Tired of losing in fantasy football? Try a data-driven approach

A value-based strategy for drafting a dream team

Football season is coming up fast, and with the season comes an abundance of games for fans to play — some with the potential to win big, like fantasy football. Competition can be fierce, and the punishment for losing can be humiliating. Last year, the loser of our league had to wear a clown costume during the Super Bowl party.

In an effort to avoid such a terrible punishment (and maybe win a few bucks), we’re going to look at some advanced approaches for draft strategies. The primary metric we’ll focus on is VOR (value over replacement), which measures the value a player provides compared to an average player at that position.

There are a few other metrics we’ll include in our draft strategy, including Expert Consensus Rating, Average Draft Position, and Risk. These concepts can be tricky, so our team built an application anyone can use while drafting. Feel free to head directly to the app or continue reading to learn more about the metrics used, our methodology for calculating these metrics, and how to use this tool during your draft.

How to find maximum value in your draft

In a snake draft, every team can select one player per round. The order reverses every round, so the team to pick last in one round will pick first in the following round. The pool of players gets smaller with each round, and a team manager must determine which player provides the best value compared to the remaining pool.

The best way to approach a draft like this is to think of players like stocks. The goal is to pick players that others undervalue and maximize potential upside for bench players while minimizing risk for starters.

Brad Pitt and Jonah Hill using “sabermetrics” in Moneyball

To accurately determine a player’s value, we need to compare their projected points to the average player at the same position. There are many ways to determine what an average player is, but for our purposes, we will define an average player as “the player whose position rank is equal to the number of players at that position who are taken (on average) by pick 100”.

Once we’ve calculated the average player’s value by position, we can compare each player’s value to the average to calculate VOR. You can find a more detailed explanation of the VOR calculation and all other metrics used in the app in the documentation for the ffanalytics R package.

VOR highlights the problem of focusing only on projected points when drafting. When looking at projected points, it is easy to think that a quarterback is the most valuable player. Out of the top ten projected scorers for 2022, seven are quarterbacks:

  1. Josh Allen (QB) — 374.8
  2. Patrick Mahomes (QB) — 351.4
  3. Justin Herbert (QB) — 327.6
  4. Jalen Hurts (QB) — 320.6
  5. Dak Prescott (QB) — 316.7
  6. Lamar Jackson (QB) — 315.2
  7. Jonathan Taylor (RB) — 314.7
  8. Kyler Murray (QB) — 313.6
  9. Cooper Kupp (WR) — 310.9
  10. Christian McCaffrey (WR) — 310.4

However, when we look at the VOR rankings, there are no quarterbacks in the top ten:

  1. Jonathan Taylor (RB) — 175.6
  2. Cooper Kupp (WR) — 173.0
  3. Christian McCaffrey (RB) — 165.9
  4. Austin Ekeler (RB) — 165.0
  5. Travis Kelce (TE) — 140.2
  6. Derrick Henry (RB) — 133.6
  7. Justin Jefferson (WR) — 125.9
  8. Mark Andrews (TE) — 118.5
  9. Najee Harris (RB) — 116.0
  10. Dalvin Cook (RB) — 115.9

No quarterbacks are in the top ten by VOR because there are many high-scoring quarterbacks just below the top tier. It is easy to find a replacement for this position that can score close to the same points as the top performers. This is different for other positions, with top-tier running backs and receivers having a much larger value than the typical replacement player at each position.

The main point to understand here is that points rankings overvalue quarterbacks. Focusing on more relevant metrics like VOR helps a team manager choose players that provide the most value to their fantasy team.

There are many other factors to consider when drafting, but I’ll spare the details for another time. If you’re interested in diving deeper into any additional metrics, I encourage you to visit the Fantasy Football Analytics page for more information.

Draft advice

Let’s cut straight to the chase and talk about the best ways to approach the draft:

  • Draft your starting lineup before any bench players
  • For starters — target players with the highest VOR, a low-risk rating, and a high floor for projected points.
  • For bench players — aim for players with the highest VOR and a high ceiling to maximize the potential upside (higher risk is ok for bench players)
  • Draft Defense and Kickers last (if at all) because they are less predictable and score fewer points

How to use our draft app

Use this application to optimize your draft strategy! The most important part of this app is the data table, which contains a list of players ordered by VOR.

To properly use the draft app, follow the steps below:

  1. Set your projection settings according to desired average type (average, robust, or weighted)
  2. When a player is drafted, you can remove him by unselecting the player’s name in the Player filter.
  3. When it’s your turn, click on a player’s name in the data table to filter charts and analyze relevant metrics.

Metrics are color-coded to show the distribution of rankings on the data table. Reference lines are on the player-specific charts to show the league’s position averages. If you’d like to see a detailed analysis of a player’s scoring in specific categories, click “See Player Breakdown” after selecting a player from the table.

Player Comparisons (page 1)

  • Data Table: a detailed breakdown for each player showing all relevant metrics for selecting a player
  • Projected Points: shows the average projected points for each player, along with the minimum and maximum projected points (floor & ceiling)
  • Risk: shows our risk rating for each player based on the variance in projected points between sources.
  • Strength of Schedule: shows a position-specific strength rating for each player (1 is easiest, 32 is hardest)
  • Run vs Pass: shows the ratio of runs to passes for each team from the 2021 season.

Player Breakdown (page 2)

  • Scoring Category Percentiles: shows what percentile the selected player falls into for each scoring category
  • 2022 Outlook: contains the outlook from ESPN’s 2022 projections

In conclusion, fantasy football can be filled with uncertainty and risk, and many different strategies can help maximize value. All it takes is a bit of research and knowledge of how to use data for a competitive advantage. Hopefully, this article could shed some light on alternative metrics to focus on, highlighting draft strategies that are most effective.

We wish you all the best in your fantasy seasons! Please feel free to share this application with friends/competitors, and let us know if you have any feedback on our methodology or the performance/functionality of the application.

If you have any questions about data management, advanced analytics, or have a general interest in data — head to databenderconsulting.com for more information, or give us a follow on LinkedIn and Facebook.

Interested in working with us?

Reach out to us and we will make sure to help you achieve your business goals!