Apropos of playtesting, here’s a case study of the biggest element that changed in Ashen Stars. (By the way, we got the 200+ Stellar Nursery pre-orders we needed to fund the print run. If you haven’t grabbed your advance goodie bag yet, you can still help put us all the way in the black.)
When Simon gave the greenlight to the original Ashen Stars pitch (then called Bleeding Stars, later Into the Bleed), he asked me to consider an idea he had for ship combat. Each member on the crew would have something to do during space battles. They’d also be competing for system resources, facing a temptation to route power from other stations in order to do well themselves. Along the way we dropped the idea that players will be at odds with one another because it just doesn’t match what they want to do during a joint fight against an enemy. Their instinct, and rightfully so, is to cooperate. I could have designed some sort of kill-stealing mechanism, where there is only so much glory to go around, but there were already better systems in the game that pitted individual selfishness against group imperatives. This one we could afford to let fall by the wayside.
My first go-round at the first idea, that everybody has something to do in space combat, took the idea too literally. Each player had something different to do in every single round of a ship fight. This led to a long, hard-to-remember sequence of events. Mechanically each stage worked differently, with varying levels of influence on the outcome.
In playtest, both in-house and otherwise, we found that most (but not all) groups had problems with the complexity of the system, and the low payoff for that complexity. Although ripping the system out and plugging in a new one from scratch was more work for me and would require a more involved second playtest round, it had to be done.
The replacement system took a step back from trying to simulate the jobs of the various crew positions. This is what the first system should have done, as GUMSHOE in general emulates genre structures and doesn’t attempt to simulate imaginary realities. It became more abstract, and more like hand-to-hand combat. Now each player is the focus for a single exchange, just as they are for an attack in a fight between characters. The final version feels tactical, because the group together decides which enemy sub-systems to degrade when they hit, and what order to use their ship’s weak and strong suits. (This is the brilliant stroke of Gregor Hutton’s 3:16—it feels like you’re making tactical decisions, although you aren’t, particularly.)
It would have saved me time to arrive at the final version out of the gate. But I wouldn't have gotten there without having done it the obvious and mistaken way first. Only by cutting through its complexity to something simpler did I get the more original and more enjoyable space combat you’ll find in the published version.