Short-term results have to be pretty far out of line with expectations to move the needle much in long-term projections.
That the White Sox are 15-15 and have been in and out of first place in the American League Central has made for an interesting month-plus, but they’re still a rebuilding team. Fangraphs.com projections point toward a 56-76 record for the rest of the season.
The same goes for individual players. Without a major breakthrough or disastrous collapse, projections will suggest a player’s most likely results from this point forward will be close to his norm.
Take starting pitcher Jose Quintana, who has struggled to a 2-5 record with a 4.46 ERA and 4.01 FIP (fielding-independent pitching). For his career, Quintana has a 3.45 ERA and 3.49 FIP.
In projecting the rest of his season, his career levels — especially his most recent three seasons — are given more weight than a weak start. Opposition is a factor, and so is the 28-year-old Quintana’s place on the aging curve.
FIP is a better predictor of future ERA than ERA is itself, and Quintana has shown some erosion in FIP. From 2014 to 2016, his FIPs were 2.81, 3.18 and 3.56 even while ERAs were steady at 3.32, 3.36 and 3.20.
So for the rest of the season, Fangraphs projects Quintana with a 3.76 FIP and a 3.84 ERA — better than he has been so far this season, but continuing the FIP erosion of recent years.
To translate that into Fangraphs’ wins above replacement, Quintana is at 0.7 this season. It’s early, but he posted fWARs of 5.1, 4.8 and 4.8 from 2014 to 2016. By his current projection, he would post 3.5 the rest of this season.
Quintana stands as an example of the differences between fWAR at Fangraphs and bWAR at Baseball-Reference.com. Fangraphs bases pitching WAR on FIP and Baseball-Reference on runs allowed.
When Quintana’s FIPs were better than his ERAs in 2014-15, his fWARs were better than the 3.2 and 4.1 at Baseball-Reference. With a weaker FIP but a strong ERA last season, Quintana had a 5.4 bWAR that exceeded his fWAR. With a high ERA this season, Quintana is on the negative side with a -0.3 bWAR.
Which is more useful depends on what you want out of the stat, with bWAR regarded as more reflective about what has happened — how good was the pitcher at keeping runs off the board? — and fWAR more predictive of how good a pitcher will be in following seasons.
Fangraphs’ predictive approach allows it to update projections day by day during the season. The numbers are not hard-and-fast predictions. They’re closer to the most likely end points in a range of probabilities, guided in part by how similar players have performed in the past.
For Quintana, the most likely scenario by the numbers is that for the rest of the season he will be closer to the pitcher he has been in past seasons than the one he has been in April and early May.
Follow me on Twitter @GrochowskiJ.