I know I picked Atlanta as my upset for the previous week, but I think they will lose to Philadelphia this coming Saturday, even though the Falcons are favored by 3 points (odds as of 01/10 at ~7:30pm). In some ways the Falcons-Eagles game is similar to the Falcons-Rams game. You have one team that typically scores a lot of points on average (Rams/Eagles) against a team that scores fewer points (Falcons) but has a similar defense (graph below). [I think the dip in the offensive output by the Eagles towards the end of the season has to do with them resting a bunch of their starters against the Cowboys during their last game].
The average score curves for both teams look similar to the previous Falcons/Rams game with the exception that the Eagles defense seems to be slightly better then the Falcons defense in average points allowed, and both team’s defenses seem to be getting better as the season progresses.
Unlike the Rams, the Eagles offense tends to play the same either home or away (disregarding their last game against Dallas), but their defense typically allows their opponent to score 3 fewer points then expected when playing in Philadelphia based on their overall allowed points average for the season (below graph). The Falcons offense, on the other hand, tends to play worse away than they do at home by about -2 points.
Using a t-test to determine whether there are any differences in either a team’s offense or defense performance (based on a team’s actual points allowed or scored compared to what they should have scored based on who they were playing for that week), I find a significant (p=0.02) home effect for the Eagles defense (below graph). All the other p-values looking at performance differences for the Falcons defense and both the Eagles and Falcons offense are above 0.20, so I would expect that the Eagles defense would be the variable that has the greatest location effect.
As I mentioned in my previous blog post, I think the amount of turnovers and sacks a team gives up/takes away plays a huge role in the final outcome of the game. I think these two variables are tied together (the amount of pressure one puts on a quarterback in the form of sacks can lead to more interceptions), but as a whole, I think the turnover ratio plays the largest effect in the final score. When you look at the average number of turnovers both teams give up/take away, you can see distinct differences in both the offenses and defenses of both teams (below graph). The Eagles defense has been particularly good at getting turnovers at home, averaging about 2-turnovers/home game at the end of the season. The Falcons defense is averaging about 1.3-turnovers/away game and this number has been pretty consistent for their last 6 games.
There is a clear difference in the number of turnovers the Falcons and Eagles offense gives up on average regardless of whether they’re playing at home or away. I think what makes the Falcons a great team is the fact that their offense hasn’t give up the football very much lately (they had zero turnovers compared to the two lost fumbles for the Rams in their last game). The Eagles offense at home is averaging around 1.4 turnovers per home game at the end of the season, and there appears to be an increasing trend in this average at home after week 13 (around the same time they lost their starting MVP candidate quarterback, Carson Wentz). Will the fact that the Eagles defense gets more turnovers at home make up for the fact that their offense typically gives up more turnovers at home? I think whichever team has the best turnover ratio (take vs. give) will win the game.
I don’t see too many differences in the home and away averages in sacks for both the offense and the defense for both teams (above graph). The Falcons offense might give up fewer sacks at home, but since the game is in Philadelphia, this fact shouldn’t matter. As for season totals regardless of location, the Falcons offense allows 1.6 sacks on average per game (1.1 at home and 2.0 away), while the Eagles offense allows on average 2.2 sacks per game both home and away. I predict the amount of sacks for each team will be around 2 and shouldn’t be a factor in the game as far as who wins or loses.
Based on my calculations, I would expect the Eagles to score somewhere close to 21.4 points and the Falcons to score close to 16.6 points this Saturday. I realize that an NFL team can’t score a fraction of a point, but I use it in my final calculation to emphasize that I’m taking the middle value out of a distribution of numbers and the final number is an approximation of this distribution.
Just like any game of chance, there are many variables that can affect the final outcome of the game (injuries, weather, key penalties, etc…) that I cannot predict with my system. Given everything equal, I would say the Eagles should be able to beat the Falcons at home based on the strength of their defense playing better in Philadelphia.
-Michael Edwards, Bioinfo Solutions LLC
* Microsoft Excel for Mac 2011 and JMP Pro ver.13 were used to collect, analyze and display NFL data.