Here we might have another different view point. For me, if you have the choice between a battle with 51% win rate and positive TUV (even +1) vs doing nothing, so long as it doesn’t have a negative impact on your position - you take the battle.
You take all battles with positive expected outcome. And of course you’d favor the 52% win battle with 0.1 more TUV over the 51% one.
Unlike some/most gambling (I gamble on politics at PredictIt), there aren’t any “house” fees - so the tiniest imaginable edge is useful.
Unless you expect to beat your opponent by a large margin, in which case taking fights that are only slightly favorable to you might be detrimental.
The critical fact about attacking with 50/50 odds, is that with strategic withdrawal the battle simulators might be under-estimating your TUV gain (it’d be interesting if someone were to write a calculator that would handle that!). Admitedly this is a rare situation and you are much better off attacking at 95%+ win odds most of the time.