D@W Campaigns: Playtest Thoughts

This weekend was the fourth time I've run something out of the pre-release rules for Autarch's Domains at War, and the first time I've tried to use the operational-scale movement system from Campaigns. While I thought all the rules for the tactical system all worked fairly well, I still have some reservations about how the movement and recon system is being implemented.

Some aspects of the system are satisfying. The quick resolution system for combat was straightforward, and resulted in some nontrivial tactical choices. And it was fun seeing the invasion of an enemy domain trigger a morale check that resulted in an unexpected peasant uprising.

Here are a few criticisms:

1. The effects of splitting armies aren't well-described. The rules mostly assume that each player runs a single army, with a single general. There's one mention of the option of breaking an army into divisions to prevent from suffering the size-based movement penalty for marching with a very long column. The only advice given there is to treat each division as "an army, for movement purposes", which I guess means that it continues to be a single army for initiative, supply and recon purposes (?).

But there are many other reasons to want to break up any army into smaller components and move them separately. Cavalry can cover much more ground, and effectively run away from almost anything that chases them. Arriving on any new map, the logical approach is to scout it out rapidly with multiple cavalry forces. This creates lots of little fast-moving army groups, and makes it obligatory to resolve action on the day-scale, instead of the week-scale.

Do they each get separate recon rolls? Separate initiative rolls? Do they form their own individual vanguards and rearguards? The rules are mostly silent about this. If the answer is "well, obviously yes", then it feels almost mandatory to break down every army into its component divisions, to give them all lots of extra recon roll chances and maximize movement flexibility. This is bad, because...

2. Recon rolls are extremely time consuming. On an average map, there are probably a dozen domains within four 24-mile hexes of any army. This means that any time a player army moves, the referee is effectively making two dozen rolls behind a hidden screen. Each roll has at least five different modifier types applied to it, and they all need to be calculated separately. This amounts to about two minutes of furious dice rolling every time a player finishes moving. A player move itself rarely takes more than a minute, so the player-to-GM time ratio is something like 2-to-1. That is, for every 20 minutes of actually doing stuff, the players need to spend 40 minutes listening to me mutter at dice!

This would be fine if the recon rolls routinely had an interesting effect, but instead...

3. Most long-range recon rolls repeatedly generated the worst possible outcome. With a small forward cavalry force roaming around the board for other similarly small domain garrisons at the maximum possible range, the most common situation was a -1 or -2 penalty for size, and a -3 or -4 penalty for distance, and often also a -1 for terrain. On a dice roll of 2d6, this results in the average roll of "7" being converted into a 2, and anything lower becoming less than a 2. This is supposedly a recon "Catastrophe". In the rules-as-written, this is described as a situation where the judge is supposed to generate a "fake Major Success" result, pretending that the result is a brilliant recon breakthrough and reporting lots of false information.

But the idea that a modified 2 could be "disguised" as a modified 12 to fool players isn't very plausible. If you have any negative modifiers, a modified 12 ("Major Success") is totally impossible, even as a modified 2 becomes quite likely. So all of those supposed Major Success results (and you'll get a half-dozen a turn, from all those enemy armies at max range!) are instantly recognizable as fakes. I pretty quickly realized I would need to just replace them with a neutral "nothing found" result.

Even with that, the bottom line is that there's ton of rolling (while players sit around doing nothing), but virtually nothing happens from it.

4. Small groups keep passing like ships in the night. Even when directly adjacent, there's still fairly high probability of small scout forces (of a company or two) getting several "Marginal Success" results in a row, due to the -2 modifier for size plus a -1 modifier for woods terrain. A "Marginal Success" only gives a rough estimate of enemy position. This results in scouts circling one another, but never finding exactly where to go to initiate attack. In fact, I had armies literally running through one another's hexes multiple times. The rules only specify that they fight a battle if they end movement in one another's hexes, so the battle would never actually happen until someone got a "Success" (which for me usually required a natural 10). I regard all this as an unfortunate artifact of timescale granularity.

This is a subtype of a larger issue...

5. Lots of dramatic events are hidden from player observation. In a standard RPG, the enemy forces are constantly moving in plain sight, causing dramatic tension even as players are waiting to take the next turn. Will the giant spider fail its attack rolls? Will everyone make saves against the fireball? But in a double-blind system, everything is under the fog of war. That means that the tension of having lots of armies creeping around a map is mostly lost, aside from occasional sonar pings that amount to "something is somewhere, but you don't know what or where".


Next up, I'll try to make some suggestions for how I'd modify the system to reduce the severity of some of these issues.