Monday, April 12, 2021

How Do You Handle the "Inside" of a Hex?

I have noticed an unspoken disparity in the way people seem to use hexes in the context of a hexcrawl, and I think it deserves some attention. That is: do you bother with precision in the movement that takes place within a hex OR do you treat the space within them as fairly nebulous and concern yourself only with the movement between hexes? I'll explain the difference, and I'll also talk a bit about "sub-hexes" later. And if anyone has seen discussion of this somewhere already, then please point me in that direction.

Version A

In the case of the former interpretation, I'd like to point you to two separate articles that take this assumption and run with it. The first is the famous "In Praise of the 6 Mile Hex" and the second is the more recent "The Properties of Hexes and Mapping," both of which I encourage you to read if you like this sort of thing. But I think I can make my point just by stealing one image from the first article. BEHOLD:

To some people, the utility of this image is immediately obvious. To others, it's confusing what purpose it would serve in play. The second article delves into this kind of math for a range of hex sizes, including 5, 6, 7, 13, 19, 20, etc. That's because the people using this interpretation are actually pulling out the yard stick and measuring the party's precise location within each hex, or they're measuring the exact route they move through each hex to keep a running total of the distance traveled, down to the mile (or half-mile, even). And if you play some of the oldest RPG hexcrawl content, it seems necessary. Take a look at these movement rate charts from OD&D and from the Greyhawk Gazetteer:

As you can see, OD&D was only concerned with your mode of travel, whereas Greyhawk added in the factor of terrain type to modify this. More importantly, both of these systems are really granular. They are interested in measuring your movement mile-by-mile. It doesn't matter if the hex size is 5 miles because you never really move "1 hex at a time" or something. You're always moving 1 mile at a time, and the 5-mile hex is just there to help you measure. So in this interpretation, "hex" isn't really a unit in any game terms. It's presented as such in the OD&D chart, but that's not really what they're for. The hex grid is just an interesting type of ruler that's overlaid directly atop the map to help you measure on it more easily.

Some of you are going "well, duh." Others are like, "what the hell?"

But, see, you need to look at the kind of hexes these folks are playing with. Here's some Dyson Logos hexes:

You're looking at 7 hexes here, but each one is filled with content. So the exact path that the party takes through a hex is important in determining which sites within it they come across. I'm quoting that article about 6-mile hexes here: "From a navigation standpoint pretty much any route through the hex in general is covered. Enter from the vertex and leave through a face? You can approximate it pretty easily." Those sentences only make sense if the way you play hexcrawls involves breaking out the ruler and actually tracking movement within a hex.

Version B

So the other interpretation is that movement is always measured in a number of hexes. It's always a discrete amount, like "3 hexes per turn." In fact, it's usually exactly 1 hex, unless there's terrain or vehicle modifiers included somehow if a bit of extra granularity is needed. And you're also either in a hex or not in a hex, and that's all there is to it. Each is a node and you're always located at exactly 1 node. In this case, I usually find that they set the scale of 1 hex = 24 miles, because that's the distance a human can walk in a day (people walk at 3 mph with surprisingly little variation, and most people who aren't living an unhealthily sedentary lifestyle can walk for about 8 hours per day without getting exhausted). Necropraxis does 1 hex = 1 day of travel, as does Matt Colville. One of the only hexcrawl adventures I've ever played in was Fever Swamp by Luke Gearing, which also does 1 hex = 1 day of travel but the hex size is 18 miles, because it's a swamp and it's all difficult terrain.

But it's not always measured in days. Some games are more zoomed-in and are interested in the hour-by-hour. Both Hot Springs Island by Jacob Hurst and Neverland by Andrew Kolb take place on jungle islands, and they both use a standard of 2-mile hexes which take 4 hours to traverse. But again: rather than having 1 turn = 1 hour and you have to keep track of where the party is located within the 2-mile hex, they instead scale it so 1 turn = 4 hours. Thus, the party is always definitively within 1 discrete hex. And in both of these adventures, the content of the hexes is given in a table of sites/encounters whose precise location is kept vague but are accessed in a sort of abstract way. For example, here's one such hex page from Neverland:

Here's the procedure: every hex has one major feature that it's named for. If you spend a 4-hour turn traveling to that hex, then you will wind up at that feature at the end of the turn. Along the way, the DM can roll for an encounter. They roll 1d4 to determine during which of the four hours spent traveling that the encounter occurs (if you roll a 4 then it happens at the destination), and then they roll 1d20 on the table in the top right to determine what sort of encounter it is.

If, later, the party wants to continue exploring that same hex rather than traveling to an adjacent one, they can spend another 4-hour turn doing so and the DM rolls on the red "D6 EXPLORATION" table in the lower left to determine what they find. That table can also be used to determine where they end up if they get lost. If the party wants to search for something specific or return to a past discovery within the hex, then the navigator player rolls 1d6 + relevant bonuses. The result is applied onto the Exploration Table for that hex. It forms a window of results beginning with the d6 roll and ending with the d6+bonus roll. For example, if the navigator rolls 1d6 and gets a 2, and they have a +3 in Survival, then the two relevant results are 2 (the d6 roll) and 5 (the d6 roll + the bonus). Whatever results on the table fall within that window, they may be deliberately traveled to as the destination of this turn.

Notice how this system abstracts all traveling within a hex with some kind of mechanic that represents movement and navigation. At no point does the player have to actually look at a map, see the sites of interest, and chart their route to get there. At least, not at the sub-hex level.

Personally, I've always been a fan of 3-mile hexes, which can be traversed in 1 hour. That's what defines a "league," actually. After all, since your light sources are measured in hours of fuel, and your Mage Armor is measured in hours, and short resting takes 1 hour, and the limit to how much you can travel before getting exhausted is a matter of discrete hours... why not just make the standard low-scale hex take 1 hour to traverse? 

I also think most board games that use hex maps assume this more abstract interpretation. You'll see it in games like Scythe, Twilight Imperium, technically Settlers of Catan I suppose (in as much as you "travel" in Catan). And video games too! Look at Civilization!

And yet, I've seen this interpretation prompt the "what the hell?" response from people who are used to Version A. I once tried explaining this idea on Reddit and the response was, "do you think it would be feasable [sic] to handle traveling in less than ideal sitiations [sic] as fractioning the total number of hexes you go or penalizing it?" and I'm like, "dude. You go one hex. There aren't fractions." No shade to that guy, it's just a difference of understanding and experience.

What's This About?

Some things are, as far as I can tell, kinda ambiguous about which interpretation they assume. For example, the famous setting/adventure Carcosa by Geoffrey McKinney seems to assume the more granular, precise version of a hexcrawl. But each hex is provided 1 or 2 sites/encounters that are just nebulously located within the hex and are rolled for on whatever turn the party finds themselves in that hex. Hubris by Mike Evans is the same way, although he explicitly recommends that you might play the setting as a hexcrawl and use 1 encounter per day traveled in such-and-such region (even though the map of the setting doesn't have a hex grid). But I guess in both cases, you're just given content and you're told to deliver it by whatever means you want to. What people don't talk about enough is the means, though. Two groups using Version A versus Version B of hex crawling would have wildly different experiences in the same adventure.

I'm fairly certain that most actual old-schoolers use Version A and it's mostly the young people like me who are used to Version B. I feel like most maps and hexcrawl products rely on you using Version A, but I just... I don't like that style of play. I mean, yeah, if you want to play OD&D as written then you need that level of granularity in order to make use of the varying travel rates. That's one of the few variables in their wilderness rules that you can play with. If you cut it out and streamlined it to 1 hex = 1 day or 1 hour or whatever, then there wouldn't be much of a wilderness ruleset left.

But my problem with this is that "distance traveled" is a shitty, boring variable to focus on in a game system. Like, it just isn't that interesting to deal with. Random encounters? Awesome. Weird weather? Affects the game in a cool way. Foraging and hunting? Not always applicable but it creates interesting choices when it is. Getting lost? Depends how you do it, but sure. But measuring the exact number of distance in miles you need to travel so you can calculate the shortest time? Come on. Is that fun? Really, is that anyone's idea of fun? It's usually a 1-person job anyway, and it's not like you actually make much of a choice. You just calculate the shortest route. You're solving for X. That's it. You might be in a situation where you have two possible routes, where one is shorter (3 days away) and more dangerous and the other is longer (3.2 days away) but safer. But unless the difference is pretty major then it usually isn't actually that meaningful a decision. Especially if you're just traveling from one city to another then usually the only meaningful consequence of how long it takes is how many rations you'll need to consume... which is a measure of a discrete number of days it takes, not hours.

I understand that granularity adds realism and the exact terrain type and mode of transportation used was a pretty important factor in ancient warfare, for example. But I just don't think it's actually very good as a game element. It doesn't really create interesting choices after a certain point.

Look, I'm glad to have "difficult terrain" hexes that take twice as long to traverse and also vehicles that let you travel twice as fast, but anything beyond that won't add much more to the experience. Rather, a good travel or exploration ruleset should rely on other variables to make things interesting rather than obsessing over adding more and more detail to just speed.

But I could be totally wrong. Maybe there's a reason lots of people prefer Version A. But I think one of the more important factors to consider here is the role of "sub-hexes" when they're implemented. You can find lots of hex grid templates that include sub-hexes like this:

If you're using Version A of hexcrawling, then the purpose of sub-hexes is once again obvious. They're just a finer ruler to use when measuring distances. Sure, the dimensions of a 6-mile hex are great and all, but if you divide it further into half-mile hexes then you can be even more exact in your measurements. Each large-hex has 12 small-hexes across, and the interesting sites within the large hex can be placed in specific small-hexes for convenience. Thus, lots of folks don't choose one hex size. They choose two. A sub-hex size that's evenly divisible into the larger hex size, such as 6 miles within 30 miles, or 5 miles within 25 miles, or 1 mile within 12 miles, or whatever.

If you use sub-hexes in Version B, however, then you need to describe exactly how they function mechanically, at least in relation to their larger hex. After all, if the sites within the large hex are abstracted in their location or means of navigation, then what are the sub-hexes for? Sometimes I see the only ones actually being mechanically used are the sub-hexes, and the larger ones are just there to help you divide up the map into some manageable chunks that are otherwise meaningless.

Personally, I've always been inclined to do 3-mile sub-hexes and 24-mile super-hexes, so that you can have each super-hex measure 8 sub-hexes across. You can call a 3-mile hex a "league" and a 24-mile hex a "province" if you need to write rules that don't reference exact numbers, such as if you want to apply the same rules to a pointcrawl system or just a freeform map. You can also redefine them as 5 km and 40 km, respectively, and it works out pretty cleanly. The 40 km province will still divide into 8 "metric leagues" across!

When do you know which scale to play at? 24-miles when you're just traveling overland to a known destination, but zoom in to 3-miles when you're doing finer exploration or you're in the wilderness. Correspondingly, you change from 1-day turns to 1-hour turns. What if there are multiple locations or encounters or whatever within a single 3-mile hex? I guess I would use an abstracted system of navigation like in Neverland. I think this works, at least.

In fact, I've ended up moving in the direction of combining a hexcrawl with a pointcrawl in a fashion. Specifically, that traveling by paved roads is done at the "province scale" but functions like a pointcrawl system, where the roads act as a pointcrawl overlaid atop the hex map. Here's an image of such a map I have in progress, followed by a version that only shows the road network:

The first map is more political than geographic, hence the weird colorful-ness and the lack of, like, trees and mountains and whatnot. But more importantly, you can see how the roads and settlements form a pointcrawl overlaid atop the hexes. You still measure turns and actions and whatnot with the hexes (e.g. going from the Crownlands city southeast to the capital of Ironwall is 7 hexes away, and thus 7 days), but you can otherwise treat the travel in much the same way you do when using a pointcrawl system. After all, if you're traveling by road then you likely know your destination up front and where it's located, so you don't have resolve the journey hex-by-hex, day-by-day, turn-by-turn.

But when you're exploring? Don't know your destination? Looking for a lost tomb? Alright, time to start resolving each turn 1 hex at a time. And if you're off-roading, that means zooming in to the "league scale" and looking at the contents of the province up close. One problem? If you do the "zoom in" from province scale to league scale, how do you know which sub-hex the party starts out in within the super-hex? Honestly unsure still.

I will admit, I have misgivings about all of this. It's always a tradeoff, naturally. I'd love to hear others' thoughts and experiences.



  1. I think you have a couple of really good insights in here.

    Your point about travel speed being the least interesting part of travel is a really good one. One very good reason to make travel-times simple is to take attention off the boring part (calculating minute differences in travel times) and focus on the interesting parts (sights, encounters, etc).

    I also think you're on to something interesting when you note that many people have two preferred hex sizes that they like to switch between.

    I am very much in the Version B school of thought you described. I think the two purposes of hexes are to regularize travel times, so that each hex represents 1 unit of travel time, and to demarcate locations that take no travel time to move between. If two locations are in the same hex, it means that you can pass freely between them without "traveling" the way you do to get between two locations in different hexes.

    Because of that, I would say there are only two reasons to worry about traveling inside a hex. The first is if the hex contains an outdoor adventure site, akin to an open-air dungeon. In that case, as you note, the solution is to zoom in to a smaller scale map of that adventure site. The second reason to worry about traveling inside a hex is because you've made your hexmap at the wrong scale and need to either make a new one or rethink your travel rules.

    The "size" of a hex is a bit like the "length" of a turn. They're fictional numbers that are mostly there to give us verisimilitude. The size of a hex is however far the characters can go in one travel turn. And the length of a travel turn is based on how many "events" you want to have per adventuring "day". Do you want to see one location and have one encounter per day? 2-3 per day? Once per hour, up to 8 times in a day? Those, I think, are the most important considerations. The numbers are whatever you need to say they are to convince yourself the characters aren't moving "too fast" or "too slow" based on how many hexes they cover per day.

    1. What's funny is that, once you're already working at the 3-mile scale and you then arrive at something like an outdoor adventure site, I wouldn't even bother "zooming in" to an even smaller scale hex map. I'd just describe the location up close, the way you begin describing the dungeon once the players arrive at it. At that point, NO granularity of measuring distances and "travel time" is gunna be meaningful enough to care about.

      And as for what you say about pacing, I think something I probably should have talked about was the difference between systems that let you go 1 hex per turn vs. multiple hexes per turn, regardless of the hex size. For example, I once heard somebody say they hated 24 mile hexes (1 turn = 1 day) because it makes it feel like you're moving at a glacial pace across the campaign map. They'd rather have 6 mile hexes but STILL have 1 turn = 1 day, knowing that they'd go the same "real" distance across the map, but they'd FEEL like they got to do more that turn. They picked a route that was 4 hexes long and felt just a little more mobile. Maybe there's something to that.

  2. I really like the hex-point crawlmaps that you made. They look evocative and seem like they'd actually be pretty useful.

    I've taken a pretty different approach recently. Probably more akin to your type A, with a small number of super detailed hexes. My game (that's been running for ~20 sessions) has 9 "hexes". Here's what the interior of one hex looks like:

    The overall map:

    I'm curious how you use your political hexmap. Is it player facing?

    1. Thanks for the kind words. BTW I really liked your map you made of Cove Town, I used it as a good-ass example in a post I made a while back.

      Going by what your image looks like, I think it's arguably a third take: the inside of the hex is neither abstracted nor is it mapped in precision. It's just a pointcrawl at that layer, which is interesting since that's almost the opposite of what I was thinking I'll be using (i.e. low-scale is hexagons and high-scale is mostly pointcrawl).

      The reason my hexmap is set up for politics right now is because I wanted to prioritize having the factions figured out before anything else, like the 7 Kingdoms of Westeros. It could totally be player-facing, since I really like transparency and taking steps to get players thinking the same way the DM is thinking.

  3. The hardest question I always have with hex crawls is whether the map should be player facing or how to deal with the "fog of war". Do I make a detailed hex map and then hand them a blank one? I haven't played with anyone who enjoys doing the work of filling it in as we go. Also, in my limited experience, they will fill it out wrong. Do I fix it?

    The difficulty is that the GM is the "world around" and a person navigating in the wilderness needs many many touch stones to actually navigate. That means tedium for both parties (A) or abstraction (B). I think abstraction is better. I also think octagons or squares are a better shape for this. Probably squares. players in the wilderness are going to move based on landmarks or cardinal directions and hexes are neither of those.

    All that said, I liked some of the alternative hex sizes you mentioned. 4-hr hexes giving 2 hexes travelled per day and the option for 2 more if you travel through the night for example. You also get the option of choosing where you sleep since you can stop in the first hex or push on to another. that's a potentially interesting decision, which is what we're always after.

    1. I guess all I can say is "it depends on what you're using the map for." For example, if the adventure is just going from Point A to Point B along a linear, known route, then you don't need to ever show them the map. Instead, they can travel using an itinerary, like actual medieval people did (i.e. an ordered list of stops and landmarks they should follow to reach their destination, like printed Google Maps directions).

      But if the adventure is exploration focused, then they might need to grit their teeth and fill out hexes on a map. That said, it's still good to give them a partially-filled out map to start with (maybe include the major settlements, or the hexes that the major road passes through, etc.). And also, have mechanisms in place like a mountain/plateau hex they can go to which will let them see everything around them in 360 degrees up to two hexes away, letting them fill out a huge chunk of the map at once.

      You can also give a large-scale map that has large-scale detail but then have the adventure take place on different "zoomed in" parts of it which the players don't have as much detail for to start. For example, the big map is a bunch of 24-mile hexes and the players have a decent map to work with, but when they have to find the lost Tomb of Baron Slime and it's been narrowed down to one hex, it's time to zoom in to the 3-mile hex scale and search within that 24 mile area.

  4. I see one advantage to method A. You can make your hexes as big or small as you like (or is apropriate) with out having to come up with rules how fast you move through them. You have an overworld map with 24 miles hexes, but an local adventure with an map with five mile hexes? No problemo, you just can use the same rules.
    While I am personally agree with you that method B offers the nicer game experience I see how method A is more scalable in an mix and switch approach to DMing.