How I played like a daily fantasy sports shark for only $195 and it worked…kinda

During football season Tuesday morning is often met with two different feelings, that of vindication for making the right decisions by selecting players that scored 10 + touchdowns each[1], or that of failure, where only 72 hours earlier you were toasting yourself for building the ultimate lineup.

I’m currently more in the failure boat. So let’s start from the beginning, last week I wanted to experiment with volume entries — submitting multiple combinations of lineups into one competition. This is how pros, a.k.a., sharks play and I wanted to give it a try — but I wanted to see if it could work at a much smaller scale — pros play with thousands of dollars and up to 500 lineup variations for just one competition — I was going to play with just $195 and up to 45 lineup variations.

I decided I was going to focus on the Sunday/Monday Primetime competitions on DraftKings and FanDuel because I wanted to limit the player pool I was using to create my lineups — with the idea that if I had less players in the universe, I had a higher chance of creating a lineup that could pay off. My investment for the competitions was limited to $135 on DraftKings ($3 per entry) and $60 on FanDuel ($5 per entry).

The competition included 2 games, Patriots vs. Colts and Giants vs. Eagles for a total player pool of 82. But in reality, the player pool was more like 52 players, with the remaining players not significantly contributing over the first five weeks of the NFL season. Even by limiting the player pool, I didn’t have enough resources to randomly generate lineups and just enter them all — so I had to further limit the pool — and this is where the distinction is made between gambling and a game of skill — I had to make data driven decisions on who would be in my pool and who would not.

I created a spreadsheet and looked at the first 5 weeks of scoring for all Patriots, Colts, Eagles and Giants players. From there, I sorted by points scored in week 3 and week 5 (week 4 the Patriots had a bye). Additionally, I started to look at the opponents those players faced during a given week. Based on this info, I identified trends like while Donte Moncrief only scored 0.8 points in week 5, when Andrew Luck is playing, Moncrief scores in double digits or that Eli Manning and Tom Brady are two of the most consistent quarterbacks through week 5 of the NFL.

Based off research, I cut the pool from 52 players to 30. The breakdown was:

QB: 4 | RB: 7 | WR: 11 | TE: 3 | K: 2 | Defense: 4

With those numbers, I looked at the number of combinations that could be made from the pool. For DraftKings, I could make a total of 166,320 combinations and FanDuel a total of 332,640 combinations. Granted, those are just the raw numbers that don’t take into account salary cap — so the number reduces by a lot (I don’t have the specifics) and most competitions limit the number of lineups submissions to 200, 500 or 1,000.

With all these possibilities, and without the time and the mental capacity to come up with that many combinations that also fit the salary cap, I decided that I needed help, so I used Fantasy Cruncher. Fantasy Cruncher is a tool that optimizes lineups based on projections, it allows you to select from a pool of players and look at the combinations. For what I was doing, it was very helpful, and while I don’t put stock in projections, the tool itself did everything I wanted — specifically giving me lineup options that I could pick and choose from. Fantasy Cruncher gave me 500 possible combinations and

from there I copied and pasted them back into my excel spreadsheet and went through them and picked the best 45 for DraftKings and 12 for FanDuel. But while selecting the lineups there were questions that still loomed, like will Odell Beckham play? He is one of the top wide receivers in the league and was going up against the 2nd worse secondary in the NFL — great match up, how many points will the Patriots hate score on the Colts, will Andrew Luck look more like last season, was Andre Johnson’s outburst of points last week just an outlier?

All of these questions and more were running through my head, and while I play a lot of fantasy sports, this was by far one of the more stressful things I have done because while looking at all these variations, I was talking myself into and out of players. For instance, I talked myself out of Riley Cooper. While he was in the pool of players, I didn’t select any lineup with him in it. This proved to be a costly mistake because Cooper was the second highest scoring Eagle behind DeMarco Murray.

And so I watched the Pats vs. Colts game and felt good going into Monday night, I was breaking even with lots of players still going. I had played Blount, Edelman and Gronkowski in a good number of lineups. A costly mistake that I made was that I overplayed Dion Lewis who was in 38.10% of my entries on FanDuel and 42.86% of my entries on DraftKings.[2] Any other week and this would have been a good play, but Lewis only scored 5.4 points on FanDuel and 6.9 points on DraftKings.

So Monday night came, and Vegas had the Giants vs. Eagles pegged as one of the more high scoring games — and in the first quarter it looked like it was going to be, with Odell Beckham Jr. scoring a touchdown fairly quickly. I was happy about this because I had Beckham in 14.29% of my lineups on FanDuel and 4.29% of my DraftKings lineups.

But then, the Giants pooped the bed, and players like Shane Vereen (28.57% usage rate for my FD entries and 35% usage rate for my DK entries), who had 21 points the previous week on DraftKings, but only scored 1.6 points on DraftKings, 1.1 points on FanDuel this week, and Dwayne Harris (33.33% usage rate for my FD entries and 41.43% usage rate for my DK entries) scoring 3.8 points on DraftKings and 2.8 points on FanDuel essentially killed my entries. I picked my combinations wrong and it was a costly mistake.

Come Tuesday morning I was on the failure ship of crappiness with a -$51.94 profit on FanDuel and -$87.00 profit on DraftKings. You can see my full results and player usage percentage for DraftKings and for FanDuel.

So what does this all mean? What was the point? The point was I wanted to see if I could play like a shark but on a smaller scale by limiting my player pool and making multiple entries in a single competition.

Was it a success? If I were to judge it based off profit, then absolutely not, I finished the weekend negative. But if I were to judge it based off learning a new aspect and strategy in daily fantasy sports, I would say without hesitation, yes.

I learned that there is an advantage in playing multiple lineups (duh) but there is a threshold to where that pays off. On FanDuel, I didn’t meet that threshold, 12 lineups out of 137,931 (0.008% of total entries) wasn’t enough to make a significant difference in my chances but on DraftKings, the 45 lineups out of 38,333 (0.12% of total entries) did give me a slight advantage. And while those numbers are ridiculously small, what it really came down to was having the right players in my player pool and not overvaluing or undervaluing specific players, like what happened with Vereen and Cooper. The advantage sharks still have is that they don’t have to limit their player pool as much and have more room for mistakes.

And there is some silver lining here, think of it this way, while you may feel like you are at a disadvantage because you don’t play with thousands of dollars and submit nearly 500 entries for each competition, those sharks have a lot of bad lineups and their contributions help inflate the prize pool.

Ultimately it gets down to quantity vs. quality. Us smaller players have to focus a lot more on the quality of our lineups, while sharks have the advantage of churning out bad lineups and not being punished as badly for it.

Will I do this again? Of course. This weekend I am going to take what I learned and apply it to the Sunday night, Monday night games — and one thing I know for sure is that I am avoiding Eagles players like the plague — they are too unpredictable, so good luck.


[1] not really that many, but we can wish

[2] I know this because I use huddlehive and it breaks down usage percentage for my entries. Try it out now at http://huddlehive.io!