Interesting. Everything else offensively suggests Cave. Cave had the higher OPS .797 to .727, higher OPS + by far, 113 to 95, and BR predicts .259/.317/.445 (.763) for Cave vs. .235/.317/.419 (.736) for Kepler in 2019. Pretty wide offensive disparities all around. Kepler is supposedly better defensively but it is interesting that Molitor tried both in CF last year and chose Cave over Kepler. Suggest at least he thought Cave was the better OF. Besides, these guys are playing RF since Buxton has been "'promised" CF - bad idea to promise anyone who hasn't preformed anything other than an opportunity to compete in my view - so I would say offense is more important than defense.
I think Cave wins on the potential issue, at least based on results so far from Kepler. He's had over 1550 ABs and he hits like a 4th OF - .233/.313/.417 .(730), not like a good corner OF on a winning, contending team. Cave was better last year at the plate, actually much better. Who knows what we'll get this year from him, but he deserves the chance to play over a guy who has had his chance and not been able to show he can consistently hit MLB pitching. . My point is that mediocre teams pick winners and don't change when they don't perform as hoped and/or perform worse than someone they didn't pick as the winner. Let's not be that team.
Without digging into a stat that I don't know much about, I'm assuming they are banking on Kepler's low babip's being a product of bad luck.
In other words, with neutral luck, Kepler "should" put up better offensive numbers than he has in the past.
I'm not necessarily agreeing with that. There are more variables that go into babip than just luck. But, theoretically, Kepler's numbers could be the victim of poor luck on balls in play, in which case those numbers should be expected to improve, even without any actual improvement from Kepler.