nope

DraftKings Millionaire Maker: Why Going Against Popular Picks Gives You An Advantage

DraftKings has an awesome tournament a week and a half from now they are calling the Millionaire Maker, which will pay out $1,000,000 for first place with only a $27 entry fee. The tournament is sure to get a large overlay as DraftKings NFL guaranteed prize pool tournaments have routinely failed to get full entrants this NFL season.

The tournament is also extraordinarily top heavy, as nearly 50% of the 2.2 million dollar prize pool goes to first place. With such a win or bust structure, it’s important that you understand the mathematical concepts that make it possible for a lineup to have a high enough of a ceiling to beat tens of thousands of entrants. Max wrote a great article a few days ago outlining how correlations play a big part in lineup upside. Today I will write about how and why going against popular picks gives you a larger chance of winning.

Players Will Always Have Bad Games

It’s nice to think of our prediction skills as so good that everyone we have a strong feeling about will have a great game. But the reality is prediction is not clairvoyance. There will always be luck related events that cause a good pick or great pick to have a bad game. This is especially true in football, where games are so short and injuries are so prevalent that truly anything can happen on any given week. Two weeks ago, the highest picked receiver A.J. Green got a toe injury in the first quarter and left the game, leaving a zero for everyone who had him in their lineups. Even in times where a player doesn’t get injured, the defense can play well or he could not get a lot of looks or a great player will have a bad game for seemingly no reason at all.

Player Usage Percentage

In any given daily fantasy contest, regardless of sport, there are going to be some obvious great picks that a lot of people use in their lineups. Last week, those were guys like Drew Brees and Calvin Johnson who had great matchups.

You can see player usage percentage in the lobby of any contest, assuming it isn’t a head to head contest. On FanDuel, you can see everyone’s usage % as soon as a contest starts. On DraftKings, lineups aren’t revealed unless a player in the lineup has a game that has started. Often in large GPPs you will see players used as highly as in 50% of lineups, in smaller tournaments you can see as high as 80% or 90% usage.

Why Going Against High Picked Players Is Advantageous In Top Heavy Payout Structures

To understand why not picking popular picks for your lineup gives you an advantage, we will look at a toy game that will illuminate the issue more simply.

Let’s say we are playing in a $10 daily fantasy football contest with 10 players total, and the top 2 entries will be paid out, with first paying out $70 and second paying out $30. You only get to pick one player, and it must be a QB. That will be your whole team. Let’s say Drew Brees is such an overwhelmingly good pick, everyone else in the tournament will choose him. You decide to choose Aaron Rodgers.

Let’s say Aaron Rodgers is a much worse pick than Brees, he will only do better than Brees 35% of the time. With these numbers, we can find out the EV of each lineup.

Rodgers Lineup EV: (.35*$70)+(,65*0)= $24.5 – $10 = $14.5 Profit.

Brees Lineup EV: (.35*$3.33)+(.65*$11.11) = $8.39 – $10 = ($1.61) Loss

Wait, how is that possible? Brees was the better pick! Who cares how much a player is used by other entrants? I just want to pick the best player!

The issue is that by picking against the grain, you essentially are competing against less entrants. In this toy game, you were really only competing with one lineup. If Brees did badly, you would win everything, while if Brees did well, you would lose. You leveraged your chance of winning with one event occurring. Because that event occurred so often (having a 35% chance to win a top heavy payout tournament is gigantic), you turned an incredible profit, an 145% ROI.

Okay, yes, a lineup consists of much more than one player. And it’s rare that a player is picked over 50% of the time. But when we are talking about a tournament where nearly 50% of the prize pool is paid out for first place,  to make a positive EV lineup you need to mostly care about getting first, so this factor can really play a big part in a lineup’s profitability.

Predicting Used Percentages

Probably the best way to figure out used percentages is to play Thursday games on FanDuel and check out the used percentages in the tournaments you joined. The highest used players in Thursday games are very likely to be the highest used players on the Sunday contests on either FanDuel or DraftKings, assuming the Thursday NFL game was not popular in lineup choices, and assuming that there isn’t a huge salary discrepancy between the two sites.

Another way to predict used percentages is to look around the internet for daily fantasy football advice. Guys you see picked by multiple advice columnists should be highly used players. As you get more experienced, you will be able to predict player usage pretty well yourself, you just have to get a gauge on how the market functions.

Leveraging Further

Because player performance is often correlated or reverse correlated to each other across a team or game, you can further go against the grain by picking players that will do well when the player you’re targeting against will do badly.

A good example would be a #2 WR or a TE going against a popular #1 WR. That would be like going with Golden Tate when Calvin Johnson will be an extremely popular play.  Choosing an opposing teams RB against a popular RB play can also be useful. RBs have fantasy production largely tied to their teams ability to have a lead in the game. Teams that are down by 10 points or more will often ditch the running game completely. By picking an opposing teams RB, you are leveraging an event where the opposing team wins the game by a large margin, and therefore will make it extremely unlikely the popular RB will have a good game with his team trailing so large. For more, look to Max’s article linked earlier as well as my previous article on correlations.

Summary

  • DraftKings has a tournament in a week and a half where $1,000,000 will be awarded to first place.
  • Player performance are not as predictable as we would hope, and great daily fantasy picks will often have bad games.
  • In GPPs and smaller tournaments, popular picks can be used in well over 50% of lineups.
  • We can mathematically show through toy games that picking against popular picks increases profitability in top heavy payout tournaments.
  • You can help predict used percentages for Sunday games by playing Thursday contests. You will also be able to make prediction yourself by looking at advice columns and using your own market instincts.
  • Using players who will have performances with inverse correlations to popular picks will further leverage your lineups performance.

 

View all posts by Daniel Steinberg
Daniel Steinberg

About the Author

Daniel Steinberg Daniel Steinberg is a former bond trader at a multi-billion dollar proprietary trading firm in Chicago. He uses his knowledge of statistics and his creativity from his career as a poker professional to create the most advanced Daily Fantasy statistical analysis that you will find anywhere. Follow him on twitter @DanielSingerS

no banner found

Daily fantasy sites