Monday, August 09, 2010

So Far, Not So Good

We are now 4 matches into the 50 match, multi-sport, daily fantasy sports challenge, and things are not going very well. It is early, so I am not overly concerned yet, but I would have liked a better start. I won the first fantasy baseball match, and then lost the next three. I worked hard to get up to speed on fantasy baseball, but my record has been pretty horrible on the site we are playing at. I keep telling myself that it is just variance, but I have no proof that this is true. My biggest problem so far is in hitter selection. I think I am doing a pretty good job here, but my hitters just don't seem to produce much no matter how good they are, or how high their daily projection is. My hitters have pretty much not shown up for three of the four leagues. In the one league that they did do well, my pitcher bombed for the first time, and I lost a close one. One big issue is the cap structure for individual players at the site we are playing at. It is just pure crap, and takes a big chunk of the skill aspect away. I don't know who came up with the formula that they are using for individual player prices, or even how it works, but it manages to misprice just about every player on the site. They should fire that guy! The mispriced players force you to draft a team from a narrow sliver of players that have any reasonable value at all, and ties your hands as to the types of teams that you can draft. For starting pitchers this is a HUGE issue. The top pitchers for the day, cost about half as much per projected fantasy point as any fantasy hitter on the site. As a result you almost have to take the top projected pitcher. Our pitchers have exactly matched in two of the leagues, and in the other 2 leagues Buffalo66 had my second best pick for pitcher. Keep in mind that there are typically 30 different starting pitchers to chose from, so this would not be the case without mispriced players. The site just does not allow you to choose a lower projected pitcher. If you drop down about 1/2 of the way down in pitcher cost, you will have enough cap left to pick any hitter you want. If you select one that is cheaper than that you will be left with tons of cap money that can't be spent. It is pretty pathetic. At Fantasy Sports Live the cap values are simple and transparent, and allow you to pick any one from the highest projected player to a scrub who is going to do better than average without any penalty.
This allows the your skill to shine and lowers the variance.

Fantasy Baseball just looks to be high variance in general, and taking skill away just increases the variance. I will need to work on my lucksacery a bit more I guess. My approach to daily fantasy sports projections is that past performance history, can help determine future performance. Intuitively this makes sense, and if it was not true, you really would not be able to beat fantasy sports long-term. So, I have built up a pretty advanced projection model for fantasy baseball based on that assumption. So either, my model needs some work, I have just been unlucky, or Buffalo66 is just better than me at this and I will not get there no matter how hard I try. My model has not been validated like my model for Fantasy Football was last year. So I will spend sometime validating my baseball model, tweaking it and seeing what happens. If I catch back up, it is going to be luck, skill, or some combination of both.

Labels: ,