For starters, I have made a pivotal assumption. Rosario was absolutely fantastic in the last half of 2017 and the first half of 2018. The last half of 18 he played at replacement level and has not been even remotely as good every since. The assumption is the league adjusted. I welcome any other theories if you care to reject my assumption.
You are assuming my position is Rosario sucks and Kepler is great. My position is Kepler improved significantly on his weakness which was left hand pitching. Rosario has not adjusted and therefore Kepler has been better offensively. The number are crystal clear if you accept something significant changed after Rosario's exceptional performance from the 2nd half of 2017-1st half of 2018. Add to this Kepler is much better defensively. He can even play an important role in backing up CF and his cost was about half of the arbitration estimates for Rosario. It's not simply Rosario vs Kepler. It's Kepler + $5M to be used toward a SP or one of the better utility guys or a couple BP arms vs Rosario. For some reason this reality is be ignored by many.
I agree that the comparison between Rosario and Kepler has to include Kepler's superior defense and lower salary, and that it is rational to save the additional roughly $3-5m Rosario costs and could even help the team IF its used on other player improvements. We will see if that's the case; I hope it is.
I think what I and others on this site don't get is the idea that Kepler is anywhere near equivalent as a hitter to Rosario. Outside of 2019, he simply isn't even close. Comparing their OPS+ for the last 4 seasons (2017-2020) - Rosario 119, 116, 107, 115. Kepler - 95, 97, 123 (2019), 108. Why is Rosario easily better each year except 2019? Because Kepler can't hit left handed pitching, except for 2019. Kepler's splits vs. LH (2017-2020) - .152/.213/.240 (.453 - horrible), .245/.313/.422 (.745 - not bad, except better than he hit RH ptiching so overall not great), .293/.356.524 (.880 - great), and .128/.208/.170 (.378 - horrible again). So, is that evidence of steady improvement with 2020 a blip because of "unusual circumstances", or is this evidence that the guy can't really hit LH very well and had a career year in 2019? Who knows? What we do know is that his OPS+ tells us that he's a below average hitter for a corner OF, who in MLB should be at or above .785-.800 year in, year out.
Don't get me wrong, Kepler is a fine player and I'm glad we have him long term. But let's not kid ourselves about what he is - he's a very good to excellent fielding corner OF who can be an average or slightly above CF in a pinch, who runs well, has power, but is overall a below average hitter for his position. He's not a great leadoff hitter and he's not going ot be able to replace Rosario in the middle of the order. The Twins are going to have to find someone else to do that job.And that's why some of us are concerned that Rosario was let go and are hoping that the plan is something other than a hope that a rookie will be able to repalce his production in the middle of the order. Kepler is not that guy. He should hit 6 or 7 after Sano, who should hit 5 or 6 depending on whether Cruz hits 3 or 4, assuming Cruz is back.I wonder about our batting order if Cruz isn't back. It could be ugly.