D&D General Setting a campaign on a river

[Yora]

I think that's possibly a good idea, though you might be able to introduce some "succeed but" or "succeed and" consequences if you want to have them roll but spice things up. For example, if they fail DC 10 they still get food, but you roll a random encounter. If they exceed 15/20, they get food but they also get valuable pelts/herbs/poison frog venom they can sell downriver.

At the very least, automatic success could still require the normal time for foraging. While it wouldn't lead to situations where PCs are lost in the wilds without supplies and have to go on an adventure just to get some food, it might still end up as factor for tension for groups that are being hunted. In that case they might not really have much choice and just accept that they will have to deal with an additional random encounter check that might lead to their pursuers finding them, instead of trying to press on with penalties for starvation, but it still becomes a factor in the story.

I'm not sure how common bad water is in wild rivers with heavy circulation, but introducing the idea that PCs have a small chance to get sick if they drink the "default" water instead of foraging for "premium water" could be an interesting addition to river travel.

 

[Yora]

A few observations I made over the last days that might be useful to some.

When creating a river system for a West Marches game, there will likely be a lot going back and forth from a starting town, and there will be much more adventures taking place close to that starting town than further away from it.
The river that players are going to explore can't really be a river that is also a major trade route. It seems not very plausible that you would have large numbers of relatively easily accessible ruins filled with valuables right next to busy traffic. The starting town should be at the very edge of the sandbox, and at the end of the regular river traffic route.

I felt very uncertain about the scale of the river system I was making, as I was putting stuff more clustered together near the starting town, and more spread out the further away you go. But that resulted in adventures to more distant and higher level ruins being very long, and getting increasingly bigger numbers of random encounter checks along the way.
But I realized that this isn't that much an issue when you let higher level parties have access to faster boats. Paddling in a small canoe is not that fast, especially when you go against the current. They are likely the only thing that new characters can afford, and its not an issue when their adventures are short and close to home. Upgrading to faster boats can be a form of progression that makes journeys to more distant places much faster and more viable. It's like an early non-magical version of getting access to flying carpets and teleportation and such.
In BECMI D&D, a river boat has twice the speed of a canoe, and a sail boat twice as much again. (Though oddly, the river boat costs twice as much as the sail boat.) In 5th edition, there's the keelboat, rowboat, and sailing ship, but their speeds seem very unrealistically slow. (I think they should be double what's shown.) Upgrading to a faster boat should feel like a milestone, and I think they should be expensive because of it.
I do random encounter checks based on time, and not by distance traveled, and doubling the speed means cutting random encounter checks in half.

I also got another issue all of you could help me with: What would be good ways to distinguish settlements as they are getting smaller and more remote in regards to what services they offer?
NPCs with healing magic would tend to be found with the more powerful ones in the larger cities near the coast, where the big temples are, and less powerful the more out you go into the wilderness. Treatment for the more serious conditions becomes harder to come by as the party goes into more dangerous areas, and might even force some PCs out of action for an adventure or two as they get passage from the forward base camp back to a bigger settlement closer to civilization.
The other thing would be shipyards. You can buy two new canoes in every settlement with some kind of market, but for a nice sailboat you need a larger town, and big cargo vessels could only be bought in the big ports at the coast.
I am considering the idea that the further you get from civilization, the worse the prices get that traders demand and pay. They have additional costs getting their goods, so their stuff is more expensive. And being the only trader in a hundred miles means their customers are unlikely to make a detour to get better prices somewhere else. But I'm not sure how to make that in an easy to handle calculation factor. (With XP for treasure, players should always get the full value of non-coin treasure, even if they can only sell it for less.) I guess a smaller inventory in heavy armor on the frontier would also make sense, as there's little demand. Unless someone where to have his armor destroyed by slimes 500 miles from the next armor smith.

 

[Yora]

I got an idea for including navigation in a river campaign.

Making a complete map of an entire river system spanning hundreds of miles with all its little side arms really isn't practical. You could theoretically let players give it a shot on a blank hex map with very small hexes, but I think that would be very tedious and not feel like it reflects the kind of maps actual river explorers would be using for their notes.

Instead, I want to go with an entirely skill check based system to navigate through the networks of small side branches that fork of from the main waterways that are depicted on the main overview map. My own GM map only shows branches up to the third order, and I intend to let players find their ways on those without navigation checks. It's only for the rivers even smaller than that that this system comes into play.

Maps are items that characters can find or sell that have instructions on how to reach certain hidden places from an easily recognizable and unmistakable landmark. Every map has a dificulty based on it's quality. Using a very good map is an easy task, while using a poor quality is a very hard task. The difficulty is further modified by how far the destination is from the clearly identified reference point on the main rivers. Since I have all my travel times in increments of 10 miles, (1 mile per hour times 10 hours per day), I increase the difficulty of the navigation check by +1 for every 10 miles that you try to follow the map.
If the navigation check is a success, the party reaches the destination in the shortest time possible given the distance and their travel speed. If the check is a failure, they still get to their destination, but for each number that the check fell short of the difficulty, the travel duration is increased to require one additional random encounter check. I do three random encounter checks for each day of travel, plus one check per night. So missing the difficulty by three adds a whole day on the water searching and backpaddling, and you also get another night to rest and potentially have another encounter before you arrive at your destination. Since I usually have random encounters at a chance of 1 in 6 for every check, getting two or three checks added to the journey generally shouldn't be much of a problem. But for journeys deeper into the smaller rivers, having someone with a good navigation skill and paying for high quality maps can become really appreciated.

The fun part comes with the additional use for navigation checks to make your own maps of the unknown rivers you explore. These maps can be very important if you want to find a place again after having left it, and can be sold to other characters. To make such a map, a character makes a navigation check. The quality of the map and the difficulty to use it depends on the result of the navigation checks. For Wiorlds Without Number, I've decided to make it 20 minus the navigation check result, with the minimum difficulty being 6.
WWN makes skill checks with 2d6, so I think it's a great idea to let the player roll one of the d6 either open or in secret, and the other d6 gets rolled by the GM. That way the player has a clue for the final quality of the map, but can not be certain how accurate it really is. The ultimate difficulty for using the map remains secret for the GM, at least until the players trying to navigate with the map have reached the destination and will have found out for themselves.

The one question that remains open is how players would discover unknown sites by going into these small rivers blindly. I guess one approach would be to simply roll a d20 and the result is the number of random encounter checks until the party finds either a small randomly generated site or a larger site whose exact location on the river has remained undefined until a party randomly discovers it. Since you might always need a monster lair or pirate camp if players try to track randomly encountered enemies back to their hidehouts, it's a general good idea to have a couple of those ready at hand anyway. And players can be required to tell the GM that they plan to go on a random exploration a few days before the game.

 

[Eltab]

If you have money-minded players ...
THE Outlander background "map in my head" feature
PLUS Proficiency in Cartographers Tools
PLUS a scroll case and 20 pages of paper
PLUS trained in Survival
EQUALS that really good map (with the best skill check bonuses)
Now for sale at Adventurers Hall to the highest bidder!

Or adventure plot: kidnap the NPC described above and get him to make you a map of the way to the McGuffin.