The star ratings thing is really just relative to the talent in your league. So a guy with average ratings in an MLB league could play in a lesser league and be a 5-star guy. The ratings are deceiving, unfortunately.
I would say, evaluating the stats and just look at the relative strength to talent in the league to determine what you think of said player.
As for the ratings changes and him being on the DL, the DL likely caused his ratings to drop, which would mean his stars would drop. The things you'll want to look at in your league are, the
AI evaluation setting. That setting will determine how the AI evaluates players. I think SkyDog's NABA has them set differently, but I don't remember.
I've been using 71/27/1/1 for my current solo league, because I like that it doesn't overemphasize past performance and yet, it keeps older stars in the game a while and doesn't result in that wackiness.
The other thing, and what was likely the culprit in your situation is the
Talent Change Randomness which is defaulted to 100. In Ronco's settings, which I think that NABA file uses, the number is 200. Meaning that players talent changes twice as much as the default. It'll result in more lower round draftees getting talent bumps that make them stars or good major leaguers.
But it'll also result in your better players having talent drops that don't make a lot of sense. I've lowered it to 50 in some leagues and in FOOL, we play with it at 1. Sure, lower round draftees won't develop into superstars and that's lame sometimes. But..I'd rather have some form of predictability or at least, less than often talent drops that don't make sense..even if an otherwise good player goes on the mend for a while.
Hope that gives you a 'worthwhile' primer for your 'money'.
