I posted a quick collection of thoughts regarding Dragomeet 2016 over on my G+ page but I wanted to give the Project Cassandra feedback a post to itself.
The setup for the game was the same basic questions that I used during the Stratigicon playtest, that Apollo 11 had discovered something and President Nixon was due to be assasinated prior to announcing the discovery. The discovery this time, shapeshifting aliens on the moon who had already infiltrated the USSR and were now trying to take over the US government. The assasin was non other than the vice President who had already been replaced. Thrown into the mix was a Soviet Null, immune to the powers of the party (but not, as it would turn out, regular bullets).
The adventure went by fast. Too fast, as the players blew through every challenge with ease thanks to a combination of difficulties that were too low, great teamwork and some amazingly inventive use of Powers.
The primary feedback was pretty unambiguous, actions weren’t challenging enough, which resulted in the group not failing for the first two thirds of the game. This had a knock on effect in that it prevented other mechanics from coming into play, namely conditions and spending premonitions on re-rolls. It was only towards the end of the game that I started raising the difficulty that we started using all of the mechanics but by then it was too late to have a major impact.
The second negative was with the skills themselves and compounded my GM error with the difficulties. Essentially the players found that they either had too high a skill level for a given task or were lacking the skills entirely (and therefore didn’t attempt actions). This is something that has come up in playtests before and I had hoped that the current skill list addressed it. Unfortunately this does not appear to be the case.
It’s a blow to get this close to having the game finished before running into a major issue but on the other hand I’m glad it went badly. It’s the first major catastrophe I’ve had but it also feels like it is workable rather than an impassable issue.
Beyond this the feedback was quite positive. The players enjoyed the scenario and being able to influence it through the questions plus liked the fact that the powers weren’t mechanically constrained to prevent them being overpowered. As these are all aspects I’d put thought into I’m glad that the players picked up on my design aims. Following the game I’m also convinced that the single scenario design (of saving the President) was the right approach. Once the game is finished I may add a bonus sheet on running additional highly defined bonus scenarios (I already have ideas for one based around the Berlin Airlift).
I’ve been pondering how to change the skills since Dragonmeet but before I dive too deeply into it I thought it important to look again at the probability tables, the results of which are plotted below. The x-axis plots the number of successes and the y-axis the percentage chance of rolling at least that many successes for a given skill level.
Looking at those numbers it’s clear how off my perception of the difficulties was. With a skill level of 3, which the players were regularly achieving, there is still a 66% chance of rolling 3 or more successes. In my head 3 successes should have been difficult and definitely not in their favour so often. Those odds rises to a staggering 90% at a skill level of 4 and drops to 32% for a character with a moderate skill level of 2.
It’s clear therefore that the first thing I need to do is adjust my idea of difficulty levels and then add explicit descriptions to the game. My current working template is:
1 – Trivial – Only worth rolling if the individual is unskilled
2 – Normal
3 – Challenging (with intention of this being a typical roll for the game)
4 – Hard
5 – Heroic
6 – Impossible
The second approach is a limit on the maximum skill level of a character. The players at Dragonmeet suggested setting it at 3, which I’m thinking of implementing. At this level a skilled individual will pass a Challenging roll most of the time but still fail at a noticeable rate.
The biggest change, which I’m still working on, are the skills themselves. Under the current design players add up related specialities to get their skill level and tend to either end up with a high level or none at all. If I keep the current system the specialities need completely rewritten to provide a wider breadth skills with only a small number that overlap enough to give a high skill level.
The alternative is simply list a set it skills with a rank by each of them. That has the advantage of simplicity and also makes it easier to deal with edge cases as I can give each skill group a rank for when no specialities apply. For example using brute strength might just fall under the general umbrella of the Physical skill set at rank of 2 for the strongest and 0 for the weakest. The downsides of this approach are flavour and rigidity so more thought will have to go into it before I settle on one over the other. I may also split the skill groups further by adding Social to the existing mix of Mental, Physical and Specialist.
All in all I’ve got a lot to think about and a valuable learning experience for future projects.