As the fantasy football regular season wraps up this weekend, our attention turns to NFL Playoff Best Ball Tournaments. Underdog has launched several playoff tournaments, including The Guantlet which has a $100,000 first-place prize.
After fading the 2021 first-round pick Jonathan Taylor in 2021 playoff best ball drafts – who was the biggest bust after the Colts’ epic collapse and elimination from playoff contention – Michael Dubner provides his 2022 Playoff Best Ball Model.
Maximizing Profit: Projections and Contingency-Based Drafting
Before we get to the model, we first need to discuss playoff best ball game theory.
Just like in DFS, there are two primary levers that drafters can pull to improve their chances of taking down first-place in a playoff best ball tournament: Projection and Game Theory. The secret to fantasy sports and winning life-changing money is the realization that we don’t need the highest score possible. We really just need the highest score compared to our opponents. Because nearly all of the money is won in the Super Bowl, we need to maximize our chances of having the highest score in the Super Bowl. In order to have a shot at first-place, Playoff Best Ball rosters need to be constructed around fielding a full starting lineup in the Super Bowl. Drafting four players on one NFC team along with three players on two separate AFC teams is a good place to start. Correlation is the most important factor in playoff best ball.
Ultimately, in Playoff Best Ball Tournaments, we need to lean into contingency-based drafting. Here are some of the many things we need to consider:
- Stack teams to build a Super Bowl matchup.
- If we stack a team with a first-round bye, then we may need to draft the studs on teams with a lower chance of advancing but that can help us survive the Wild Card round.
- There needs to be some balance of having the best team in the Super Bowl but having a team that is strong enough to get there in the first-place. This is especially true if we stack a team with a first-round bye.
- If we draft a highly owned stack, then consider mixing in a few lower owned players that can differentiate our rosters.
- Either by drafting players that largely go undrafted (e.g. a player with an ADP > 55), or by reaching earlier for a player in order to build a unique combination (e.g. drafting a 7th round ADP player in the 5th round).
- Because ADP will push up players on teams more likely to make a Super Bowl run, it’s really hard to build full team stacks from the best teams (e.g. Chiefs). Meanwhile, you can build full team stacks of all the studs from teams with lower Super Bowl odds (e.g. Buccaneers).
As you can see, contingency-based drafting is much more important than simple points per game projections. The extremely binary outcome of players either advancing or being eliminated makes game theory and optimizing for the Super Bowl the key element to maximizing profit in playoff best ball.
Dubner’s Playoff Best Ball Model
With that being said, projections still matter and can help guide our rankings and identify unique angles. Instead of just simply ranking players based on points per game over position replacement, we can use Sportsbook odds to project how many games each NFL team is projected to play. This will help us balance the quality of the player (e.g. player points per game projection) with the quality of the team (e.g. how many games they are expected to play). We can identify ADP inefficiencies where the market may be over- or under- confident in entire teams. And we may find late round sleepers that the market undervalues due to their low points per game projection but the Model likes because of their odds of playing a lot of games.
Projected Games Played
2022 Playoff Best Ball: Projected Games Played
@Dubner_Michael
| Conf | Team | Odds to Play in Super Bowl | Proj. Games Played |
|---|---|---|---|
| AFC | BUF | 39.5% | 2.39 |
| NFC | PHI | 36.6% | 2.17 |
| NFC | SF | 32.1% | 2.22 |
| AFC | KC | 30.6% | 2.32 |
| AFC | CIN | 19.7% | 2.19 |
| NFC | DAL | 18.0% | 2.11 |
| NFC | MIN | 9.6% | 2.39 |
| AFC | LAC | 9.0% | 1.57 |
| NFC | TB | 8.9% | 1.20 |
| AFC | BAL | 7.6% | 1.71 |
| AFC | MIA | 4.9% | 0.98 |
| AFC | JAX | 4.5% | 1.14 |
| NFC | GB | 3.9% | 0.45 |
| NFC | NYG | 2.7% | 1.32 |
| NFC | WAS | 2.4% | 0.54 |
| NFC | DET | 1.9% | 0.31 |
| NFC | CAR | 1.3% | 0.33 |
| NFC | SEA | 1.3% | 0.31 |
| AFC | NYJ | 1.2% | 0.25 |
| AFC | TEN | 1.2% | 0.50 |
| AFC | NE | 1.0% | 0.22 |
| NFC | NO | 0.7% | 0.06 |
| AFC | PIT | 0.3% | 0.07 |
| AFC | LV | 0.2% | 0.01 |
Model uses a combination of odds from FiveThirtyEight, Pinnacle, and Playoff Status. Last Updated 12/31/2022.
Battling for the AFC No. 1 Seed
The Elite teams have a clear projected games played advantage, as they are most likely to play in the Super Bowl. But they also get dinged early from having a potential first-round bye. Before playoff seedings are finalized, stacking an elite team that is currently fighting for a first-round bye but is ultimately forced to play in the Wild Card round (particularly the Bills or Chiefs) is a major advantage. The team that earns the No. 2 seed in the AFC is likely going to be have most valuable players in Best Ball Playoff Tournaments as they have a legitimate chance to play four games.
Not All Games Played Are Created Equal
The ideal scenario is to stack a team that plays in all four-games, as this gives you the best chance of advancing out of the Wild Card round while having a shot at first-place in the Super Bowl. But don’t overweight the projections earned from the Wild Card round, as these are the easiest points to replace. The harder points to replace are the ones earned in the later rounds once players are eliminated. One shortcoming of the model is that it does not weight the importance of each game. One could argue the Super Bowl should be weighted 20 times that of the Wild Card round since the minimum cash of the Super Bowl is 20 times that of finishing last-place in the Divisional Round. Therefore, while the Vikings and 49ers are actually projected to play more games than the Eagles, the Eagles have a better chance of playing in the most important game. For that reason, I have included the Super Bowl odds in the above chart as well, as we need to weight that game more than the other games. But while one could simply focus on odds to play in the Super Bowl, it is also important to consider total projected games played, as we first need to advance to the Super Bowl. There is a balance of trying to draft the best Super Bowl team while also having the best chances of reaching the Super Bowl in the first place.
Any Given Sunday: Stacking Studs From Lower Seeds
We’ll see this once we get to the model rankings and look at individual players, but from a macro-lens, once you get past the top two or three teams in each Conference, the projected games played only differ by about a half-game played. If you miss out on the stacks from the best teams — or if you simply want to switch it up after your 100th draft — feel free to onslaught the best players from these lower-quality teams. Whether you stack up the Chargers, Buccaneers, or Jaguars, it doesn’t make all that much of a difference in terms of their odds of making it to the Super Bowl.
Be sure to check out RotoViz Overtime where Shawn Siegele and Colm Kelly discuss the 5 tips for winning your playoff fantasy contests.
Playoff Best Ball Rankings
The Playoff Best Ball Model is an objective way to rank players based on projected points per game and projected games played while adjusting for positional value. Using the RotoViz screener, I found that regular season Weeks 5-17 half-PPR points per game had the strongest correlation (on average) to playoff ppg. We can then use the above playoff projected games played to create a playoff cumulative projection and calculate points over position replacement.
- Underdog requires 1 starting QB, 1 starting RB, 2 starting WR/TE, and 1 Flex.
- Value is the difference between Underdog ADP and the Model Rank.
2022 Playoff Best Ball Model
@Dubner_Michael
