Why Sell Low on Nova?

Nova vs TOR 2015 I had a four year-old Chevy Tahoe with a lot of life left, but a fender-bender left it dinged, scratched, and missing a side window. The needed series of cosmetic repairs would’ve cost maybe $3K over a few months, and by using it over those months, I could’ve verified that nothing structural was wrong. But I was kind of sick of it, so I left it looking like a Sanford & Son junkyard prop, and I sold it for $9K less than bluebook value.

OK, I made that up. I just envy the southern/midwestern trial lawyers who sell their points with homespun anecdotes about good ol’ American cars, horses, and cheese grits. I was really talking about Ivan Nova: I don’t understand the sell-low strategy of trying to trade a potentially useful player exactly when his value is at a low point.

The problem isn’t just trading him after a bad half-year. The bigger problem is that we don’t know whether his 2015 performance (1) was just bad because he’s bad now or (2) was suppressed by injury recovery and rust that he could get past in 2016. And the biggest problem is that possible trading partners have less idea whether it’s #1 (bad) or #2 (rusty) than the Yankees do.

Let’s say the team's assessment is that Nova has a 50% chance of returning to second/third-starter form, but a 50% chance of remaining bad. That may be a decent expected value – call it “back-end of rotation,” on average. But another team has to wonder: is it because the Yankees know he’s not improving that they’re posting a “Buy It Now” Nova on eBay?

This asymmetric information problem is why it’s hard to get good value trading a player with injury recovery uncertainty: the acquiring team has to be wary that maybe the fellow is for sale because he’s like a lemon of a used car, a dud that the selling team knows to have low odds of returning to form. You just can’t easily trade your assets for good value when you have a more confident, inside-information-driven valuation than the other team does.

So if the Yankees trade Nova, I’ll be disappointed, because it’ll be for a weak return, which will mean one of two depressing things. First, it could mean the Yankees know he’s unlikely to return to form, so getting a sack of used catcher’s mitts in exchange is a solid deal. Second, it could mean the Yankees are willing to suffer the loss of getting a below-expected-value return when you barter an asset with uncertain value to a trading partner with asymmetric information. I’m not sure which of these two possibilities I’d find more depressing: the conclusion that Nova is washed up; or the conclusion that they’ll suffer the loss of trading him at a low point in his value.

And whether regardless of whether Nova’s odds of improvement are low or high, the team needs a sixth and seventh starter in waiting, given their injury-risky rotation (Tanaka’s/Eovaldi’s elbows; Sabathia’s knee; Pineda’s shoulder and second-half collapse). Even if Nova 2015 is the permanent Nova, his 5.07 ERA was slightly worse than predicted by his 4.87 FIP, and it was still good for 1 WAR/yr – as good as most teams’ fifth starters. It would be a mistake to dump a useful spare part, who also has potential to be more, out of disappointment with the falloff from the 2011-13 rising star Nova to the 2015-16 spare-part-plus Nova.