A person might be tempted to think that DFS outcomes might fall along a bell curve. I thought so too at first. But it doesn't. Remember we are dealing with Casino types odds in a game that involves a modicum of skill. The graph of the outcomes will have some features not unlike a Bell Curve, but there are significant exceptions.
I broke down all my playing sections on Draft Kings where I played 5 or more games and extrapolated that information with a 10 game Norm. The idea was to be able to visualize DFS outcomes base on playing 10 simultaneous game. There are 11 possibilities for the day (Win10, Win9, Win8 down to Win0). And the numbers reveal a interesting picture with some interesting features.
Basically, this is a chart of possibilities of outcomes should I play 10 games during a DFS Session (All Day, Turbo, etc). The bottom row represents the number of wins during a given session, while the left column shows how many times out of 100 a particular number of wins occurs.
The middle part does resemble a bell curve with a slight shift toward the victorious side. Based on the numbers used for this chart, winning more games than lost occurs 44% of the time, and winning at least half the games occurs 66% of the time. I think this shows a reasonable level of success on my part at DFS.
The other more interesting part of the graph are the two "wings" set at the the very best outcome (winning all games) and very worst outcome (winning no games). I believe this is caused by the value of the lineup you set: on occasion, you will pick a lineup that scores so well or so poorly, one beats them all or can't beat anybody. This also accounts for the very rare occurrence of winning only one contest or losing only one contest. If you field a good or bad lineup, you are more likely to take them all, rather than simply losing one game. For example, on a day you put up a really high number like 138 Draft Kings points, you are more likely to win all your contests rather than have the misfortune of have someone you played have a better outing.
Enough of patting myself on the back. Here's the downside. I've stated before that winning half or more of my games 2 out of 3 times feels like success. I'd bet you had agreed with the statement as you read it above. Well it isn't. This particular area graph is a recipe for losing $60.40 after wagering $1000. It's basically a return of -6% over a course of 3 months. I've created an additional graph to help you visualize this.
The blue line is the exact data set listed in the other chart, but the red line represents the dollars won or lost when each outcome occurs. Because of the house rake, a clean sweep returns 80% while a blank slate loses 100%. Or winning 6 of 10 nets you $0.80 while losing 6 of 10 loses you more than 3x as much with a total loss of -$2.80. Heck, even a 5-5 day is a loss of a dollar.
It takes a really high percentage of victories to win money here, unless you get extremely good at never picking a lineup where a good pitcher has an off night, or a star batter takes an 0-fer. I can't do it, and I don't think you can because I'm pretty sure everybody has the off days. The percent of days where I field a lineup that loses 80% of its games is a mere 13%, which I think is quite a solid performance. However the odds at this game are never in your favor. You might believe that all I have to do is get a little better at this and I will win more. That's not really the case: I need to win by a very large percentage to see winnings. The math says that I would have to take 10% out of the losing half and put it in the winning half in order to approximately break even. The winning at DFS is marginal, though bad days at DFS carry the same lose-it-all possibility.
No comments:
Post a Comment