May I repeat that historically - as I noted above - on occasion mounted did fail to break LF types and were "bounced". This, however, was unusual.
My FoG experience so far is that LF can on occasion "bounce" mounted, however, this is unusual and usually the LF get ridden down if they manage to stand in the first place.
So far my experience is that FoG accords fairly well with history - and I've played quite a bit now as have the others posting who appear to share my viewpoint
I think there is a fundamental problem here with a couple of people appearing to assume that the unusual case in the game is in fact the likely one.
However, if anyone would like to stand with large amounts of LF in front of, say, my Seljuqs I would not stop them - although I don't easy victories all that enjoyable I prefer to be stretched.
All these relationships were extensively tested (10.000 runs on simulators) and then related back to history to the best of our knowledge, and common sense checked by the 3 of us.
Our view on this one was boradly as follows FWIW:
1. LF will usually go DISR when hit by mounted (or worse). In the open, if the usual Ave UnProt LF, they will be at -- POS vs Cv every time and have 1/2 as many dice. So typically 4 Cv dice plays 2 LF dice. If the LF have a swarm of numbers then at best 4 vs 4. In the 2 vs 4 sitution the avreage hits are 2.67 vs 0.67 before re-rolls (usually for the Cv only), or vs 1.33 if there are 4 dice worth. So typically we will have the foot taking CTs on -2 for losing the combat, maybe another -1 if lancers. It will take a 9 to avoid going DISred. Should be needed twice, so about 1 chance in 15 of them not being DSRed. You can improve that a fair bit with rear support or general, but even with both you will be DISred roughly 2/3rd of the time after 2 rounds.
2. We felt this was reasonable and if any mounted felt such good reistance from such unlikely culprits they would back off warm up and charge again to make use of their momentim as they would anything else proving resistive. This makes sense as it would be their way of avoiding getting swamped by numbers.
3. Impotantly this set up also gives Superior LF a much better chance of escape as they will hang on much more often. With no break off there is little to choose between them asthey will die like any others but with a break off the superior ones have a much better chance of survival (though still not good).
4. The - for MF in open in the CT works in conjuntion with the odds of losing. Strong MF in open do well, but not good when they get on the back foot. The 1/2 dice for LF and lack of POAs makes it much much more likely they lose in the first place and by a fair bit. So no need for additional -1, but then again it would do no great harm either if it did apply. The -s and odds of needing to test always need to be comboned.
Overall I think it is important for the LF to have some small chance of escape, and then some even smaller chance of victory. I am sure it happened in reality on rare occasions. Many things will have happened that we did not read about in the press of the day. It will be a rare event by design but will happen - see it as a fun challenge when it does.

I doubt many of our favourite generals had the benefit of POAs and a way to calculate their odds - they had many a surprise to deal with
By the way, if we had we had no break off my betting is I would now be answering a stream about how someones cavalry charged LF, didn't win and got gradually overwhelmed by numbers .. why is it that Cav don't break off vs LF!!!
Si