Projection Accuracy: Late March Hitter Counting Stats

After diving into early draft hitter projections, the late draft season hitter projections get their time in the sun. First up is the counting stats that are heavily influenced on accurately guesstimating playing time. As with the early projections, the Bat and the Wisdom of the Crowds stand out with the addition of the Pod projections joining the others near the top.

For the projections, I pulled the following ones from the morning of March 30.

• Steamer (FanGraphs)
• ZIPS
• DepthCharts (FanGraphs)
• The Bat
• The Bat X
• Davenport
• ATC (FanGraphs)
• Pod (Mike Podhorzer)
• Masterball (Todd Zola)
• PECOTA (Baseball Prospectus)
• RotoWire
• Razzball (Steamer)
• ZEILE (Fantasy Pros)*
• Paywall #1

I didn’t run the values on CBS even though I pulled them. They were missing quite a few players and I messed up not pulling the Utility-onlys. Additionally, I pulled the ZEILE projections which are an average of several projections.

To create a list of players to compare for accuracy, I took the NFBC Main Event ADP (players in demand at that time) and selected the hitter in the top-450 drafted players (30-man roster, 15 teams in TGFBI). To determine accuracy, I calculated the Root Mean Square Error (RMSE) for two different sets of values. RMSE is a “measure of how far from the regression line data points are” and the smaller a value the better.

• All hitters in the subset.
• Removed six who missed quite a bit of time (Lewis, Mondesi, Trout, Ozuna, Hicks, Rendon).

Additionally, I created an average of the projections. Here are the results (a lower RMSE is better).

Thoughts

• Even with a horrible showing in the plate appearance projections, the BAT’s were near the top of the playing time projections. The projections performed great when the playing time was set to the league average, so they will likely stand out again.
• The projections that combined three or more projections (ZEILE, ATC, and the average) stood out and this should not at all be a surprise. Combining the minds of several sources should produce a better result than just one. Simply, a playing time average is probably the best option.
• Remember that these results are heavily based on getting the at-bats right. Projections must guess on three different variables: playing time, rate stats, and run-scoring environment.

Next on the list, is to look into the rate stats late in the draft season. It’ll be interesting to see how projections handle new lineup spots along with team and league run-scoring changes.

* ATC, Clay Davenport, Depth Charts, Harper Wallbanger, Luke Gloeckner, Mike Podhorzer, CBS Sports, ESPN, RotoChamp, Razzball, numberFire, Steamer, THE BAT, and ZiPS.