The best way that I have found to come up with Fantasy Baseball projections is to take an existing statistical projection, and translate those numbers into fantasy value.
Projections that use statistical models free players from fantasy analysts who have biases against players; “I’ll never own player X again after last year”; “He won’t be able to produce on that offense.” These are mistakes I see fantasy analysts make all of the time, and in the former assertion analysts make assumptions about a player’s future performance based off of a meager sample size, and in the latter situation analysts attack a player’s team, which matters some but not as much as some would like you to think.
At the beginning of last year I took every statistical projection I could find (Zips, Steamer, Oliver, Marcel, FANS), averaged them together—statistical projections that aggregate a group of projections have proven to make forecasts on a particular topic 20% more accurate—and then translated those numbers into player rater value with the conventional means that rotisserie value is calculated. Last year the fruits of this labor produced Paul Goldschmidt and Adam Jones as top fifteen hitters when few other pundits had their pecking orders calibrated in such a way.
Because the only projection system that I could get a hold of at the moment is the Steamer projection, these numbers should not be the end all be all, but it is still interesting to look at why these rankings are constructed the way they are? Said another way, what do these projections see in player X from a statistical prospective, that qualitative forecasters miss the boat on?
Here are the Steamer batter rankings for the 2014 MLB season (These rankings are not adjusted for position, because I did not feel the desire to manually enter each player’s position into a spreadsheet for the next hour, but these rankings are still directionally correct):
[table id=8 /]
The top ten has some jarring projections in it, so lets take each player in the top ten and give a one-sentence quip as to why Steamer made their projection; let’s get inside the head of Steamer.
Devon Jordan is obsessed with statistical analysis, non-fiction literature, and electronic music. If you enjoyed reading him, follow him on Twitter @devinjjordan.
We came up with this list of gifts for card collectors because we know it…
The first couple years of Topps cards were pretty great, but the 1954 Topps Baseball…
These 2024 Topps Update boxes are so hot, it forced me to create this post,…
Looking for the best Topps baseball cards from each set, each year? It's a common…
Grabbing the best 2024 Topps Update rookie cards is always a good idea, since the…
Earlier this week, Topps made 2024 Update Series baseball boxes available on its website for…
This website uses cookies.
View Comments
I feel like this list is a tad bit incomplete without Carlos Gonzalez....
Not saying it is complete, just something to roll around in your head. Steamer also projected Encarnacion for 21 games, Bautista for one game, Pujols for one game, Craig for one game. This must be an error by Steamer, or they're predicting injuries, which I don't think the systems does; so it must be an error.
If they were predicting injuries then Ellsbury would not be that high
Someone on Reddit noted that Steamer does not project any players that are crurently on the DL
David Wright?
Torii Hunter is rated HIGH.