The whole argument assumes that the best thing to do for a fascist president is to lie about receiving three fascist policy cards. The moment that becomes the assumed behaviour of a fascist president, then it becomes beneficial to lie about the card chosen by the chancellor (i.e., "I passed one of each, they selected the fascist one!"). They also complain that the best strategy for Hitler is to play super liberal, but again, the moment that's the assumed behaviour, doing elsewise becomes a better choice; also, liberals should be trying to become chancellor, because the only person a liberal player is sure isn't Hitler is themselves. Most of the complains seem to be a group think problem.
Depends on what you are comparing it to. I understand the complaint about the policy deck being random. But I don't think, say, Resistance with more than 7 players has more real information than SH. And Resistance does have the "first leader randomly chooses team of only rebels" problem.
First of all, that's a terrible game design argument that designers of bad social games commonly deploy. I call it the Metagame Will Solve argument: my game has a broken strategy, but it's OK once everyone is aware of the strategy because then the Metagame Will Solve for it. Not only does it assume that you only play a game with the same group of people over and over again, and that you're OK with a bunch of ruined games to start ... but also you can make games without broken strategies, and the metagame is still interesting!
Second of all, it doesn't work in SH. For example, even if liberals know that Hitler is going to act liberal, that still doesn't change Hitler's incentives! Because "acting liberal" doesn't mean spending the whole game ranting about male privilege. It just means not acting suspiciously. And that's always going to be the winning Hitler strategy because by definition liberals will always vote in someone that they least suspect of being Hitler. So playing Hitler means doing your absolute best to forget that you are Hitler. This isn't a strategy you can counter even if you're aware of it, because it's a strategy that depends on doing absolutely nothing. It's a Nash equilibrium of boring.
The whole point of a hidden loyalty game is that a traitor acting non-suspiciously is easy
but that you need to take risks in order to maximize your gains. That is the land of interesting decisions and compelling games. A hidden loyalty game where a traitor is best rewarded by playing it safe defeats the entire point of the genre.
I've only tried Avalon once, and I didn't like it as much as Resistance, because it feels very hard for Merlin not to get assassinated. But a lot of people seem to think it's the better one, so I must be missing something.
PS: I love Resistance, by the way.
I actually hate this part of Resistance too. I posted about it on BGG. Then I was reminded why I left BGG. https://boardgamegeek.com/thread/1397174/modifying-how-merlin-identified
Base BSG has the problem that you can't convincingly claim not to have an XO.