ZiPS and Steamer Rest of Season Projections
Over Saturday, I aggregated last week’s ZiPS and Steamer rest of season projections (found on Fangraphs) into one list, and then translated those numbers into Fantasy value using Jeffrey Gross’ E.Y.E.S. formula to come up with what one may consider the most thoughtful rest of season rankings.
When I say, “most thoughtful,” I simply try to find a word that distinguishes these rankings from the rest of the rankings you will find on the Internet.
Before the season started, I performed a similar task to what I described in the previous paragraph with aggregated Marcel, FANS, ZiPS, Steamer, and Oliver projection systems found on Fangraphs, and drafted my 12-team 5×5 mixed league according to those results. To say I have been pleased with those results would be an understatement. The formula I used ranks batters in the traditional five offensive categories: runs, runs batted in, home runs, stolen bases, and AVG.
Here are the results of those preseason-aggregated projections:
As you may see, there are several outliers on these rankings relative to the mainstream media’s perception of some players (Especially when you consider these forecast were made preseason.): Giancarlo Stanton, Billy Butler, Adam Jones, Bryce Harper, Starlin Castro, Anthony Rizzo, and Paul Goldschmidt.
Before I show you the results for the process I performed on Saturday, let me preface those rankings with three notes.
1. These rankings attempt to project performance from now until the end of the season, and do not incorporate numbers already accrued by a player.
2. ZiPS and Steamer are not capable of predicting playing time, so if there is a player who is currently hurt for an extended period (i.e. Troy Tulowitzki), these projection systems do not have the bandwidth to take prolonged absence into consideration. (I just started a new job, and that is why I have not written in two weeks, but everyone at this company loves to use the word bandwidth. Bandwidth. It’s a funny word. What’s even more comical is I’m writing this article at work right now! OK, back to work).
3. These projections, unlike the projections I made with the preseason forecasts, are not adjusted for position. Weighting players for position requires a lot of extra work because the CSV download from Fangraphs does not include position. In the preseason I took the time to go through each player and fill out his position, but it was a beautiful day on Saturday and, well, you know. Positional adjustment does not drastically change a player’s ranking, and at most, changes their spot by a few positions.
Here are the rest of season aggregate ZiPS and Steamer Fantasy projections:
A few thoughts:
Carlos Gonzalez was not originally ranked in the top 30 for the preseason projections and is now fourth. Projection systems are very conservative when it comes to altering their forecasts, so ZiPS and Steamer must think there is a considerable reason to believe the CarGo will continue his first half performance.
Chris Davis (3, 43), Everth Cabrera (5, 151), Jean Segura (6, 84), Carlos Gomez (9, 82), Domonic Brown (16, 106), Dexter Fowler (27, 50), Starling Marte (29, 45): The first number after each player represents their current rank on ESPN’s player rater, and the second number represents the players rest-of-season rankings on the aggregate ZiPS and Steamer projections.
Like I said about Carlos Gonzalez, projection systems are conservative with alterations to their forecasts, and while ZiPS and Steamer have bought into what Carlos Gonzalez has done, they are pessimistic about the players listed at the beginning of the paragraph. The MLB season is not even half way through, and to drastically alter a prior belief about a player over half a season of baseball is foolish.
Billy Butler (129, 25), Matt Holliday (47, 22), Anthony Rizzo (65, 11), Giancalro Stanton (371, 10), Robinson Cano (31, 7): The first number after each player represents their current rank on ESPN’s player rater, and the second number represents the players ranking on the aggregate ZiPS and Steamer rest of season projections.
Matt Holliday’s BABIP is 40 points below his career average.
Stanton has been hurt the majority of the first half, and if you’re reading this article you’ve probably waited too long to buy low on him; Stanton hit two home runs the other night, and has his hit three long balls in his last few games.
Under the assumption that you would not have to pay very much at all to get him, Anthony Rizzo is my favorite buy low. Rizzo’s ability to walk evaporated in May (3.4 percent), but it appears to be back in June (22 percent). The resurrection of his plate discipline will also lead to a rejuvenation of his power numbers, and prove his two-HR May to be an aberration.
Devon Jordan is obsessed with statistical analysis, non-fiction literature, and electronic music. If you enjoyed him here, reading about these rest of season projections, follow him on Twitter @devinjjordan.