After finding the book in my local game store, and having the game recommended by an acquaintance, I got inspired, recently decided to set up a Torchbearer campaign with IRL friends.
I got quite hyped (part of the problem, maybe) about the game and put quite a bit of effort into preparing it, going to the lengths of creating cards, and various table-tools and cheatsheets to deal with the rules complexity.
Despite the level of effort put in, it took barely two sessions for the enthusiasm to deflate and settle into disappointment. I credit the learning I have done at this site for being able to identify this as early as I have.
The text sells itself as an intricate gearbox that once learned promises to deliver amazing gaming moments — my experience is that most of this is false advertisement.
In practice what I've noticed is that the intricate rules require a lot of text-referencing, since precise rules exist for any situation, while leaving very little to the GM's interpretation. This is not bad in general, but in this case playing the GM-role felt like being a passive executor of the mechanized wankery — at the same time demanding and boring — with little space left for GM-agency.
At the same time, there are moments where it seems that the system does as much as possible to avoid constraining outcomes — which is resolution mechanics ought to do. In particular, the conflict system has turn-by-turn clever double-blind action selection system, but all it manages to achieve is to prolong resolution, without having a significant effect on conflict outcome, or even on narration — we adapted to barely describe actions to keep the conflict time manageable.
In the moment where outcome authority had to be applied, I felt completely abandoned by the system, which for simple rolls offers a fail-forward system that allows the GM to choose between giving a condition and allow the roll to succeed on failure, or generate a "twist" and allow it to fail. On the other end it seems to encourage players to attempt to pile as many dice as possible on any task to ensure they pass it. The result is that failure often becomes something unexpected and undesirable, and that as a GM I felt often compelled to rob the rolls of their meaning, lest I break the finely-tuned machine.
For extended conflicts, there's a clear winner, but the system offers a sliding scale of compromises based on how many disposition points (conflict HP) were left for the winning side, with very little constraints — conflict outcomes are explicitly negotiated between GM and players.
The most bizarre rule was that in Kill conflicts, compromises include people dying on the winning side — with absolutely no constraint to determine who dies! So, the GM and players need to negotiate on who desrves to die more, with the player "captain" having the final say. I wonder if this rule was even playtested, it seems so bizarre and we had a lot of trouble applying it.
The Belief/Impulse stuff also felt kinda carrot-and-stick and players didn't really engage with it at first, only starting to do it to get the brownie points they needed to spend to do cool stuff. It felt forced.
The entire system felt to me it's an intricate set of levers designed for the GM to make play run "smoothly" towards the desired game experience — essentially more of a widget than a game. I felt I couldn't contribute significantly to the game as a GM, and despite having all the power to walk PCs anywhere I wanted, I was singing my own song.
That said, there are a few things I like — how skills improve by using them and characters end up reflecting the things that they have done in past sessions. But they's really minor details.
Is this a review of the game? Not really. I'm open to considering that I've completely misunderstood it. But I'm only going to tolerate this kind of play for a few more sessions — I can already see my gaming group is getting tired, compared to our successful experiences with The Pool.