Last week, I took a look at five hitters for whom player forecasting systems strongly disagreed in their 2014 projections and offered my take on which of two divergent opinions makes more sense. Today, it's time to take a look at some unpredictable pitchers.
Using the Steamer Projections created by Jared Cross, Dash Davidson and Peter Rosenbloom and the Oliver system designed by Brian Cartwright, I isolated every pitcher whom both models project to throw at least 100 innings in 2014. (I use Steamer and Oliver not because they are necessarily better than any other systems but because their 2014 numbers have already been published and the data was easily accessible on FanGraphs. It is worth noting that Steamer provided the most accurate preseason predictions for pitchers in 2011.)
I then compared the two systems' projected ERAs for each player in search of the biggest inter-system disagreements. Here's a look at five of the most interesting names about whom the projection systems disagree -- and which one is more likely to be right. (A full chart showing the 20 most divisive pitchers can be found at the bottom.)
2013: 3.27 ERA
Steamer 2014: 4.18 ERA
Oliver 2014: 2.90 ERA
At first glance Oliver's projection seems much more reasonable than Steamer's despite the ominous gap between Weaver's ERA and his peripheral numbers in 2013. Weaver has a history of defying FIP (after eight MLB seasons, his career ERA is 86 points lower than his career xFIP) and it has been five years since he last had an ERA above 3.27.
But as Mike Petriello explains well, Weaver has been trending the wrong way in his velocity, strikeouts and batted-ball profile three years in a row, to the point where he no longer fits the profile of a pitcher who's likely to outperform his peripherals. An ERA over 4.00 seems overly pessimistic, but Steamer's projection probably will be closer to reality than Oliver's.