[chriskelly]

I am 1 week into 2019, real rosters, Marlins. Scouting accuracy is on High. trade dif is high / favor prospects. 50% ratings, then 35%, 10%, 5%. I just stared, and with accuracy on high, i feel computer players are currently very close to their real life equivalents.

I mainly trade solely by shopping players. What I'm finding:
-Many teams who are not good will trade solid youth for a 38 year old Curtis Granderson. For example, baltimore is somehow in 'win now' mode and will give me their young starting 3B. No idea why.
-BOS is trying to get rid of 1 of their 3 catchers. that makes sense. but, one of the players they are willing to take back.....my backup catcher
-Teams seem willing to trade quite a bit for guys that won't crack their starting lineup.
-I have a feeling AI teams aren't considering when their injured players will return. For example, NY will trade Greg Bird for Granderson. Is this because Hicks is currently hurt? I can see how bench depth might help them, but bird is better rated as a bench player than Grandy, as is their best AAA OF.

I end up saving, making a deal, simming a few days, and seeing what the AI does. if it seems they traded for a guy and gave him a big role, then fine. if not I alt/ctr/delete back and undo it.

I'm tempted to switch to very hard. Have you guys found a best fit for ratings vs stats in AI evaluation? also, how does that work at the beginning of the season? Does the AI then value ratings or prior year stats more than it would later in the year? I almost never do more than a 2 for 2 trade, or it seems too easy to steal