DraftKings Millionaire Maker: Why Correlations are the Key to Winning a Million Dollars
UPDATE 8/26/2019: This article was written in 2014. During this time, there was no such thing as a lineup builder. Projections weren’t publicly available. You had to build lineups by hand and by feel.
We’re in a much different landscape now, where most players use lineup builders and projections to create the best lineups for tournaments.
The content of this article is still very much true, maximizing for correlations is the key to winning a million dollars on DraftKings. But now, we don’t have to figure out how to maximize for correlations ourselves. Sabersim.com has a lineup builder that does it for us, using actual game simulations to calculate both correlations and upside and builds lineups maximizing for those two essential stats. Right now, Sabersim is offering a 3-day FREE trial for anyone who signs up. Just go to Sabersim.com and sign up. You won’t regret it.
And if you’re interested in learning more, check out this video explaining how the Sabersim lineup builder works:
Roughly 2 weeks from now, DraftKings will run their highly anticipated Millionaire Maker Fantasy Football tournament. The Millionaire Maker is a $27 to enter, $2.2 million dollar fantasy tournament with a 1st prize of one million dollars, one of the biggest prizes ever awarded in a fantasy tournament, and certainly the biggest prize in a tournament with a buyin so small. I will personally be putting in over 100 different lineups in the Millionaire Maker in an effort to win that top prize.
Usually, DraftKings payout structures are pretty flat, paying out 20-25% of the prize pool and awarding only about 10% of the total pool to 1st place. DraftKings Millionaire Maker, however, has a wildly different structure, here’s a snapshot:
As you can see, while 1st place is $1,000,000 as advertised, 2nd prize is only $100,000, and 3rd is only $50,000.This discrepancy is totally ridiculous, but clearly this was the only way DraftKings could support a $27 tournament with such a high 1st prize. While it’s fun to sit there and simply laugh at this massive difference, this factor is something to think about as it completely changes how you should set your lineup in the Millionaire Maker. Since almost half the prize pool is given to 1st place, we need to maximize our chances of getting 1st at the expense of of placing in the middle. But the question is, how do we do this? One way to help our cause is using correlations.
A Lineup Full of Correlations, Without Reverse Correlations
In football, there are correlations between players and events throughout a game. A simple example of this is a quarterback and a wide receiver: If a receiver catches a ball, it means the quarterback threw him the ball, therefore a QB and WR will accumulate points together. This is the strongest correlation in fantasy football. There are also reverse correlations, where a players performance negatively affects other players performances. A simple example of a reverse correlation is defenses and players on the opposing team. If a defense does well, it means the players on the opposite team are doing poorly. Simple right? But some correlations and reverse correlations are not as obvious. Here are a few:
a) Quarterback and WR. I explained this in the above paragraph. You absolutely should be pairing your QB pick with a WR on his team.
b) Defense and Running Back. In general, if a running back’s defense is doing well, he should get more oppourtunities since teams tend to run more often when they have a big lead. He should also get better field position which will lead to TD oppourtunities.
c) Receivers on opposing teams. When one team is scoring a lot of points, the other has to play catch-up, which almost always means more passing, shorter drives, and higher scoring games. Using players in the same game, but on opposite teams, can ensure that you’ll either have 2 duds at WR, or 2 fantastic games, which is good because we do not want any mediocre performances.
a) Defense and opposing RB. This one is quite obvious as well, you absolutely do not want to use any player from a team your defense is facing, since his production will take away from your defense’s points and vice versa. For RB, it’s even worse because teams that haven’t scored usually will abandon the running game.
b) Skill players on the same team (WR/RB/TE). If a WR, RB or TE touches the ball, it means that a WR, RB, or TE is NOT touching the ball. While sometimes there’s enough offense to go around and two players can have big games on the same team, it’s unlikely. It’s possible someone will win the Millionaire Maker using two skill players from the same team, but to maximize your chances, you should not be doing this.
c) RBs on opposing teams. You almost never see two RBs in the same game have great games and it’s for several reasons. Running kills the clock which gives the opposing team less oppourtunies and as I stated before RB performance is correlated with big leads. The only exception to this is RBs like Reggie Bush and Giovani Bernard, who are mostly pass catchers, since being down actually helps these RBs get more touches and better opportunities.
d) D/ST and QB. If a D/ST is doing well (which hopefully means some pick-6’s, fumble returns for a TD, or special teams TDs), the Quarterback is going to get less opportunities on offense and there will be less of a need to throw.
In general, we want to construct each of our teams to have as many correlations as possible and avoid reverse correlations entirely. For an example of this, I’ll construct a lineup for the week 4 Sunday Million on DraftKings (since the salaries are not yet out for the Millionaire Maker):
This lineup has the makings of a great lineup, we have a QB/WR combo in Johnson and Stafford who both have very high upside, and we have a WR on the opposing team in Kerley, who’s a deep play threat. But there are a few reverse correlations on this team. A good performance from Khiry Robinson will take away from Graham’s TD and catch opportunities. And Golden Tate almost certainly will not have a great game if Calvin Johnson lights it up like we need him to. By making a few slight changes, we can make a lineup with a much better chance of winning.
We replace Jimmy Graham with Zach Ertz, who doesn’t have the upside of Graham but is capable of a big game. This gives us more money to spend and we use it wisely pairing Lamar Miller with the Dolphins D, two pieces that are highly correlated. We also replace Tate with Crabtree, and now we have two high upside receivers from the same game but on opposite teams. Our lineup is littered with correlations and we’re primed for a boom or bust day on DraftKings.
View all posts by Max J Steinberg