Wait, Hermit/MS will almost always be able to get 5 of the hermits in a mirror or against anything else. And 5 is enough. Will that really lose to feodum? When you kick off, you gain the remaining feoda, which will be ~3 by my estimation, and then more good cards. Market squares are at most half empty I would guess, so you should have enough time to catch up with a superior deck. I mean I've never tried this, but that's my guess.
Why would we assume that you will "almost always" get 5 Hermits? I get that it is the most common outcome, but no by some massive percent.
There are quite a number of cases where P1 ends up with an easy 6/4 split:
1. One player gets a 5/2 split, particularly P2. If P2 gets the bad split here, at most they can nab 4 Herms by the end of T4; P1 will have very high odds of being at 5 by then and then dropping to 6.
2. P1 hits Herm/3C on both T3 & T4. T3's expected hand is 2.92 Coppers, .83 Herm, and 1.2 Estates. Going with the nearest whole numbers that is 3C, Herm, 1E. For T4 the expectation numbers are then 2.86 C, .71 Herm, and 1.4 E. Sure 4 Hermits on T3&4 for P1 is not the majority case, but it is the mode outcome.
3. P2 hits Herm/<3C or no Herm on both T3 & T4. In which case, again the turn order advantage goes to P1.
4. Both of one player's Hermits get buried to T5.
5. Both players singleton on T3/4 & P1 doubles (as they are down an estate or two) on T5 for a 6:4 split.
I'm curios what the sims actually bear out, but I doubt that 5:5 is going to pop up 90+ % of the time.