We are roughly 27% through the MLB season and I wanted to analyze my results and try to draw conclusions about why I'm doing well on certain sites and doing poorly on other sites. Since I won't be playing til next Tuesday, I'll release one blog post a day analyzing each site.
Overall Season Results
I'm up $22,616.69 with an 8.78% ROI on the season. This exceeds my expectations as I came into the season unsure if I was a winning player and expecting to quit at the first sign of resistance. Despite being pleased with my overall results, my site by site results are troubling, so I wanted to take some time to analyze each site in detail and wonder why if there's something other than variance to explain the results.
Yahoo
Yahoo has been my best site so far this season. I actually have a touch higher ROI on Draftkings, but I play over 1.5x as big on Yahoo, so a lower ROI is normal, but given the increased volume the ROIs are much closer than expected. These results are not a surprise for a few reasons.
No Touts for Yahoo
Touts are DFS players who charge subscription fees in exchange for providing daily advice. They are usually moderately successful professional players: the best pros would probably lose more money by giving away information in exchange for their picks, and unprofitable players are usually found out quickly. Moderately successful pros on the other hand, could use the extra income from touting, and give out solid advice: they break down stats, mention players they like, and help guide others to make better lineups.
Most touts give out plays for Draftkings or Fanduel because they are the two biggest fantasy sites. Fanvice and Rotogrinders also have live shows specifically oriented towards FantasyDraft plays. But none of the touts are giving out information for Yahoo. So with a different scoring system and player pricing, my opponents that try to use advice given for the main sites may choose overpriced players or overlook underpriced ones not mentioned.
Very different player pricing and scoring
On Draftkings, the minimum priced hitters are $2000, and the maximum priced hitters are roughly $5600 - less than 3x more expensive. On Yahoo, the minimum priced hitters are $7, and the maximum priced ones are $28, a difference of 4x. Add in a different pricing ratio for pitchers/position players, and very different scoring, and the lineups are built in a very different way from the main sites.
Building lineups on Yahoo is completely reliant on math and optimization. The brain does a horrible job of estimating what the best lineup is when player pricing has a larger disparity, so using math over hand building gives a significant advantage.
Weaker Competition
Not only are there a lower percentage of strong players in the larger field contests, but some of the strongest DFS players play poorly on Yahoo due to the reasons listed above. The result is only 3-4 guys setting consistently elite lineups every night.
Occasional overlay
Overlay is when guaranteed contests do not fill. For example, if Yahoo runs a 6 man guaranteed $1050 buyin 50/50, Yahoo puts up 6k for the prizepool, and each player puts in $1,050 to play in the contest. If it only attracts 5 players, then Yahoo has to put up the remaining $6000-($1050*5) = $750 in prize money for not filling the contest. If all 5 players are equal skill level, they each profit $150 in expectation by playing the contest.
Early in the season and on some of the smaller slates, Yahoo's contests were not filling. Although the overlay is infrequent, it probably added about 2% to my ROI on the season thus far.
I'm most likely the best player on the site
This is more of a combination of the previous things, but is worth a standalone mention. There are two players who are probably better MLB DFS players on the main sites, but one max enters GPPs on Draftkings as well as playing Fanduel so his attention is spread thinner, and the other is more of a feel based player, and that style is not suited towards Yahoo. Being the best player means every contest is +EV, so I can enter more buyins than I would on other sites and really push my edge.
Mistakes made so far this season and other notable choices
I wanted to scroll through my day to day results and make note of any possible mistakes I made or interesting things to note that could raise my ROI further.
4/5: Played Michael Pineda and Jacob deGrom at pitcher, did not play Chris Sale. At the time it was probably, but in hindsight this seems insane. The swap was Sale+punt 1b vs. Pineda+Goldie, on a day where Goldie was the #1 overall play. Given my projections for Sale now, the Sale side wins no matter what, but at that point it was at least close.
4/13: Unfortunate late swap. Was down 3.5 points with my team having two at bats remaining and my opponent's team having two slightly better at bats remaining, and both of us had Madison Bumgarner. His opponent, Jon Gray, is an underrated pitcher. I chose to swap because I thought it was more likely Gray beat Bumgarner by 3.5 yahoo points than for my players to outscore his in two remaining at bats.
For yahoo purposes, this required my guys to do 2 things: single+run, two singles, a double, walk+steal, etc. I figured it was .65*.65 = 42.25% they both do nothing, likely another 10% they only do one thing and I lose, and then another 20% my guys do things but his do more and I lose anyway. I estimated I was ~72.25% to lose by not swapping, so I figured since Bumgarner was not a heavy favorite (-150), my EV was higher by swapping to Jon Gray.
What happened was almost unbelievable:
- one of my players homered
- neither of his did anything
- Jon Gray was beating Bumgarner through 3 innings before coming out of the game due to a toe injury
- Bumgarner stayed in, and beat Jon Gray by 10 points and I lost by 1.5. Had Jon Gray stayed in for one extra inning I win.
I don't think this was a mistake, but it was intellectually stimulating enough to worth discussing.
4/17: Forced in Danny Salazar at pitcher in a potentially rainy game. After this I started focusing entirely on math instead of making lineups by hand.
4/20: Rain in Baltimore game, and wanted 3 players from it. Removing those players gave up 3 points of equity, but likely a 15-25% chance game gets postponed. I think this was probably a correct decision, as when it PPDs I almost certainly lose, if it plays out and I fade I could win, and overall point EV wise if they PPD 15% it's break even with removing them from lineups.
4/23: Rogue played Chase Utley over Kolten Wong at 2b for no reason. Wong has more power and more steal upside, only chose Utley based on a slightly better lineup spot. Wong finished about 20 yahoo points higher.
5/2: Missed on Hanley Ramirez at first base.
5/9: Had close choices between Charlie Morton and Julio Urias lineups, should have split them instead of going all in.
5/11: Accidentally played DJ Lemahieu over Brian Dozier at second base because my hand made lineup was nearly identical to what I played.
5/19: Hated my lineup with McCutchen. Need to adjust his base rates, no longer an MVP level player.
Conclusion
Happy with my Yahoo results in this quarter season review, and don't think there is really much room for improvement. Made a couple slight errors here and there, but they were generally close spots, and I may have run slightly above EV.
Will dive into Fanduel tomorrow and leave Draftkings/FantasyDraft for last.
My name is Ryan Daut and I'd love to have you as a follower. Click here to go to my page, then click in the upper right corner if you would like to see my blogs and articles regularly.
I am a professional gambler, and my interests include poker, fantasy sports, football, basketball, MMA, health and fitness, rock climbing, mathematics, astrophysics, cryptocurrency, and computer gaming.