There are five NFC teams (CHI, GB, NO, NYG, and PHI) ranked ahead of the Falcons in the latest installment of the M-P rankings despite the fact that the Falcons have the highest winning percentage in the conference and are tied for the lead league in wins and winning percentage (with NE). The Dirty Birds have played three of the NFC teams ranked ahead of them, and beaten two of them. Of the 11 total teams ranked ahead of them, the Falcons have played five and gone 3-2 against that set – beat the Packers, Saints and Ravens, and lost to the Steelers before getting worked by the Eagles. The graph below charts the M-P rank of the Falcons (bolded Blue line) as well as the five teams ranked ahead of them in this week’s pole. After 13 weeks of play in the NFL, the road to the Super Bowl passes through the Georgia dome (where the Falcons have yet to lose)…so, why no M-P love for the Dirty Birds?
The proof is in the pudding. As a start, check out ATL’s category scores through week 13 in the graph below. Very good teams should score positively in the x-axis items that end with the letter “o” (for offense) and negatively in the x-axis items that end with the letter “d” (you guessed it). Off the bat, it’s easy to see that the Falcons don’t run the ball well nor do they defend against the run well (in other words, a good running and run defending team would have those scores flipped). The same can be said of passing offense and defense, which is also flipped in the wrong direction for a good team. This is especially surprising given the season that Matty Ice is having. No points for 4th quarter comebacks, I suppose.
What matters most, however, is that ATL ranks only 9th in the most significant category of offensive play success (z-score of .36, versus .98 for first place in category NE). Moreover, ATL is way at the bottom (ranked 22nd) in the third most important category of defensive play success (.22 versus -.79 for first place in category NYG). The Birds make up for it a bit by ranking 2nd in scoring offense, which is the second most important category (.43 versus .59 for first place NE).
How does ATL compare to the teams ranked ahead of it that it has played? Those results are even uglier, especially against the NFC squads. The graphs below are broken up just to make them easier to read, and the results speak volumes. Based on the explanations above, readers should be able to figure out what’s going on here. Versus the NFC competition, ATL looks good only on scoring offense. Against the NFC, ATL holds the scoring offense and offensive play success advantage against PIT and BAL…which makes sense given the offensive snooze fest the AFC North rivals put on this past Sunday night.
NFL talking heads, a pantheon I can join only if I keep this up for many years, speak often of the supposed parity in the NFL this season. There are many ways to both qualitatively and quantitatively define parity. For instance, does parity mean that teams that observers expect to lose a particular game because their winning percentage is less than their opponent’s actually win that game 50% of the time? In other words, statistically speaking, does NFL parity mean that game results are as predictable as a coin flip?
Hardly, but we thought it would be interesting to look at parity through the lens of the M-P ranking system. Arguably, because our system rates the relative factors of a particular team’s success versus its opponent and the rest of the league, the M-P rankings should be more valuable at determining league parity rather than results parity.
The graph below tracks the variance in points earned across all 32 teams after each week of the NFL season from after play in week 3 through after play in week 16 in each season from 2000 to 2009 and week 3 through week 10 for the 2010 season. The 2010 season is in bolded blue (the graphs are easier to make out if you click on them to enlarge). The variance of M-P points earned by each team after each week is a good indicator of parity because the points represent an aggregate of z-scores reflective of our offensive and defensive play-by-play inputs. In seasons where the NFL is competitive, one would expect teams to cluster more closely around the mean (a z-score of 0) than in seasons where the league was dominated by a particular conference, division, or team.
This season does appear to feature the greatest parity through weeks 3 to 10, although it is clearly following a well established pattern from other seasons. It makes intuitive sense that as the season goes on the good teams separate from the bad (even in years when the league is very competitive) and variance increases week to week. Ironically, the gulf between good and bad teams was the most severe at nearly all points last season.
So, through 10 weeks of play in each of the last 11 seasons, what season has the greatest parity? It turns out to be the 2010 season. Through last weekend, the average variance in points earned in weeks 3 through 10 was 10.99, compared to 11.32 in 2000 (second) and 20.02 last season (greatest disparity). The results are featured in the table below.
Another way to think about parity is to ask whether parity exists to a greater degree among the above-average teams or the below average teams. The graph below compares the range between the number one ranked team in each week and the mean and the last ranked team and the mean. This season, not surprisingly considering that our last place team has been the same team that has gotten consistently worse since after play in week 2 (tough year for folks in Carolina), the league’s last ranked M-P team is “worse” than the league’s top ranked M-P team is “good”. So does that mean that greater parity actually exists at the top?
That theme holds true for the group of teams above average and the group of teams below average as well. More often than not week to week, the below average teams are further below the mean than the above average teams are above it. As a result, the variance in weekly points earned by above average teams is less than the variance in weekly points earned by the bottom half. The greater in amount the mean or variance of points for each group is also helps determine the degree to which a season has more or less parity. Compare the y-axis scales of the 2010 and 2009 season graphs below, which are the two seasons in the last eleven that have featured the greatest and least amount of parity through week 10, respectively.
On paper there are some colossal mismatches this weekend. Consider the following:
(5-2) PIT at (2-5) CIN (1-6) DAL at (5-3) GB (5-3) NO at (1-6) CAR (5-2) NYJ at (2-5) DET (6-1) NE at (2-5) CLE (4-3) CHI at (0-7) BUF
I became curious about how these look from M-P’s point of view, so I did some research. In some cases, these matchups are as one sided as they appear. The biggest gulf is the Steelers-Bengals game where PIT holds a 22-spot advantage over CIN in the M-P rankings (second versus 24th). Of the games above, the most closely matched from M-P’s point of view is the Bears-Bills game, where the Bears hold an 11-spot lead over winless BUF (19 v. 30). The average separation is a little more than 16 places on the M-P scale following play last weekend (week 8).
Most surprising to me is how the matchups breakdown category-to-category, especially since there are quite a few instances where the underdog is ranked higher than the favorite in a particular category. CIN, for instance, is ranked four spots better than NE in M-P’s highest weighted category, offensive play success (14th for CIN, 18th for the Brady Bunch), and a touch higher in rushing offense. The second biggest difference between M-P ranked teams is 19 in the Saints – Panthers game, yet the Panthers rank higher than New Orleans in three categories including defensive play success (third most heavily weighted category).
And of course the Cowboys bring some irony to their matchup with the Packers, even when it comes to M-P rankings. The only matchup where the “favored team” (in terms of both M-P rank and win percentage) is ranked higher than the underdog in every M-P category is the GB-DAL game. This despite the fact that the old NFC rivalry features teams separated by “only” 14 spots on the latest M-P pole, second least in the sample above.
Are we learning? Two weeks ago we took a look at average rank variance for both computer based and subjective ranking systems and found that M-P had the highest variance. Checking in after week 7, we find that our average rank variance has decreased by more than five, while ESPN’s has gone up by almost two. Covers has improved the most by nearly eight. So what, if anything, does this mean?
Perhaps it’s too early to look back into this issue, but it appears as though ESPN has become more sensitive to week to week performance over the last two Sundays while the teams that perform well based on the metrics that matter for the statistical models have started to solidify themselves in each systems ranking universe. That is, teams are falling into line, becoming “who we think they are” to paraphrase a former Arizona Cardinals coach.
Of course, there are always surprisingly big moves in M-P each week, and that brings us to our look back.
A look back: OAK v. DEN The Raiders moved up 9 spots this week to claim the dubious achievement of being our biggest positive mover this week. Their opponent, the suddenly hapless Denver Broncos, tumbled 7 spots (biggest loser). That’s what happens when a division rival drops 59 on you at home, I suppose.
What can the M-P model say about the way a game was won? For one, the Raiders like having Darren McFadden around. He scored enough points on his own to have beaten the Broncos by four points. The Raiders jumped from 14th to sixth in M-P’s rushing offense rankings. But what moved the Raiders ranking was the improvement in their scoring offense and play success on defense, our second and third most important categories. The silver and black moved up 9 spots in both. Perhaps not surprising given the 59 points they put up and only 14 they allowed.
The Broncos, on the other hand, got dominated in every category one might imagine: rushing defense, passing defense, and offensive and defensive play success.
Bring up the rear I noticed while compiling results of our stats-based model competitor Covers.com that the bottom three teams were far behind the rest of the field in terms of points. The 30th ranked San Francisco 49ers are 3.8 points worse in this week’s Covers.com poll than the 29th ranked team. This is the largest point differential between teams in the ranking. Additionally, the difference between the 49ers and the 31st ranked Covers.com team, the Bills, is only 0.3 (one of the smallest differences) and between the Bills and Panthers 0.9. In the Covers.com poll, the really bad teams are really in a class by themselves.
The disparity between the bottom 3 and the top 29 in Covers.com got me wondering if the same trend exists in the M-P rankings. Indeed, the largest difference between team points is between the 30th ranked Bills and the 31s ranked TB Buccaneers (1.90). The difference between the Bucs and the last ranked team (you guessed it, Panthers) is a below average 0.33. So yes, M-P “dislikes” the bad teams in its rankings, too. But what’s really interesting to me is that this week Tampa Bay Bucs coach Raheem Morris called his 31st M-P ranked Bucs, “the best team in the NFC.”
A look ahead: NYJ v. GB This is a big matchup for a lot of reasons. Some consider the Packers to still be the class of the NFC and the Jets are coming off a bye and look to keep rolling. It's also a pretty interesting matchup on paper.
Almost halfway there At the close of play Tuesday morning, a few teams will have played half of their games. As such, it's time our readers check out our projections broken down by conference. New York fans will be pleased.
It's been a busy week, so I am just posting these without any write-up. Thought I should first point out, though, that there have been 4 games where Pinnacle has moved the line at least one-half point since Wednesday, and 3 of those 4 games the move has been toward Massey-Peabody's number. This is not surprising--in fact, this has been the trend throughout the season. If our system has predictive power not contained in the point spread (and historical data indicates it does), we would expect these kinds of line moves, since the closing point spread is generally more efficient than the opening (and mid-week) point spread. So, to summarize, if line moves tend to be predictive, and Massey-Peabody has predictive power controlling for the closing line, we would expect line moves to be toward the Massey-Peabody line more often than not, so betting opportunities should be better early in the week.
Ok, on to the plays...here are the strongest, which I would feel comfortable blindly betting:
Green Bay Packers -2.5 vs. Minnesota Vikings (M-P Line: GB -7.5)
Which is the most complete team? Forgetting for a second that a team’s ranking in each category is weighted in determining that team’s overall ranking, I thought it would be interesting to see which team is the most “complete” based on the play-by-play categories we consider in our analysis. Given that only 4 of the top 10 teams come from the NFC and only two of those teams (NYG and NO) have records above .500, it makes sense that, coefficient weights aside, the two most complete teams as far as the M-P rankings are concerned come out of the AFC. PIT, ranked number one overall, has an average category rank of 9. NYJ comes in just a hair ahead with an average rank of 8.75.
The Jets excel in running offense (#1), defensive play success (#1), scoring defense (#4) and scoring offense (#5). The Steelers have the top ranked running defense, scoring offense and scoring defense, and rank third in passing offense. PIT’s 20th and 24th ranked offensive play success and running offense, respectively, bring up their average relative to the Jets.
However, the reason the Steelers climbed to number one this week while the Jets fell a spot to fifth is because the Steelers far outrank the Jets in two important categories. PIT ranks first in scoring offense (second most important category) and third in passing offense (fourth most important), whereas the Jets sit fifth and 29th in those same categories.
Persistent stupidity At least three teams either over the course of the season or last week alone caught our attention as being heavily penalized: DET, DAL and SD. On TV, it’s easy to see how these penalties prevent teams from achieving success in the win/loss column. Take for instance the Lions, who stopped the Giants on 3rd and goal, only to receive an unsportsmanlike penalty at the end of the play giving the Giants at least three more shots at the endzone. Three plays later, the Giants scored. For the Cowboy, Miles Austin torched the Minnesota secondary, but only by committing offensive pass interference for which he was flagged. The Cowboys ended up punting.
Where do the effects of these penalties over the course of a game or the duration of a season manifest themselves in our rankings? Indirectly, persistent sloppiness shows up in the categories of scoring offense and defense. For Dallas, who has been plagued by penalties since week one when a penalty on the last and potentially game-winning play of the game cost them victory against the division rival Redskins, this is painfully obvious. The team’s unweighted average category rank is 11 when scoring offense and defense are omitted and 15 when they are not.
Detroit appears to be sloppy on the defensive side of the ball, which detracts from the big numbers put up by Calvin Johnson and the rest of the team’s offense. For San Diego, poor discipline on both sides of the ball has begun to chip away at their meteoric rise in our rankings as documented last week. San Diego slipped from 7th to 12th after this past weekend when their comeback attempt against the resurgent Rams fell short.
A look back: Chiefs/Texans One of the more intriguing matchups of last week for us was the Texans and Chiefs. The game lived up to excitement expectations with the Texans furiously fighting back in the fourth quarter to take it in the end. What made this game particularly interesting to M-P was the direction each team was heading in entering play. From a rankings perspective, HOU needed to play well to stave off their impressive M-P plunge and QB Schaub’s huge day helped the Texans jump from 21st to 18th overall. The evenly played game (the Chiefs outgained the Texans by 4 yards and had the ball for 6 more minutes) and QB Cassel’s similarly impressive performance benefitted KC from our point of view as well.
A look ahead: Giants/Cowboys A Monday night matchup of top 10 M-P teams grabs the headlines this week, albeit top 10 teams heading in different directions at different rates the last two weeks. DAL may be (1-5) and occupy the bottom spot in the NFC East, but the Cowboys impress M-P by ranking fourth in both offensive and defensive play success (the first and third most important categories) and sixth in passing offense (fourth most important). The Giants counter well by ranking ninth and second in the same two play success categories, respectively. NYG’s sixth ranked rushing offense versus DAL’s 17th ranked rushing defense and DAL’s sixth ranked passing offense versus NYG’s 15th ranked passing defense will be key matchups for each team to exploit in the march to victory.