[Zeyes]

Originally Posted by kq76

I did some editing testing:

-changing to reliever does have a huge affect on increasing the ratings... which I guess IRL it should, but does OOTP really give a SP a boost in performance if used as a reliever? I doubt it does... even if just looking at players in the same role there's an obvious other problem

I'm not sure what you mean here. The increase in rating doesn't imply anything (certainly not an improved performance). It works exactly likes stars did in OOTP6 as far as I can tell, i.e. the rating is calculated by comparing the player to the positional average. The average reliever has worse ratings than the average starter, so the same pitcher will come out looking better as a reliever. That's all.

-lowering the GB% has a large affect on increasing the ratings and increasing it has an even more drastic affect the other way... if anything I'd say it should be the opposite, high GB% is usually better... I'd say this should definitely be looked at

That permeates the whole game though. Take a look at how the GB% affects the estimated pitcher ERA in the editor...high GB% = large ERA, low GB% = small ERA. So the OVR rating at least follows along with everything else in the game, even if the underlying model might be questionable.

-Stuff and movement both have significant affects with control slightly less, but please note that this should depend on the lg environment (if you have a lg with a lot of homers, movement should have a big affect and vice versa) and I doubt it does

I think this is automatically built into the calculation because it's based on the positional average. If League A has a lot more homers than League B, it stands to reason that the average League A pitcher has a lower Movement rating than the average League B pitcher, so a pitcher with good Movement should automatically be rated better in League A than in League B.

Of course, this assumes that the positional averages are calculated separately for each skill. If the game calculates an overall number for each pitcher, and then normalizes it to the league average, it won't work as I described it.