This interactive dashboard explores the relationship between two separate NFL markets: the daily fantasy sports (DFS) and point spread betting (spread) markets. It is updated weekly during the NFL season and the data currently goes back to the beginning of the 2021 NFL season. The first update for the week occurs on Wednesday at 12:10pm eastern time.

Motivation

The DFS and spread markets both take in money from market participants betting on NFL events (i.e. player performance and game outcomes respectively) and both market makers spend significant effort in modeling and predicting the same in order to make their offerings profitable. Vegas spread lines are often cited as being very accurate predictions of game score differences. Given player performance and game outcomes are logically correlated, there may be valuable insights to be had by studying the correlation of these markets. This dashboard lets you view the overall correlation derived from historical data as well as the details pertaining to the current week games.

Markets Overview

DFS

In DFS, market participants are given a salary cap which is used to build a team of NFL players who are assigned salaries by the market maker (e.g. DraftKings). Market participant teams score points based on the individual players’ statistical performance in the week’s games, and their competition is other market participant teams. Put simply, it is fantasy football played at a weekly interval, with salary caps limiting which players you can draft on your team each week.

Spread

In the spread market, market participants bet on a range of outcomes occurring in NFL games, where the range offered on each game is fixed by the market maker via the point spread. For example, in a simple, non spread wager between New England Patriots vs The Chicago Bears, someone could bet on New England to win the game, which covers any outcome where New England wins (scores more points than Chicago). In a spread market, if the market maker makes the spread for the game 7, where New England is the favorite and Chicago is the underdog, a participant betting on the New England side is limited to a range of outcomes starting at winning by more than 7 points but not including anything below 7. So if New England were to win by only 3 points in this scenario, a New England bet would be a losing wager. In this example, New England is -7 (giving 7 points) and Chicago is +7 (getting 7 points), and 7 is the spread.

Analysis Overview

The goal of this analysis is to compare how the spread market maker perceives the strength of a team relative to its opponent to how the DFS market does in order to see if there are any discrepancies that can be exploited in one of the markets.

Extracting the Information from the Markets

In the spread market, it is easy to understand the evaluation of the market for each team: the higher the spread number is on the favorite, the stronger the team is relative to its opponent. For example, a -7 favorite is stronger relative to its opponent than a -3 favorite to its opponent.

To get a similar evaluation from the DFS market, an assumption needs to be made: In the DFS market, strong teams relative to their opponent will have higher player salaries than their opponents’ players, and the salary gap should be proportional to how much stronger the team is. The first analysis task given this assumption is to group DFS players by their teams. Given the nature of the DFS salary pool and football in general, the groupings in this dashboard are done as follows for each team:

• Highest salary QB
• Defense/Special Teams salary
• Next 7 highest salary players (not QB)
• Players on the injury report as Doubtful or Out are discluded

Next, for each NFL game the player salaries of each team need to be compared. To do this, the salaries for each team are simply summed and the difference between the two teams is the salary differential. Teams with a salary sum higher than their opponents are stronger and have a positive salary differential. This salary differential is the measure of the strength of the team from the perspective of the DFS market. For example, a team with a salary differential of +$7,000 is stronger relative to its opponent than a team with a salary differential of +$2,000 is to its opponent.

Visualizing the Salary Differential

Each week this dashboard is updated to show the salary differential of each NFL match in the Weekly Salary Comparison Stacked Barchart. The 9 salaries for each team are grouped in the barchart by value: P1 being the most valuable player on the team with the highest salary and P9 being the least valuable player according to the DFS market. The stacked bars easily display what team in each match has the higher salary sum, and the magnitude of the salary differential is visualized. The chart also has interactive capabilities that allow hovering over a bar to see more information about an individual player and filtering out player groupings.

Comparing the Market Evaluations

Given the analysis done here, there are now two variables representing team strength for every team and every game, one from each market: the point spread and the salary differential. Plotting the two variables for each team on a scatter can test the assumption about the DFS market. The Spread Salary Scatter Plot does that for almost every team and every game starting from week 1 of the 2021 NFL season. Each point represents a team in a game which has an associated spread (x axis) and salary differential to their opponent (y axis) and is shaped/colored according to the betting result of a spread market participant wagering on the team (win/loss/push). Looking at the graph we can see a fairly strong negative linear correlation, which is expected. That is, as the spread value (X) becomes more negative (higher favored team) the salary differential value (Y) becomes more positive. To put in simple terms, a highly favored team will have players that are expected to perform better statistically than the underdog team. A best fit line is drawn which acts as an expected differential line for any spread. Given this analysis base, we can now group games based on their spread and DFS salary differential characteristics and look for trends.

The Spread Salary Inverse System

The spread salary inverse system was one of the first indicators I created after this analysis. The definition: a team is an underdog (i.e. getting points in a spread market and therefore on the right side of the y-axis), AND has a positive salary differential in DFS markets (i.e. has players that are expected to perform statistically better and therefore above the x-axis). This is opposite of what we expect given our analysis. These points are easy to see on the The Spread Salary Scatter Plot, as they are in the first quadrant (top right) surrounded by golden axis lines. In the dashboard you can filter out all games that were spread salary inverses by using the dropdown button. As of this writing (12/20/2022), spread salary inverse teams are 57-34-1 against the spread, or 62% winners.

Comparing two markets can be a powerful and interesting thing. There is much more one can do with this data, this is just a start.

Disclaimer

The accuracy of this data (and all data on this website) is not guaranteed. Data collected, analyzed, and shown anywhere on this site can have errors. All data and analysis shown on this site are not intended to be used toward any financial decisions.

Hi,
Thanks for checking out my site! If this information is valuable to you, please consider a tip to help cover my costs in creating it and hosting it. Any amount is appreciated!

You can make a tip on Twitter (Venmo, Ethereum, Bitcoin, etc.):

@LukeHeide1

You can tip on Paypal:

You can also support by following me on twitter:
Main Account – @LukeHeide1
For stocks – @grindStocks
For sports – @grindSunday