FanGraphs Feature Focus: On-Pace Leaderboards

The FanGraphs Depth Charts. ZiPS. Steamer. ATC. THE BAT. OOPSY. These are all fantastic projection systems that I look at on a regular basis, and you should too. But what if the “projection” system was just… basic division and multiplication? Enter the On-Pace Leaderboards:

While the On-Pace Leaderboards are located within the Projections section of the site menu, I cannot stress enough that the stats you’ll see aren’t based on any projection system whatsoever. They’re merely a rest-of-season extrapolation based on what a player has done to that point in the season, with the scaling changing depending on whether you select Every Game Played or Games Played % in the menu:

In either case, each player’s season stat line is first reduced to his per-game numbers; the difference is in the multiplier, and therefore how many games the player ends up with on the leaderboard. Every Game Played mode prorates a player’s statistics as if they’re going to play in every one of their team’s remaining games, regardless of how few they’ve played to this point, or whether they’re currently on the injured list or in the minor leagues. That leads to some ridiculous results, including such small-sample darlings as Enrique Hernández (who played in just two games after being activated off the injured list before he suffered a severe oblique strain) and Rafael Flores Jr. (who got into a couple of games as the Pirates’ backup catcher a few weeks ago before being optioned back to the minors):

For pitchers, “every game played” is considered pitching every four and a half days for starting pitchers and every two and a half days for relief pitchers. The proration doesn’t consider when a pitcher last pitched, so you may notice slightly inflated games started totals for pitchers who pitched within the last couple days:

The Games Played % mode, on the other hand, does consider the ratio of games played to team games for each player, which makes for a more normal looking extrapolation that doesn’t magnify the effect of extremely small sample sizes:

This view additionally softens the totals for starting pitchers, since it’s aware of how often a team utilizes a pitcher and is implicitly understanding of six-man rotations and extra days of rest:

With relievers, the every-two-and-a-half-days methodology tamps down totals for arms who have been frequently used, but the Games Played % mode applies the exact current rate of usage to each player individually:

To reiterate: This is not a projection system. No reasonable model would tell you that Rafael Flores Jr. projects to put up MVP-caliber numbers for the rest of the season if he’s called up again. These leaderboards are just a multiplying and dividing tool so you can more easily see who’s on pace to drive in 100 runs or pitch in 86 games and risk his arm falling off (hello, Mason Fluharty) rather than having to do that math on your own. The leaderboards are filterable to Default, Standard, Advanced, or Fantasy stats, and Members can export to Excel as with any other leaderboard.