Daily Fantasy NFL: The Amorphous Patriots Offense
This week the Patriots face the Buffalo Bills. In the the three NFL seasons I’ve played daily fantasy, one thing has been clear about the Patriots: They are hard to predict. You don’t know if they are gonna throw 60 times in a game or run 40 times. They adjust their game plan based on their opponent.
Against the Bills earlier this season, the Patriots threw the ball 59 times, while running backs only ran the ball 10 times. Against the Jets, running backs only ran the ball 5 times while passing 54 times. But then against the Redskins, they handed the ball off 34 times compared to only 39 passes. The theme is obvious. The Patriots will run vastly more against bad run defenses, but exclusively pass against the very good ones.
I wanted to check if the Patriots were in fact one of the most amorphous offenses in the league. In concrete terms, did the Patriots have the highest standard deviation of pass/run percentage in the league? Did they change their play style a lot week to week?
I used 2014 data to look at this, because I have some clean 2014 data that makes this easy to research. While this may not be too useful for this season, I figured it would be interesting to look at to see what kind of teams had a ton of variability.
As I’ve wrote about on this site before, scoring margin and pass percentage have in inverse relationship. A team’s pass percentage goes up the greater the margin of defeat, and goes down the greater the margin of victory. Teams who have had large margin of victories and margins of defeat are likely to have the most variability, so I adjusted pass percentage based on margin of victory, using the formula we use in our NFL Sportsbook Projections. An example of how this works: A team that wins by 21 points and throws 70% of the time is much more pass inclined than a team that loses by 14 points and throws the ball 71% of the time, despite the fact the latter team threw the ball more. Adjusting for scoring margin makes this clear.
Below are the teams that have the largest pass variability. PassVariability is the standard deviation of margin adjusted pass percentage. The Bills were the largest in 2014, with an 11.4% standard deviation in margin adjusted pass percentage. Included is a box plot graph that shows distributions of teams pass inclinations throughout the 2014 season.
The Bills lead here because of a QB change in the 2014 season. They started off with EJ Manuel, who they handled with care, but let Kyle Orton loose when Manuel was benched. Chicago is another anomaly. They had so many large blowouts that they didn’t even bother to pass to catch up, which our formula did not handle well. Denver started rushing the ball much more later in the season after airing it out with Peyton Manning early on. The Jets are interesting. They are the first team we see who drastically adjusted their play style game to game. They would not stop running the ball against the Miami Dolphins late in the 2014, despite eventually losing the game. They ended up passing the ball only 20% of the time, by far the lowest total for any team in the season. The Jets are high on the list because of their penchant to load up on running the ball when their running game has success.
The Patriots show up in 5th, but they only are behind Buffalo by a hair. You could call the Patriots the passing version of the Jets. They had the top 2 most pass heavy games of the 2014 season. One was against the Lions, where despite winning the game by 25 points and getting out to an immediate lead, the Patriots had a 72% passing percentage, a 53/20 pass to run ratio. This was the largest margin adjusted pass percentage in the league. The other was the divisional playoff game against the Ravens, where Brady threw the ball 50 times while running backs only had 7 carries for a near 80% pass percentage. The Patriots rarely ran the ball very often in a game, unless they had a huge lead against a bad run defense. However, even in the most run heavy Patriots games, they did not run significantly more than you would expect given their margin of victory.
So in one sense I was wrong, the Patriots weren’t the most amorphous offense. But they did often employ a essentially pass only strategy that made them an interesting team to target in certain matchups. However, they weren’t the only malleable offense in the league last season. The Jets were the inverse Patriots. When their running game worked, that’s all they did, even if they didn’t have a big lead. The Broncos changed throughout the season, they moved from a pass first offense to a run first offense with the emergence of C.J. Anderson.
What does this mean for 2015 NFL daily fantasy and for Week 11? A few things. First, despite the Patriots going pass heavy in game one against the Bills, I’m not sure this will happen again. The Bills have been a pretty average run defense this year, with some analytics such as DVOA showing they have been one of the worst run defenses in the league this season. In the first matchup, Blount was coming off a one game suspension and wasn’t quite integrated into the offense yet. I would bet against the Patriots abandoning the run this game unless they fall behind.
Teams with young or poor quarterbacks and good run games like the Jets could also be the kind of team that will run the ball relentlessly if they are running with success. The Vikings come to mind for 2015, but any team starting a bad QB with a good run game is a threat to pound the ball.
The data also shows us that teams can change their style as the season goes on. St. Louis is a great example. They were a balanced team in the beginning of the season, but with the emergence of Todd Gurley they have been the most rush heavy team in the league. Tampa Bay is another one. They were a run first team early on with Jameis Winston, but as Winston’s play has improved, they have been passing it a lot more.
Teams can often have their play styles set in stones, but when teams adapt week to week, opportunity can arise.View all posts by Daniel Steinberg