"You stats shouldn't play so rough. Somebody's gonna start crying."
Some teams bark all day, little doggy. And some teams bite.
Week 15 was supposed to be an easy caper for the playoff-bound teams in the NFL. Minnesota, New Orleans, Philadelphia, and Indianapolis were all looking to get in, do their job, and get out--unscathed and without too much gun play. Only two of these teams (Philly and Indy) actually escaped with a win.
As usually happens when jobs get botched, the better team isn't done-in by the play of the other team. No, that's rarely how it happens. Usually it's an inside job.
This was certainly the case with your Denver Broncos. While the Oakland Raiders showed they had some ammunition, it was the Broncos that turned on themselves. The Broncos were too careless (missed tackles, bad angles, and missed blocks). They were too cute (a pass to Ryan Clady and a QB sneak on 2nd down). And in the end, there they were - wounded from field goals and gashed from the running game, while JaMarcus Russell walked away with the cash.
Welcome to another week of The Stats That Don't Lie, your weekly rendezvous where we pick our statistical aliases. These are the stats that don't believe in tipping. They are the stats that don't talk, no matter how much you torture them. As always they are Turnovers, Field Position, Time of Possession, and 3rd-Down Efficiency.
This week's edition is real special, too, and not just because I'm making Reservoir Dogs references. It's because I'm finally going to exhaust Denver's 3rd-down woes. (Note: I tried to link all references for Broncos fans born after 1990).
The Big Picture & 3rd Downs
Usually I put up a graph each week expounding upon the turnover margin in Denver's win/loss. This week, Denver won the turnover battle and lost the game, so it's a good time to look at what has been another key area for Denver (we´ll deal with turnovers shortly). It´s no secret that Denver has struggled on 3rd downs. But just how bad has it been? Not as bad as taking one to the gut, but let´s take a look:
The blue line is the league-average per game since the Broncos' bye week. It has ranged from 37.13% to 37.79%. This is normal. The league average usually runs between 37% and 38%. The orange line is Denver's 3rd-down percentage per game through the bye week as well. In Denver's last 8 games, they've manage to best this league average but one time. In 3 of the 8 games, they didn't even hit 30%. As you'll see in the rankings, they are averaging a paltry 34.53% per game.
Clearly, this is a huge problem. Consider the 15-week average per game of the playoff teams if the season were to end today, in addition to the four biggest threats to Denver's wild-card spot.
- Indianapolis: 51.61%
- San Diego: 41.81%
- New England: 42.45%
- Cincinnati: 41.71%
- Baltimore: 40.08%
- Denver: 34.53%
- Jacksonville: 43.08%
- PIttsburgh: 39.11%
- Miami: 47.10%
- Tennessee: 40.00%
Someone on this list doesn't have a cool-sounding 3rd-down rate, and we are not talking about the Colts. Even though the Broncos are within one standard deviation (about 6%) from the mean of 37.71%, the Broncos are fortunate that their defense has produced some timely turnovers this year. One might even wonder how a team with such a bad 3rd-down percentage is in the position of making the playoffs. For now, let's chalk it up to orange-and-blue leprechauns and move on.
3rd Downs - Hooked On a Feeling
The real question is, "Why have the Broncos done so poorly on 3rd downs all year long?" Stats can only do so much of the heavy lifting, but we'll give it a shot, anyway.
Many have postulated that it's the offensive line, the play calling, or Knowshon Moreno himself. Some view Peyton Hillis as the bionic man. Perhaps he should be getting more carries. Or some want to blame a certain bad ass with the finest neck beard this side of the Mississippi. I'm hearing none of it.
I have a different theory. It's less complex. And it doesn't require pointing fingers.
Simply, it's this team's first experience with "Erhardt-Perkins" offense. As Tom Brady has repeatedly alluded, getting used to this complex offensive system, installed with the Patriots by Charlie Weis, takes at least two years to get used to, if not more. But let's not take Brady's word for it, let's look at the numbers.
We simply need to go back to the 2000 Patriots, when Weis installed the offense and look at the Patriots' 3rd-down conversion rate between 2000 and 2001. In 2000, the Patriots were 35% on 3rd downs. In 2001, that number jumped 6% to 41%. Now, there is one big difference between 2000 and 2001. And that's Tom Brady at quarterback. But it was his second year in the system, and I don't believe Brady's numbers would have been the same had he had that first year as a rookie to learn Weis's system.
The Football Outsiders, who many stats geeks worship like I do Pam Grier, would agree. They've had a theory going about 3rd downs for several years:
Teams which are strong on first and second down, but weak on third down, will tend to improve the following year...over time, a team will play as well in those situations as it does in other situations, which will bring the overall offense or defense in line with the offense and defense on first and second down.
This year Denver has rushed for 977 yards on 1st down, good for 4th-best in the league in that category. Here is the precise breakdown on 1st and 2nd downs when compared to 3rd down:
- 1st Down-Average Running Play: 4.2 yards; Average Passing Play: 7.02 yards
- 2nd Down-Average Running Play: 4.71 yards; Average Passing Play: 5.99 yards
- 3rd Down-Average Running Play: 2.62 yards; Average Passing Play: 5.19 yards
Your eyes do not deceive you. On 3rd down, Denver struggles. If the Football Outsiders' theory is correct (it has been over the last 7 years) and history repeats itself with the Broncos' learning curve with respect to the "Erhardt-Perkins" system, we are looking at a team next year which is poised to improve drastically in 3rd-down efficiency. Take it to the bank.
Stuck in the Middle With You on 3rd-and-Short
Since we are beating the 3rd-down horse, it's also useful to bring up the play calling on 3rd and short. A lot of fans have been lamenting the Broncos' struggle on 3rd and short and hoping the team would pass more often. While this is completely understandable, just know that the percentages actually favor running in these situations--by a good margin. Once again, I present to you one of the little gems from the Football Outsiders:
On average, passing will always gain more yardage than running, with one very important exception: when a team is just one or two yards away from a new set of downs or the goal line. On third-and-1, a run will convert for a new set of downs 36 percent more often than a pass. Expand that to all third or fourth downs with 1-2 yards to go, and the run is successful 40 percent more often. With these percentages, the possibility of a long gain with a pass is not worth the tradeoff of an incomplete that kills a drive...
...Teams pass roughly 60 percent of the time on third-and-2 even though runs in that situation convert 20 percent more often than passes. They pass 68 percent of the time on fourth-and-2 even though runs in that situation convert twice as often as passes.
As bad as Denver has been in short-yardage situations this year on 3rd and 4th down, it's still not swimming upstream against these percentages. And if you don't think McDaniels knows the percentages, you are fooling yourself. Still want more proof? Consider that on 3rd-and-3 (or less):
Kyle Orton is 7 out of 15 passing (54%)
Knowshon Moreno is 10 out of 15 rushing (67%)
Moreno just happens to be converting on 3rd-and-3 (or less) about 24% more often than when Orton throws in the same situation. This is slightly higher than the Football Outsiders would estimate, but we are dealing with a relatively small sample size. However, the point is still valid. McDaniels is generally making the right call when he continues to run the ball on third and short.
For those that believe Peyton Hillis is Charles Bronson, ask yourself if he could have done any better than 10 out of 15? If you can look yourself in the mirror with any seriousness and answer, "yes," than we'll just have to act like professionals. Maurice Jones-Drew has picked up 17 out of 22 (77%), so you'd be putting Hillis in that kind of company. Given that Hillis has only averaged 3.4 in his 5 carries this year up the middle, I'm skeptical. Moreno has had 62 carries up the middle and averages 3.7.
The Offensive Line - Hold It...A Little Big Longer
If you want to point to the offensive line and losing Ryan Harris, then we've got some common ground. This loss is a lot bigger problem than McDaniels would like you to believe. Here is what Denver was averaging per carry on the ground (pre-Oakland) through 14 weeks:
Left End Left Tackle Left Guard Middle Left Middle Right Right Guard Right Tackle Right End
4.2 4.6 5.2 3.6 3.4 4.4 5.5 4.7
If you were scouting the Broncos and you saw these numbers, it would become clear to you very quickly that the strength of this offensive line was on the right side of the ball in general and on the outside edges--at least with respect to running. It's even clearer when you look at the right-end runs and left-end runs, because Daniel Graham is shifting in the formation and these numbers are directly related to him. If the left side of the line were as good as the right, you would expect the numbers on the right-end to be similar to those on the left-end. This isn't a knock on Ryan Clady--far from it if you look at the numbers off left tackle. It's simply a recognition of the importance of Ryan Harris (and also Polumbus, who has played well in the running game).
This is data a casual fan comprehends, but it is also backed up by a short-yardage asssessment as in BShrout's impressive analysis here. In addition, it's backed up by slightly more complex statistics as well, as witnessed by Jeremy Bolander's excellent post last week, in which he applies Sharpe Ratios to running plays behind each of Denver's offensive linemen. Bolander cleverly noted the reward gained (represented by the standard deviation) in-and-above a "safe" play of a zero yard gain:
...the ratios for running behind Weigmann, Hochstein/Hamilton are generally quite poor, at 0.14 and -0.18. Clady is better, clocking in at .300375 while Kuper dominates in short yardage at 1.78885. Too often, however, the one cut has been designed to bend back away from Kuper's stellar blocking. When the blocking scheme is more straight ahead, we seem to take advantage of his side more.
This certainly correlates to a general strategy of Denver being more effective when running to the right side.
When taken as a whole, all of this would suggest that Denver is going to continue to struggle a bit on 3rd downs. The quarterbacks and receivers continue to learn the complex "Erhardt-Perkins" scheme, and the absence of Harris and the play of the offensive line, while it hasn't been horrendous, has shown consistency issues. As other teams continue their run slants, rush blitzes, and stacking the line of scrimmage, (as both the Raiders and Colts did), look for McDaniels to counter with strategies like play action.
Here's where the Broncos stack up against the rest of the league after Week 15. As always, remember these are averages/game (in my mind a better measure than aggregate numbers):
While Denver continues its spiral on 3rd downs (the St. Louis Rams rank just below them), they still rank high in both giveaways (which I consistently label "turnovers") and turnover margin per game. Don't turn the ball over. Play good defense, capitalize on the mistakes of the other team. This was, is, and will continue to be their ticket to winning.
Given Denver's ranking in average starting field position (28.98-yard line) per game, I was pleasantly surprised in the Oakland game to see Kenny McKinley back returning. I thought he showed a great burst and quickness that will only get better. Although I had been clamoring for Alphonso Smith, I'll take what I can get. I was never sold that Eddie Royal was the longterm answer. Not because he wasn't capable, but because we'll never see another Tim Brown in this league. I think letting Royal be a receiver is the right move. And, outside of the San Diego game, the numbers weren't eye-popping (23.9 yards/kick). Royal was being asked to do too many things. When you've got jokers to your left and Raiders to the right, you need a guy who is focused on returns and rested between each series.
There's not much to say about Denver's time of possession other than they are sitting right at the league average.
From the rankings above, one team you must absolutely love is the Green Bay Packers, even after their loss to the Steelers in Week 15. I know, I know, the Steelers threw the ball all over the yard on them. But Aaron Rodgers is as underrated as Jay Cutler is overrated. They rank 1st in turnover margin, 2nd in giveaways, 1st in field position, 4th in time of possession, and 5th in 3rd-down efficiency. While the Packers are no Foxy Brown, they are still a whole lot of playoff team. I continue to think that the Packers can play with the Saints and Vikings, and I would personally love to see Rodgers send Favre back to retirement. If that makes me a Favre hater, then okay, but I prefer to think of it more like I am a hater of the Mainstream Media crush that exists for so-called "gunslingers." I mean, it's not like the the guy is taking anyone on a magic carpet ride with his arm this year anyway.
The teams that are the most dangerous from these rankings are the teams that rank high in both turnover margin and 3rd-down efficiency, because once the playoffs start, you need a team that at once doesn't make errors and which capitalizes on errors. And a team that can "make plays," which is just code for converting on 3rd downs. Right now, I would say this list includes, in no particular order:
- San Diego
- Green Bay
- New Orleans
- New England
Each of these teams is in the top ten in both categories, although the Chargers seem to have the most momentum of the whole bunch, as much as it pains me to say it.
Last week a few of you pointed out that perhaps I should finally let up on Jay Cutler as he continues his assault on history. I thought about it for one Sunday, but before I could say "Like a Virgin," "True Blue," or "Borderline," Cutler had thrown 25 picks for the year. This is the most in NFL history for a non-rookie quarterback in his first year with a new team. That's a strange stat indeed. Suffice to say the guy is to interceptions as a Tarantino movie is to F-Bombs. You lose count 20 minutes in.
Imagine how bad it would be if the Bears didn't rank 5th in average starting field position. Actually, it might have worked out better for the Bears if they didn't rank so high because then Cutler wouldn't begin so close to the red zone.
Week 15 - The Games
Let's start where we always begin, with the Broncos-Raiders game Since I was a big advocate two days ago of simply forgetting the game, I won't say too much, except that given the Raiders turned the ball over twice, it would appear that Denver would have won this game. But, if you read these
ramblings columns on a weekly basis, you know I am a big "context" guy. Yes, turnovers are bad. Yes, if you are plus-1 in turnover margin, the stats say you'll win about 70% of the time. Get to plus-2 and you're at 80%. Yes, all of this is true. However, you also know that turnovers committed early in a game can be overcome--even by a decade of washed-up first-round draft picks known as the Oakland Raiders. Oakland committed one of its turnovers on its 2nd drive (1st quarter) and the other on its 8th (3rd quarter). But they had 12 drives in the game with which to work.
And what did Denver do with these two turnovers? They put up 10 points off of them. And yet they still did not win. This is what happens when you crawl around at 26% on 3rd downs. Not even Lee Marvin could get you out of that one.
If you want another interesting stat that hasn't received any attention from the Denver game, it's the following: on Oakland's scoring drives, they ran the ball 13 times for a total of 94 yards. This represents an average of 7.2 yards per carry, which is almost what Oakland's average was for the entire game running the ball. But the standard deviation of this data set was 8.8 yards. This is my smarty-pants way of saying that the Raiders running game was feast or famine. They had a lot of 1- and 3-yard carries, but also several carries over 20 yards.
Everyone is commenting on the New Orleans-Dallas game, so I might as well throw in my two cents. New Orleans had 11 drives in the game. They committed turnovers on 3 of them. That left them with 8 drives. And the Cowboys got out to a big lead, so in the 2nd half they ran the ball effectively and killed the clock, which led to their domination of the time of possession. Did I mention that Drew Brees was 14% on 3rd down? Given that New Orleans is usually at 43% on 3rd downs, this was quite a feat. Dallas got pressure, and they got it with their front 4.
Two other games I briefly want to mention. First, the Jags-Indy game. Two weeks ago, I developed a profile of what a Colts loss would look like, based on their losses since 2005. In Week 15, the Jacksonville Jaguars almost matched that profile. They ran, and ran, and ran, rushing 34 times in the game. They tried their very best to keep Peyton Manning off the field. They matched the Colts step for step on 3rd down. And what happened? Oh, just a 93-yard kickoff return from the Colts for a touchdown. Manning lives to fight another day.
The second game I wanted to point out is the Eagles-Niners game - if for nothing else, to simply point out that San Francisco was 0% in 3rd-down efficiency. The Niners also turned the ball over 4 times. But somehow they avoided the blowout and made a game of it. They even managed an average of 5.9 yards per rush against the Eagles. The Eagles can be beaten.
2-Turnover Rule (explained here)
- 7 teams had 0 turnovers. 5 won (71%); For the season, 67/87 (77%)
- 8 teams had 1 turnover. 4 won (50%); For the season, 75/129 (58%)
- 7 teams had 2 turnovers. 5 won (71%); For the season, 56/122 (46%)
- 6 teams had 3 turnovers. 2 won (33%); For the season, 22/67 (33%)
- 2 team had 4 turnovers. None won (0%); For the season, 4/29 (14%)
- 2 teams had 5+ turnovers. None won (0%); For the season, 0/14 (0%)
Week 15 - Big Picture
- The team that won the turnover battle (removing ties) won 8 of 11 games played (73%)
- 5 games were tied in the turnover battle.
- The team that won the time of possession battle won 11 of the 16 games played (69%).
- 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 7 of the 16 games played (44%).
- There were 4 games this week in which a team won all four categories. In all 4, the same team won on the scoreboard (100%).
The winning teams this week averaged 1.25 turnovers, 31:32 in time of possession, 44.32% on 3rd downs, and their average starting field position was the 30.82-yard line.
Top 5 in Week 15:
- Top 5 in giveaways: 7 Teams Tied with 0.
- Top 5 in time of possession: 1.Carolina 2.Dallas 3.Jacksonville 4.Pittsburgh 5.Cleveland
- Top 5 in 3rd downs: 1.Giants 2.Indy 3.Jacksonville 4.Green Bay 5.Dallas/Saint Louis
- Top 5 in avg. starting field position: 1.Baltimore 2.Cleveland 3.Tampa Bay 4.Buffalo 5.Giants
Running Totals, Season (through 15 weeks):
- 224 games have been played this season
- 75.80% (or 141/186) were won by the team with less turnovers (removing ties)
- 38 games were tied in the turnover battle
- 66.52% with a better time of possession
- 68.75% by the team who won on 3rd down
- 69.64% by the team that won the field-position battle.
15-week League Averages:
- 15-week running average/game, turnovers (all teams): 1.71
- 15-week running average/game, time of possession (all teams): 30:10 (thanks to OT games)
- 15-week running average/game, 3rd down efficiency (all teams): 37.79%
- 15-week running average, starting field position (all teams) 29.91-yard line
The Look Ahead
Looking ahead to Philadelphia, 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 1.36 30:03 29.98 34.53%
Philadelphia 1.36 28:20 31.17 37.90%
The teams are even on turnovers. The good news is that Philadelphia's time-of-possession stat tells you one important thing about this game. The Eagles are a passing team. This plays into Denver's strength. It's also encouraging that as a passing team, the Eagles' 3rd-down-conversion percentage isn't higher. This again bodes well for Denver, especially if Denver can shut down DeSean Jackson on special teams.
The traditional thinking is that home field advantage adds about 3 extra points of value. In this game, I'd also remove that off the top because this is a homecoming for Brian Dawkins. With the kind of love that city has for BDawk, you might as well add three points for the visitors. Dawkins is going to put the lime in the coconut and drink em both together after this win. Ramblers, start ramblin'.
Hail, Brian Dawkins!!
Last stat: 99.99% (percentage chance that they would let Brian Dawkins have the alias Mr. Black, without fighting)
Originally posted at MHR