Discussion Board


Take the Eagles and the Points!

By Arod and BA Marks

It is time for the 2018 Super Bowl LII Prediction. 

The Wisdom of the Crowds: Patriots by 12.  Oddsharks has the Patriots winning 29.3 to 16.9 and a Vegas spread of 4.5; here are a bunch of CBS cognoscenti.

And here is ours.

Arod and Brian: This is painful…Eagles by 3!!!  The score 29 – 26.  Did we really say Eagles by 3?   Eagles by a field goal.

Our time-tested model told us Patriots over Falcons last year.  We take that same model, but we need to call in the adjustment bureau this time around.

First, we account for offensive prowess by looking at “season-average points per game.”  We make some simple adjustments to the average points scored in the 2017 NFL season.  We see each team outperformed the league average.  Can you say, “no surprise.”  The Patriots and Eagles each out-scored the average by 32%. 

Second, we account for defensive weakness of each team.  In percent terms, both the Eagles and the Patriots perform better than the league average:  -15%.  In other words, the Patriots had 15% less points scored against them than the league average; the same for the Eagles. 

The similarity between the two teams is remarkable.

Now, the adjustment bureau will perform some additional magic (after hours of debate).

The average points for the Patriots is 21.7.  For the Eagles, here is the kicker (or should we say quarterback), we adjust the average by 3 points, a field goal, to 24.8.  Say, what?  Yes, a 3-point adjustment and, yes, we know Nick Foles is no Tom Brady!  Who is Brady, but possibly the best quarterback of all time.  Eagle quarterback Carson Wentz went down in game 13.  After much debate, we adjust the Eagles average points per game to reflect the Eagles’ last 6 games  - when they were sans Wentz (instead of the season average) and derive 24.8 points per game. 

According to Martin Frank of the Delaware New Journal, Foles enters the season finale on a two-game hot streak in which he completed 78% (49 of 63) passes, averaged 9.5 years per attempt, and tossed three touchdowns with no interceptions.

For the quants . . . we use the following formula: 

Expected Number of Points = (League Average Points per Game) * (Offensive Prowess) * (Defensive Weakness)

 The following table shows the numbers:


Average Points

Offensive Prowess

Defensive Weakness

Expected Points












Our prediction, after rounding off, we predict a Spread equal to 3 points. 

And for the quants of the quants, here is a graph of the distribution of possible scores:


Here is the distribution of the spread:


In other words, the spread could reasonably range from -9 to +9, but centered on 3.3.  The distribution of the spread is obtained from randomizing the expected score for each team by drawing from a normal distribution with each team’s expected scored (Eagles 29.3, Patriots 26.2) and the mean and the league standard deviation in points per game scored (4.08).

Bottom Line:  Eagles by 3.

For the skeptics, who are pondering the validity, simplicity of the model, we remind you that the same model accurately predicted last year’s upset when the Patriots beat the Falcons.  And, one other thing may bode well for the Eagles, a sign, if you will:  we are poised to consume approximately 1.5 billion chicken wings.  So, maybe it is time for a bird to win the Super Bowl.    

Oh, and for you Patriot fans, as every economist knows, you cannot lose:  Park your emotion at the gate.  Take the Eagles and the Points too! 

Enjoy the Super Bowl

Peace and Cheers

Arod and Brian



E-mail me when people leave their comments –

You need to be a member of UNH Economics Collective to add comments!

Join UNH Economics Collective