So, Dan has been casting an eye over my Monitors draft and pointing out a few issues. One of the things that came up was the use of stats to handle "things like injury and damage", which Dan interpreted mostly as a combat system, but this wasn’t quite what I had in mind.
That's all sounding a bit coy, so let me explain a bit. Monitors, after all, was originally a preposterous example for use in my inconclusive series on soft attacks, so the ideas I have for it have always included a reasonable amount of soft attacking. Once I'd arbitrarily picked a Stats + Skills mechanic, and was moving on to think about things like Hit Points, it occurred to me that it might be possible to model both hard and soft attacks through the use of stats by making stats degradable. The first draft wouldn't work for this, since I didn't get round to revisiting my placeholder stats after having it, but let me just explain the idea.
For example, let's go back to my old pal Xerxes and his flash pistol. If it inflicted damage to a target's Perception score, this would immediately impair any skills that depended on that stat. Similarly, a tranquiliser drug could damage Reaction, psychogenic gas (or alcohol) could affect Mind, being hit with a club could damage Toughness. On reaching zero, you can no longer do anything reliant on that stat (otherwise very skilled people could see you even though their eyes don't work, and similar confusing things). You could also have a threshold below which secondary penalties kick in: maybe when Perception hits 2, you need to roll your now-reduced Spot whenever you need to see what you're doing.
Obviously, for something like this to work you'd need to tweak the stats and damage inflicted accordingly, and then look at the stat-skill interaction to make sure it strikes the balance you want. It also has the drawbacks I mentioned when discussing soft attacks, specifically the soft hit points model: notably it involves tracking several pools over time, and you need to work out appropriate thresholds for any qualitative effects.
Heading down a bit of a tangent, a carefully-balanced system could potentially overcome the qualitative threshold issue, by using large bonuses and no critical failures for most rolls, and assuming rolls are happening in the background but automatically passing.
(numbers in this example are arbitrary and don't reflect plans for Monitors)
Let's say we have a large sitting-room containing an intruder, a crab sneak-thief. It's daytime. The home's owner, a parakeet, suddenly walks in.
- Parakeets, like most ordinary civilian, have Perception 10.
- As a civilian with no relevant experience, the parakeet has no Spot skill (+0).
- Seeing a medium-sized intruder in a sitting-room in daytime is very easy (+10)
- Rolling on a d20, the parakeet needs to equal or beat... 20, so there's no need to roll. She sees the intruder, and reaches for the poker.
A mole, in contrast, might have Perception of only 2, and potentially miss seeing the intruder at all, even standing in front of white curtains in a lime-green jumpsuit.
Similarly, an owl (Perception 15) can reliably see the intruder if he hides behind the curtains (easy +5). However, if the crab lets off a flash pistol (6 Perception damage) and then hides, the owl will find it harder to track him down. After a couple of hits, the owl will really struggle with even simple tasks.
Obviously, this could work just as well with a statless skill system, and in some ways better. Blinding attacks, for example, make more sense if they damage a Spot attribute than a Perception stat that also affect hearing; and it would avoid confusion where someone's stat is depleted but they have a very high skill. However, this would be unwieldy unless the attribute list was quite small.
In any case, I'm going to revisit the mechanical side soonish and, as I mentioned last time, "go entirely the other way and cut things down to a small number of broad skills...just to see what happens".