There’s an ‘‘x’’ factor among baseball metrics that are designed to make them better predictors of future performance.
The ‘‘x’’ is short for ‘‘expected,’’ and stats with an ‘‘x’’ in front tell you they are trying to factor out possible chance influences.
One of the earliest ‘‘x’’ stats is for pitchers, where Fangraphs.com has listed xFIP for several years. It eliminates volatility in home-run rates by using league-average rates instead.
Tyler Chatwood, 2-3 with a 3.31 ERA so far in his first season with the Cubs, can be used as an example of how it works.
With the Rockies in 2017, Chatwood was 8-15 with a 4.69 ERA. One basic tool to check how much was Chatwood and how much was outside influences is FIP — fielding independent pitching.
FIP focuses only on events in which defense doesn’t come into play: walks, strikeouts and homers. The formula is (13*HR+3*(HBP+BB)-2*K)/IP, then multiplied by a constant to put it on the same scale as ERA.
Last season, Chatwood pitched 147‰ innings, struck out 120 batters, walked 77, hit two and gave up 20 homers. His FIP was 4.94, higher than his ERA. That’s a problem because FIP has been a better indicator of future ERA than ERA is itself.
But Chatwood pitched his home games in Coors Field last season. The Rockies and their opponents hit 208 homers in Coors and only 174 in Rockies road games last season.
There are other factors that make home-run rates volatile. Weather conditions can help or hurt hitters. The sample size of homers by any given pitcher in a season is small enough that swings in home-run rate from season to season are inevitable.
XFIP tries to work around the problem by substituting a league-average number of homers, given the number of fly balls the pitcher allows.
In Chatwood’s case, that lowers his xFIP to 4.27, suggesting he pitched somewhat better than his 4.69 ERA.
XFIP isn’t perfect. It can underestimate a pitcher who allows a high percentage of fly balls but can induce weak contact, and it can overestimate a pitcher who is genuinely homer-prone.
But xFIP and similar internal metrics developed by ballclubs give teams another reference point in player evaluation to go with other metrics and the scout’s-eye view.
Early in their rebuild, the Cubs signed low-cost free-agent pitchers Jason Hammel and Scott Feldman, who had looked better by FIP and xFIP than by ERA. When they performed well, the Cubs were able to trade them for younger building blocks. One was Jake Arrieta, who also had looked better by FIP and xFIP than by ERA.
It’s a two-way street. Chatwood’s xFIP last season suggested the possibility of better things ahead. This season, his 4.26 FIP and 5.03 xFIP, inflated by 7.44 walks per nine innings, suggest he hasn’t pitched as well as his ERA.
For Chatwood to have the ‘‘x’’ factor working for him, the walk rate has to come down.