Dave Caban explains the process used to generate the weekly projections that will power the apps in the 2019 season.

Throughout the 2019 season, I’ll be generating Game Level Similarity Projections (GLSP) on a weekly basis. GLSP use historical data to determine a realistic range of outcomes for a particular player in a particular game. These projections will be included in a number of our apps this season. With this in mind, let’s take a quick look at how these projections are built.

Saquon Barkley Vs. Dallas

Saquon Barkley will line up opposite the Dallas Cowboys to open the 2019 season. Barkley is an absolute monster and in an average game, I’d expect him to rush for somewhere near 85 yards while adding four to five receptions and half a touchdown. Given this line, Barkley should score approximately 22 fantasy points in PPR leagues.

When creating that stat line, I didn’t really consider the quality or attributes of the Cowboy’s defense. Doing so would likely allow me to refine this projection and help me to get a better sense of his median outcome. As a rookie, he faced Dallas twice scoring 25 and 24 fantasy points. Perhaps my quick and dirty projection above wasn’t bullish enough. Of course, it’s hard to draw conclusions from a two-game sample.

While this baseline expectation is somewhat useful, we oftentimes find ourselves interested in a player’s weekly range of outcomes. If Barkley were to really go bananas, how many points are we talking? If he struggles, how poor of a game could he have? While we only have data from two games in 2018 to work with, we can modify our search to create a larger sample. To do this, we can search for players similar to Barkley, and review their performances when facing defenses similar to the Cowboys. This is the goal of the GLSP process.

The GLSP Process

A ton of calculations are made in the GLSP process but overall it’s actually pretty simple. Let’s take it a chunk at a time:

Step 1 – Finding Matches for Player In Question

• On a positional basis, determine the statistics that are most heavily correlated with fantasy scoring as well as the statistics that are most likely to carry from week to week
  • Weight the correlated statistics in a way the prioritizes those that are heavily correlated with fantasy points and are likely to carry from week to week.
• Create an average stat line of the last eight games played for the player in question.
  • Review the last eight games played and remove/replace any in which the player’s volume (passing attempts, rushing attempts, or targets, depending upon position) are significantly lower than average.
    • For example, a game in which a quarterback who generally throws more than 20 passes per game records just eight will be removed and replaced with an older game in which the threshold was met.
      • Only games within the last 20 regular season weeks are considered.
• Compare the difference between the average stat line generated via the above for the player in question with the end-of-season stat line created by every player from the same position throughout the past five seasons.
  • Use weightings to determine an aggregate difference and then rank these differences.

Step 2 – Finding Matches for Opposing Defense

• Using the statistics identified in Step 1, compare the average stat line of the defense being opposed, from its eight most recent games, against the season-ending stat lines of every defense in the last five seasons.
  • Use weightings to determine an aggregate difference and then rank these differences.

Step 3 – Combine Steps 1 and Step 2

• Having identified players that are similar to Barkley and defenses similar to the Cowboys, search for games in which players similar to Barkley opposed defenses similar to the Cowboys.

Step 4 – Review Player Performance in Matching Games

• Compile the 50 best matching games and review the results for those in which the matching player did not meet required opportunity thresholds.
  • Remove such players from the analysis.
•   Review and compile the fantasy points produced by the matching players in the matching games and gather a range of outcomes.

Barkley’s Week 1 Range of Outcomes