Tigerleap wrote:snip
"Actually, if two people with the same priority visit, the Crusader would choose the one who chose their night action first. I believe that's how priority works in the current game. This means that the Crusader doesn't have to be RNG."As I said elsewhere, fine by me - I'm not convinced it'll add or detract a great deal from the role or from the experience of other players, but if this mollifies people who have a legitimate, understandable and well-founded hatred of RNG then it'd be a worthwhile change.
"As for the Psychic, let me try to explain why RNG is bad. Imagine that there's a multitude of parallel universes out there, and each one represents one of the outcomes that the RNG could give. In one of these universes, Town would win because of the Psychic's information. In another, Town loses because the Psychic's information is worthless. The two outcomes are very different, but the way that the Psychic plays is the exact same; the actual Psychic doesn't have to do any investigating--they just have to hope that the RNG gives them good information. Another problem with the Psychic is that it requires absolutely no skill to play as, other than jotting down the information obtained. The game's crappy RNG spoon-feeds the Psychic with potentially game-changing information while the Psychic just sits there hoping they don't die."I entirely appreciate the point, having been in both situations as Psychic and against Psychic. However, I've yet to see a credible and workable suggestion for shifting the Psychic to a pure non-RNG version which would be workable and avoid either gimping the Psychic or making it even more overpowered than it already is. Obviously, one option is to throw it into the fire and never go near its smouldering corpse ever again. This was pretty much my immediate reaction to reading the role description, but I'd prefer to be constructive with BMG if at all possible.
Speaking personally, I'm not opposed to RNG ideologically. The vast majority of ToS games are influenced by pseudo-random factors as it is - for example, decisions on N1 visits have a strong randomised element to them due to the near-total lack of information on all sides, and the tendency for the metagame to shift between people deciding to actively target, or not target, unusual names. Roles such as Medium or Disguiser can be hugely influential in the 'right' game, and wither into insignificance in others. While both have the opportunity to apply skill, they have less direct influence in most cases than the majority of roles with more 'active' abilities. It's not surprising that the roles players dislike the most tend to be those which are the most passive and/or the most reliant on situational factors beyond their control.
While this isn't *quite* in the same category as a purely RNG-reliant role, I feel there is a shade of grey that roles containing an element of RNG can occupy without automatically being considered trash. A good case in point is Crusader, which only has an RNG element as a side-effect to their main ability. I'm actually fairly relaxed about it, as the role is still primarily skill based. Pirate and Psychic cross to the other side; although Pirate isn't technically RNG, as R-P-S does have a small skill element, in practice they act as if its a random 1/3 chance of whether they kill or not. Psychic crosses for exactly the reasons you say - they're far too passive, far too RNG reliant, the skill threshold is too low and, to top it off, they're also OP in their current form.
The Guardian Angel is in all honesty the worst role that's been added in this update, and I still find it hard to believe that anyone actually likes this role. The concept of the role may not seem too bad on paper, but in a real game, it is just utter garbage. Agreed. The underlying concept of a third-party Survivor whose win condition is based on ensuring someone else survives to the end of the game is an interesting idea - and I actually quite like the concept of an anti-Exe in principle. But yes, it's also a hideously difficult one to implement in a workable form, and I agree that the current implementation is massively off the mark. The GA doesn't have the tools to stand up to a determined NK, or a majority-holding Mafia or Town.
That the GA can tactically decide to abandon protection of their target to win as Survivor isn't, in my view, a good thing. Yes, it's technically gamethrowing from their current role win conditions, but since they're almost certainly not a member of their target's faction it would be pretty tough to make that accusation stick. It's in the same vein as the last surviving Town member who technically shouldn't capitulate to the Vampires and win with them by being bitten but, y'know, does so that they convert an unwinnable game for them personally into a winnable one.
Because this alternative win condition exists, GAs are effectively played as Survivors with a random starting alignment that they can abandon at low risk after having confirmed themselves. Irrespective of any other factors, it's clearly not the way the role was intended to work. That adds rock all to the game, and makes for a potentially toxic role - hence why I feel strongly that they shouldn't convert to Survivors if their target dies, even though this does solve some potential issues about their target being unlucky/unfortunate and triggering an automatic GA loss.
I also agree it's difficult enough to keep yourself alive in ToS, let alone a random third party when you can only intervene to save them a limited number of times. My suggestion (+1/2 heals/voteblocks, die from guilt if target dies) converts it from a completely broken role into one which does at least have to stick solidly to its win conditions, makes confirming the alignment of the target role a key piece of information (which the GA can't directly get in a safe way), requires GAs to balance up the pros and cons of revealing and doesn't impose an automatic loss if they die themselves.
Your points around being killed early, lynched, executed and visiting Veterans are completely valid with only 2 vests, but as there's no reason the GA heal/voteblock couldn't prevent these from killing the target, increasing the number of vests should massively lower the risk of those things being an issue. This would also let the GA keep the target alive for long enough that they can have a reasonable shot at working out which faction they need to side with / against. Being dc'd and quitting would be legitimate gamethrows that would prevent the GA from winning, and they would suffer from this far more than other roles, but this is arguably no different from being stuck in a mafia with two poor luck / poor skill / gamethrowing partners. Blocking Exe/Jester from being chosen as the target in a similar way to Exe's not being allowed to have Mayor or Jailor as target would also make sense.
On the flip side, making the GA more powerful would mean on occasions where it effectively becomes a 5th/6th Mafia, or an actively ally for an NK, it would be a rather nasty thing to do to the Town...although, to be fair, it's not as if Town wouldn't benefit from having its win-rate dented a little.
Are these actions worth it to try to save the role? I'm not sure. It would still likely be a royal bastard to win as (it'd need testing), and removing it entirely might be the better, as well as the simpler, option.