Auto-resolver's predicted outcome seems to get worse sometimes, when Ogres use pre-battle feast

PurpleChair

Hi!

I've just finished a campaign as Greasus, and I just wanted to add a quick bug report for the way that the auto-resolve system seems to predict worse outcomes when I increase my pre-battle feast, sometimes.

I don't have a quick and easy reproduction to share, but it's something I noticed quite a few times during the campaign. A typical example would be that I'm predicted a Close Defeat at first, and then I ramp up my pre-battle feast, and I notice that the prediction changes to Valiant Defeat... but only when I'm spending exactly 3 meat. For any amount above or below 3, it changes back to Close Defeat.

I'm not even sure whether Valiant Defeat or Close Defeat is worse, but that doesn't really matter here - clearly at some point in this process (either when going up to 3, or up to 4 meat) I am spending more resources on a thing that buffs my troops, but being given a worse result? I can understand that the auto-resolve calculation may be tricky and weird sometimes, and I'm not grossly upset by the fact this is happening, but it does seem like there's a flaw in the logic when one single variable is increasing and the outcome is swinging back and forth, and I felt like it was worth reporting.

Now that I think about it, this is probably something that happens most often with Ogre Camp defenders? My main armies were usually too big to lose, and settlement garrisons don't have the feast option, but my smaller camp garrisons fought a lot of losing battles along the way.