Master Cheat Sheet
Three years ago, I wrote an article for RotoWire wherein I described building a fantasy baseball cheat sheet by combining the best algorithmic projections systems with the best market-based average draft position (ADP) from the NFBC.
From the article:
But where the light bulb went on was after I read an article about a horse-racing bettor who had a moderately successful algorithm that only took off when he combined it with publicly available market data:
A breakthrough came when Benter hit on the idea of incorporating a data set hiding in plain sight: the Jockey Club's publicly available betting odds. Building his own set of odds from scratch had been profitable, but he found that using the public odds as a starting point and refining them with his proprietary algorithm was dramatically more profitable. He considered the move his single most important innovation, and in the 1990-91 season, he said, he won about $3 million.
The idea was to mine the best information available by combining the vast knowledge contained in the high-stakes market with the most predictive performance metrics. The market handicaps who's slated for closer roles and how healthy players looked late in the season, while the projections systems aren't mislead by recency bias, buzzy players or cosmetic results not commensurate with underlying skills.
Last year, I updated this process and explained it below. Skipped the italicized parts if you don’t care how the sausage was made and just want to see the hybrid cheat sheet:
I’ve since made some changes as to how I calculated the respective values for each projection system, and how much weight I gave to each. While I gave 1/6 of the overall weight each to Steamer and Derek Carty's The Bat, and 2/3 to the NFBC in 2020, this year, I went 1/4, 1/4, 1/2, respectively, in part because I calibrated the Steamer and Bat projections in a more NFBC-centric way. One change I made was incorporating a speed premium that shows up in NFBC drafts, in addition to the two-closer, seven-starter baseline and separate catcher baseline I was already using.
Accordingly, I used recent NFBC ADP, setting the No. 1 player, Francisco Tatis, at $45 and the number 276 (23 starters * 12 teams) player at roughly $0, and made the total add up to $3,120 by multiplying $45 (and each successive draft slot value) by .9856585. (As you might imagine, it took some trial and error to get the parameters to line up.) I went with $45 for the No. 1 overall pick instead of $52, or whatever Tatis fetches in NFBC auctions, because auctions are a different ecosystem, and the top overall valued player by Steamer and The Bat, according to my formula, was closer to $45. In any event, the starting number wasn't likely to sway the overall distribution much.
To get the speed premium, I added up the values all the players projected for 10 or more stolen bases by Steamer in NFBC dollars (based on the formula above) and compared them to my original Steamer and Bat dollar values. It showed the NFBC had them worth about $268 more collectively, so I divided by the number of stolen bases projected for those players, and it came to about 32 cents per steal. Of course, I had to subtract an equivalent amount from players with nine or fewer steals to offset the extra spend, and I did so at minus 48 cents per steal below nine. So a player with 11 steals would get an extra 32 cents added, while a player with seven steals would have 96 cents taken off his value. It’s a crude way of doing it, but it more or less builds in the speed premium unique to overall contests where you have to do well in all the categories.
I made more tweaks this year which might be too dull to get into in depth (for closer rankings I used standard deviations and strikeout and win replacement baselines from the overall pool rather than the separate closer one because each closer really does contribute negatively in those categories, but massively in saves. The way I had it before, comparing closers only to closers, some setup guy with eight saves and a .85 WHIP might crack the top ten among closers, even though he’d never be drafted there. Now the list more closely resembles ADP, though of course the WHIP, ERA and K projections still do matter. As I said, probably too dull to get into.)
In any event, here’s the overall master cheat sheet as of Saturday, March 4. Note: I did not use this for my first Beat Chris Liss Draft because I had finalized it maybe 10 minutes before the draft started and hadn’t had a chance to look it over.
Keep reading with a 7-day free trial
Subscribe to Real Man Sports to keep reading this post and get 7 days of free access to the full post archives.