continue getting used to the "Erhardt-Perkins" offense (just like the Pats did in their second year, going from 35% to 41%).

3) They will improve on 3rd downs next year because they fit the Football Outsider's profile of a team that is good on 1st and 2nd downs, but bad on 3rd downs.  

4) It's the offensive line which has been the issue, not any specific running back.  This means Peyton Hillis is not Hansel.

Everything I wrote last week about 3rd downs remains true this week.  Even if the Broncos make the playoffs, I've come to terms with the fact that it won't be until next year that this team has a drastic improvement on 3rd downs.

With respect to the other four statistical categories, it's more of the same.  First, Denver's average number of giveaways and turnover margin per game continue to be the strong point with this team (Translation: they have a conservative offense and an opportunistic defense).  Second, since the Broncos have been behind in many games since the bye, their time-of-possession statistic has eroded. That's because they continue to have to pass the ball in the 2nd half in order to come back.  This causes them to abandon the running game.  Third, Denver has made a few small strides in improving their field-position stat.  They are now nearly at the league average of the 29.94-yard line.  We will return to field position shortly in discussing Denver's game against Philadelphia.

I'd like to again point out a team that I think no one wants to face in the playoffs, and it's not the Cowboys.   It's the Packers.  They are 1st in turnover margin, 1st in field position, 1st in time of possession, and 5th in 3rd-down efficiency.    If you need further proof, check out this list of Top-10 teams by scoring ratio (points for/points against), which is about the best predictor of wins that you can find:

  1. New England: 1.59
  2. New Orleans: 1.57
  3. Baltimore: 1.49
  4. Indianapolis: 1.476
  5. Green Bay: 1.475
  6. San Diego: 1.43
  7. Minnesota: 1.39
  8. Philadelphia: 1.37
  9. Dallas: 1.34
  10. New York Jets: 1.31

The Packers are virtually deadlocked with the Colts in scoring ratio, scoring 1.47 points for every point they give up. They are also ahead of division leaders like San Diego, Minnesota, and Philadelphia.  Do not be shocked if they do serious damage in the playoffs.  With the the way the Saints have struggled recently, the Packers could represent the NFC in the Super Bowl.

Surprised to see the Patriots and Ravens so high on this list?  I wouldn't be.  With the Patriots, even when their record wasn't as good as it is now, they were generally in the top ten in all four of the statistical categories. And Baltimore has just been a little snakebitten, as we've seen with their penalties.  At the end of the season, I'll be doing a piece on "luck" in the NFL.  And the Ravens are sure to be on that list of "unlucky" teams.  Still, if the Ravens get into the playoffs, they could do some damage.  Everyone in the media will say they were a surprise team.  But you'll know differently.

Think I'm giving Cutler a free pass after he had one reasonable game?  Think again.  It seems that everyone in the mainstream media is ready to again anoint Cutler the "player the Bears thought they were getting."  But that's loco, ese.  Cutler has six more interceptions than anyone else in the league and has as many interceptions as Kurt Warner has touchdowns - and Warner is 9th in the league in touchdowns.

The Bears rank 3rd in average starting field position.  Imagine how much damage they could do if Jay Cutler wasn't a runway for red-zone interceptions.

The Games

TEAMS Score TOs TOP Third Down Field Pos   TEAMS Score TOs TOP Third Down Field Pos
SD 42 1 39:07:00 72.73% 32.4   JAX 7 2 27:35:00 36.36% 22.22
TEN 17 3 20:53:00 25% 30.44   NE 35 1 32:25:00 60% 32.25
                         
BUF 3 3 25:35:00 20% 19.45   TB 20 2 36:15:00 58.33% 25.7
ATL 31 0 34:25:00 47.06% 33.45   NO 17 1 30:39:00 54.55% 23.44
                         
KC 10 2 26:08:00 46.67% 26.27   STL 10 4 25:49:00 42.86% 34.64
CIN 17 1 33:52:00 46.67% 29.55   ARI 31 1 34:11:00 35.71% 27.62
                         
OAK 9 3 29:38:00 23.08% 29.54   DET 6 6 29:22:00 16.67% 25.21
CLE 23 1 30:22:00 37.5% 31.77   SF 20 1 30:38:00 29.41% 39.07
                         
SEA 10 4 27:46:00 35.29% 33   DEN 27 1 29:17:00 33.33% 39.57
GB 48 0 32:14:00 57.14% 33.69   PHI 30 3 30:43:00 30.77% 26.08
                         
HOU 27 1 32:38:00 36.36% 31.75   NYJ 29 0 33:06:00 43.75% 35.25
MIA 20 1 27:22:00 50% 26.27   IND 15 2 26:54:00 18.18% 23.67
                         
BAL 20 3 32:33:00 38.46% 35.75   DAL 17 1 35:02:00 53.85% 27.91
PIT 23 2 27:27:00 38.46% 36.5   WAS 0 1 24:58:00 28.57% 23.9
                         
CAR 41 0 32:16:00 66.67% 33.1   MIN 30 2 36:25:00 35.71% 28.77
NYG 9 4 27:44:00 38.46% 25.1   CHI 36 1 29:20:00 35.71% 38

There were three games in Week 16 that I'd like to focus on, starting with the Broncos-Eagles game.  In this game, Denver won the turnover battle by two, the 3rd-down battle by a few percentage points, and field position by a wide margin.  Each and every week I talk about how important it is to win these four statistical battles.  Denver virtually did just that, and yet they lost.  How was this possible?

Let's remember, they did almost win the game.  In fact, they should have.  But there are several reasons why they didn't.  The first were the penalties.  They played a factor in this game, as we've already seen.  Second, even though they won the battle on 3rd downs, they weren't exactly dominant, considering the league average is 38%.  Going 33% when your opponent goes 31% is like breaking even.  If the Broncos would have even converted 1 or 2 more of their 3rd downs, they would have been victorious.  Third,  the turnovers weren't necessarily the game-changing feature in this game, but functioned as a great equalizer.  In this case, the turnovers helped an inferior team climb back into a game in which they were almost buried in the first half.  Had the Broncos been in this game from the beginning, the turnovers certainly would have given them the win.  Lastly, field position in this game was only useful if Denver was able to take advantage of it.

This last point merits further explanation.  In the 2nd half of the game, Denver had 8 useful drives.  They did have a 9th, but it came with 4 seconds remaining, and therefore, it was a non-event.  Of these 8 drives,  there were 4 drives that began at or around midfield (the average was the 48-yard line, to be exact).  These were drives in which Denver could have possibly put a maximum of 28 points on the scoreboard.  But what did they come away with?  6 points.  On these 4 drives, they ran 22 plays and gained a grand total of 53 yards.  That's an average-gain-per play of 2.4 yards, including passes.  That, my friends, is a freak fest.  I guess now I know why McDaniels decided to sneak on 2nd down in the previous game.  He was simply calling a play that might gain 2.5 yards.

For the entire Eagles game, the Broncos didn't fare much better.  Both their passing and running games yielded an average of 3.9 yards per attempt.  That's why Kyle Orton threw it 41 times in a game and only ended up with 171 passing yards. Again, if you want to win, you have to take advantage of the opportunities field position brings you.  The results are in, Amigo.  What's left to ponder?

Before moving to the other two games, don't be fooled by the comment by Phil Simms during the game in which he said that blitzing Donovan McNabb was a bad strategy.  Nothing could be further from the truth.  Unless you have a lights-out front 4, blitzing every QB is a good strategy (outside of Peyton Manning). And even with Manning, one ought to do it often.   But rather than simply disagree with Simms, I'll just give you the numbers:

Donovan McNabb's QB Rating:

  • No pressure (294 drop backs): 102.5
  • Plays under pressure (116 drop backs): 66.2
  • When blitzed (115 drop backs): 78.0

Peyton Manning's QB Rating:

  • No pressure (423 drop backs): 107.4
  • Plays under pressure (121 drop backs) 78.4
  • When Blitzed (130 drop backs) 98.3

Simms might have been a great quarterback, and he's certainly a decent color man, but as a defensive coordinator, he's not taking Dick LeBeau's job anytime soon.

Now, I'd like to discuss two other games, both of which are useful in driving home a point that I continue bringing up every week.  And it is this:  a turnover is worth more the later in the game it occurs.  Let's first consider the Tampa Bay-New Orleans game.  The Saints had a total of 9 drives in the game.  On their 7th and 9th drives, they turned the ball over (the 9th was a turnover as a result of a missed field goal).  The score before the two drives? 17-3 in favor of New Orleans.   I think you are getting the picture.  The Bucs tied the game up, got the ball in overtime, and scored.    

Drew Brees and the Saints never got the ball back again because they had no time.  Had these turnovers occurred in the first quarter, Brees would have had at least 5 or 6 drives with which to overcome these mistakes.  But because of the constraints of time, the costs of the turnovers were magnified.  And you always wondered by coaches like Marty Schottenheimer ran the ball to the point of pain in the 4th quarter.  

The game on Monday night demonstrates this even more precisely.  In the overtime period, the Vikings and Bears traded possessions.  On their 2nd possession, the Vikings fumbled inside their own territory.  Chicago went on to score one play later.  Game over.  Why? Because the Vikings would never get another possession due to the NFL's sudden-death rules. We'll leave the silly way the NFL handles its overtime period for another piece, but simply note that because overtime rewards the first team to score, turnovers (and possessions) are at a a premium.    

The Summaries

2-Turnover Rule (explained here)

  • 4 teams had 0 turnovers. 4 won (100%); For the season, 71/91 (78%)
  • 13 teams had 1 turnover. 9 won (69%); For the season, 84/142 (59%)
  • 6 teams had 2 turnovers.  2 won (33%); For the season, 58/128 (45%)
  • 5 teams had 3 turnovers. 1 won (20%); For the season, 23/72 (32%)
  • 3 team had 4 turnovers.  None won (0%); For the season, 4/32 (13%)
  • 1 teams had 5+ turnovers.  This team lost (0%); For the season, 0/15 (0%)

Week 16 - Big Picture

  • The team that won the turnover battle (removing ties) won 12 of 14 games played (86%) 
  • 2 games were tied in the turnover battle.
  • The team that won the time of possession battle won 14 of the 16 games played (88%).  
  • The team that had better third-down efficiency won 13 of the 16 games played (81%).
  • The team that had better average starting field position won 14 of the 16 games played (88%). 
  • There were 9 games this week in which a team won all four categories.  In all 9, the same team won on the scoreboard (100%).  
  • The winning teams this week averaged 1.00 turnovers, 32:45  in time of possession, 46.88% on 3rd downs, and their average starting field position was the 32.13-yard line.  

Running Totals, Season (through 16 weeks):

  • 240 games have been played this season
  • 76.5% (or 153/200) were won by the team with less turnovers (removing ties)
  • 40 games were tied in the turnover battle
  • 67.29% with a better time of possession
  • 69.58% by the team who won on 3rd down
  • 70.83% by the team that won the field-position battle.

16-Week League Averages:

  • 16-week running average/game, turnovers (all teams): 1.72
  • 16-week running average/game, time of possession (all teams): 30:11 (thanks to OT games)
  • 16-week running average/game, 3rd down efficiency (all teams): 37.97%
  • 16-week running average, starting field position (all teams) 29.94-yard line

The Look Ahead

Looking ahead to Kansas City, here is how the two teams stack up in the four statistical categories (keeping in mind these are averages/game):

Team Turnovers/Game Time of Possession Field Position 3rd-Down Efficiency
Denver 1.33 30:00 29.68 34.45%
Kansas City 1.67 28:24 29.48 26.40%

When it comes to these stats, Kansas City really is a hand model in a world of beautiful people.  Denver has the advantage in each category and should win this one big.  Interestingly enough, Denver finally found a team that is worse than they are on 3rd downs.  I guess we'll call this a Christmas present one week too late.  

Last night it was announced that Brian Dawkins, Champ Bailey, Brandon Marshall, Ryan Clady, and Elvis Dumervil all made the Pro Bowl.  But these guys know that it will take more than being professionally good-looking to beat the Chiefs. And if Denver happens to trip into the playoffs come Sunday, these guy will be ready.  Like the world of male modeling, you've always got to be ready for a straight walk-off, old school rules...for serious.

Hail, Orton's Neck Beard!!!

Last stat: 29.08% (This is the real percentage chance that Denver makes the playoffs.  Yes, really)

Originally posted at MHR

I’m glad we had this talk.  Now, vaya con Dios, Brah.

Agree, disagree, just like us on Facebook and follow us on Twitter so I can quit my day job.

Mark it ZeroStats That Don't Lie

2014 Offseason

Offseason coverage