OrcaCon 2024

Another year, another OrcaCon. Same great taste, lovely delicious food trucks.

Just so you know: This post is mostly a deep dive into the first-round development of Cold Comfort, with some scattered notes about other things. 

If you want the notes on Cold Comfort to make sense, you might want to read the current beta rules.

Background

OrcaCon is a small Seattle gaming convention, which I have attended faithfully for years. Many Crab Fragment games got their first exposure here, including Vines, Shipwrights, and Powderkeg. I remember playtesting Capital City here also. Cold Comfort was brand new this year, and it was my obsession all weekend.

Cold Comfort is a simple area-control game set in the Yukon Gold Rush. Players build hotels, entertain customers, and dominate neighborhoods, earning money and points. This much is fairly certain. But the details are a slowly unfolding mystery.

I came up with the idea in late November, and put Cold Comfort through six revisions in December, hosting three game nights at my house including one on Christmas Eve. At OrcaCon I was still figuring out which management actions were critical (building hotels), which ones were less useful (special abilities on characters?) and which ones were just a nuisance (re-rolling neighborhoods?)

One mechanic that I assumed was critical was that every time the city grew, random customers would show up and check into random hotels. There were other random events too, like rolling a die when you went digging for gold.

Because of all the randomness, I felt like there ought to be a balancing mechanic, specifically one where the players can gang up on the leader to bring them down. This is also referred to as a “catchup feature,” and even though I generally despise these, I spent the weekend trying to find the right one for Cold Comfort. Spoiler alert: I didn’t, and I still hate them.

Catchup Features

“Rubber-banding,” “Catchup Features,” “Balancing Mechanics.” Game rules intended to fix the disparities caused by random events. Randomness can hurt games by giving more “win” to one player than another, and catchup features are supposed to bring everyone closer together again.

But here’s the question. Do you want your catchup feature to completely level the playing field, erasing any gains that the leader might have made, or just make the scores look closer, without actually changing anything? Hint: Neither of these results is worth it.

Often a catchup feature is presented as a gamble, and these types seem to feel the best. If you’re losing, you can take a wild chance and maybe surpass the leader, but otherwise you will be even farther behind. But mathematically speaking, the designer still has to decide whether the odds favor the winner or the loser; on average, this is going to be the same as handing more points to the player in last place.

Now, giving players actual choices and control is not the same as a catchup feature. This is just a game with a blend of luck and strategy, where skilled play has a chance to overcome bad luck. I’m talking about a specific rule that looks for the winner and takes something away, or looks for the loser and gives them a boost.

They can be well-hidden in the mechanics of video games (everyone cites Mario Kart as a good example of rubber-banding), but in tabletop games they just feel like lazy design. What’s worse, lazy catchup features can create perverse incentives. Imagine a racing game in which, each turn, the racer in first place moves back two spaces, and the one in last place moves forward two spaces. Do I even want to be in the lead?

I prefer to balance highly random games with frequent hard resets. For example, shuffling the deck after a hand of poker (this resets the components, though not the score). Or I’d look for the source of the randomness and weed it out, root and stem, so that when a player falls behind it’s their own fault.

The other variety of balancing mechanics, also sometimes lumped into the “catchup” category, are systems where players are allowed, and even encouraged, to attack the leader. They might be asked to help the players in last place as well, though I can’t think of any examples of that. Usually, these “targeted damage” mechanics require the players to figure out who is in the lead, and then to attack that player, sometimes at their own expense.

Poker doesn’t have targeted attacks, thankfully. But plenty of other games do. And so, with this in mind, I spent a good deal of the weekend trying to find a targeted-damage mechanic representing “fire” that would balance Cold Comfort. Because if Gold Rush boom towns do one thing consistently, it is to burn down.

I can’t say that I succeeded in this quest, but I did fail in a productive way. And as I have discovered many times before, the randomness itself was the problem, and without it, the balancing mechanics aren’t so necessary.

But worry not, we shall always have fire.

Friday

Before setting out for OrcaCon, I packed a bag filled with test-ready games, though I had only three main goals for the weekend: to play a few games of Shipwrights in its “final” state, so I can send it away to DriveThru for proofing; to play a few games with the Paradise Art of the Island Deck, testing the new graphic design; and to play as much Cold Comfort as possible, prepping it for open beta.

I arrived at lunchtime, and immediately ran into my stalwart testers Lauren and Shel. We played Cold Comfort with various experimental fire rules, stopping and starting and discussing the rationale behind each change.

The game had a potato salad of mechanics derived from a single issue: When new customers roll into town, they choose a hotel to stay in. How do they decide between similar hotels? Our working solution was “walking distance,” which leads to the question of “but from where?” and also “what if that is tied?”

We used a permanent center point as the origin for the new customers, but this caused the hotels to cluster in the center. New map cards were added on the outskirts, but since new customers always came from the center, players preferred to build their hotels close to the center and ignore the outlying cards. 

This made the newer parts of town feel like a wasteland, so next we experimented with a mechanic where fires started at the center of town, just like visitors. So if you build on the outskirts, you may get fewer guests, but at least you’re safer from fire.

This made sense in theory. But in practice, placement of guests and fire was so unpredictable that this extra rule was almost useless. Guests filled the closer hotel rooms first, and then the only empty rooms were on the outskirts, and the fire found those hotels almost as easily as anywhere else. 

We talked about different ways to relocate that origin point, so that customers spawned from different places throughout the game. But this seemed a lot of effort since the only real value of counting distance from the origin was to break ties between similar hotels. And making the origin point unpredictable did not make it better.

Sometimes game designers can spend so much time fixing a tiny part of the game that they forget about the rest. And then after all that work, the other stuff becomes harder to change, because we’ve spent too long building up a web of mechanics to get around the core problems.

As a palate cleanser after Cold Comfort, I broke out Shipwrights of Marino. I’ve recently nerfed a couple of the cards, those that let you buy stuff as a free action, so they pay about 5 gold less. They seem decent at this value, and I’m happy with the result. Hopefully you will see a final set of components, and a DriveThru version of the deck and board, in a few weeks. It was quite a relief to work on a “done” game for a change, after wrestling with Cold Comfort, but soon I would be back in the thick of it.

I played Cold Comfort with a different group, experimenting with new fire rules and trying to test the balance of the character abilities. I think I’m pretty terrible at coming up with character abilities: I can’t find one thing each character can do that gives it personality, makes the game better, and isn’t over- or under-powered. Each special power feels like an iron weight forcing you to play a single strategy in what could be otherwise (arguably) a balanced game. 

So for the time being, and perhaps forever, we started ignoring the character powers. And then I went home and printed new cards.

Saturday

The day started with a mix of smaller games. I started with a pitch meeting with a local publisher, who wants family games that look good on video. I have a few ideas for that, but in general I don’t think most of the current Crab Fragment catalog makes very interesting videos. I know this having reviewed hours of extremely boring B-roll.

I played Cold Comfort with Lauren and Shel, and we talked about the idea of various “character dice,” special dice that did different things when they came into play. The Mayor could be a die that grows to a 6 when he lands in your hotel (giving you more scoring power) but he doesn’t pay when he checks out. The Arsonist gives you the power of fire - when he checks out of your hotel, you can put a fire token anywhere you like.

These mechanics were fun, but they didn’t really provide the balance that I was looking for. If fire only comes from the Arsonist, then we have to wait for access, and sometimes it ends up in the hands of the leader, who will just waste it hurting the person in second place. And even when a player got control of the Arsonist, they had to use it right away, and sometimes the moment wasn’t right (the right hotel might not be empty, etc.). So it started feeling like too many rules for too little value. But at least we were learning.

After that it was lunchtime. I tried a frybread burger from the Native American food truck. It was strange because the frybread was fresh out of the fryer, so the bun was hotter and greasier than the burger. Delicious, but wow.

After lunch I pulled out the Island Deck and played Down the Well and Powderkeg, mostly to check that the new Paradise art was legible. It’s hard to playtest the legibility of graphics directly; if you ask people to remark on the graphics, they will put on their critical hats and find flaws for the sake of finding them. It’s often better just to teach them the game, and observe if they are having any trouble reading the cards. Try both methods (asking and not asking) and you’ll see the difference.

I also played a round of Linos, because it’s a nice looking game and I like it. I’m not really testing anything particular at the moment, just keeping my eyes open for ideas. I recently printed business cards for Linos, pointing to the web page, to hand people when they see me playing in public. I did the same thing with Tak.

I played another 5-player game of Shipwrights, with at least one new player, my old boss Jason. He made the kind of suggestions that tell me he likes the game, like “have you ever thought of putting this game on Board Game Arena?” That’s not my specialty, but yes, I’m glad you like it. Also, I have trouble answering questions like “have you ever thought of” because yeah, you’re not the first person to suggest this, so obviously the thought has crossed my mind, and yet I have not acted upon it, which is what you’re actually asking.

If anyone wants to help put any Crab Fragment games on Board Game Arena, let me know.

I played Cold Comfort again with Beth and Derek Mantey. We felt like we had some pretty good fire rules, though I made a critical error early in this game and had basically no chance of catching up. This was a bad omen for the game, though everyone else had a great time. I built a big hotel early, and the Arsonist found it and ruined me. I never came back; I lost tempo, lost money, and I hadn’t even been in the lead. I really don’t like spending the bulk of a 60-minute game thinking “I made a mistake” with no hope of coming back.

I mean, as a designer it’s educational, but as a player, it feels like I’m a hostage. Everyone else enjoyed that game, and their scores were close, but I’m leery of a no-comeback situation ruining a player’s first time, whether it was their fault or not.

I played just one game this weekend that was not in my bag. At the very end of Saturday, Derek showed me Cat in the Box, a Quantum Trick Taking Game, a card game where you declare the suit of a card after you play it.

I love trick-taking games, so I’m always dubious of a game that proposes to demolish one of the pillars of the genre, like “cards are all different.” Cat in the Box turns out to be less of a trick-taker and more of a puzzle game, where the geometric relationship of cards you have played (on a suit map, which tracks what has been played, you see) might be more important than playing the tricks. This game feels like a solution looking for a problem, or an answer to the burning question “what if Hearts had three times as many rules and one billionth the number of starting hands?” That’s not an answer I was looking for.

Yep, it’s a truck serving scoops of cookie dough.

Sunday

I arrived Sunday Morning confident that I had figured out the basics of Cold Comfort. Then I played with my friend Nate, who always ruins everything, and then I felt like I was back at square one.

Sunday kicked off with The Celebrated Jumping Frog Game, a quirky little Lego game from Chief Herman’s Holiday Fun Pack. Then I played Cold Comfort with Beth and Derek, and showed The Harvest to Jessica Blair. 

The Harvest is coming along nicely, and my main question this weekend was whether it makes sense to include 18 money cards for scorekeeping. The game itself is only 36 cards, but DriveThru’s smallest tuck box holds 54. So I kept score with card-money, and in some ways it works better than coins. I look forward to designing a set of Fight City money. Obviously this decision is a strange side effect of printing all of my card games at DriveThruCards, but I don’t feel like the scoring cards were extraneous, and I’ll be glad that the whole game fits in its box.

Fellow game designer and friendly nemesis Nate Heiss just returned from two months in Hawaii, for which I am super-jealous, so we caught up for a while and chatted about our next adventures. I showed him The Harvest and he got stuck trying to make the mechanic of combat make sense. “Fighting doesn’t work like this,” he said, but honestly fighting doesn’t work like rolling dice either, so pick your battles. And I think as abstract games go, The Harvest is actually a decent model for mob politics, with underlings constantly changing sides.

Then we broke out Cold Comfort, and Nate had a deeply adverse reaction to something no one else had really complained about - the dice. Yeah, there is definitely a lot of dice rolling in this game, with layers of unpredictability that can burn out the brain of the most experienced gamer. Casual gamers find it delightful, but hardcore gamers overthink it and then get disappointed by their lack of control.

Since the current beta doesn’t have these rules, here’s a snapshot of what we were doing. Each time the city grows, roll all the dice in the dice pool, and look for places for them to stay. First they will camp on the new card, if they can, and the rest will move directly into hotels. 

This requires a pretty complicated set of rules, and it gives some pretty wild results, with large dice sometimes settling into low-valued hotels because there’s nothing else around, and flipping ownership of neighborhoods in the process.

When Nate got the opportunity to start a fire with the Arsonist, he attacked in the place where he could do the most damage, even though that wasn’t the player who was “in the lead,” and this was the moment that convinced me that giving a player the decision of whom to attack, doesn’t guarantee that they will make the right choice.

Most other groups had been pretty cooperative when deciding things like “where should the fire go?” because all the information in the game is public, and this game lends itself to cooperative discussion even though it’s a competition. (One observer this weekend even mistook it for a co-op game, I guess because he doesn’t know me very well.)

Nate, on the other hand, didn’t need any help with this decision, and he went straight for the wrong place. It did not seem wrong to him at the time, but it was at least wrong in hindsight. 

This critical failure of the rules reminded me that people have motivations outside scoring, and they don’t always make decisions in their own best interest. The concept of the Arsonist, and its complicated multi-step process for granting the power of fire to a random player once in a while, flew into the trash.

Now, to be fair, I’m not going to let a single bad playtest derail a project, any more than I’d let one good session lock everything in place. But this test broke through a dike that was already cracked; all the extra rules about training fire on the leader had pretty much failed to do that.

After a short academic debate about rules that we could just be trying out, we played one more game without any fire at all. In that game I realized some of the other things that don’t work about this game, and found a whole big bundle of rules that I wanted to tear out.

Nate wasn’t really annoyed too much by the fire rules, but by the layered randomness controlling how new dice choose hotels. It was hard to predict, and the game seemed to revolve around making the best guess, hoping you got what you want, and accepting defeat if you didn’t.

There’s nothing wrong with games like that (in general), especially if they are short, but that’s not necessarily how I want Cold Comfort to work. I’d like a practical, predictable core game with some randomness around the edges, not a random core nestled in a warm blanket of mitigation rules.

At one point in this game I felt pretty smart, after a particularly good guess paid off. But that was just one moment in a 40-minute game. The problem felt like it might actually be in the volatility of the scoring rounds, so after the last test of the weekend, I decided to remove all dice rolling from the game, and now each new card comes filled with the dice that belong on it. Players don’t get any stray visitors, just those they invite directly. 

And for now that actually works.

Conclusions

Throughout the weekend I’ve felt like “playing the game,” i.e. the core loop, should be about inviting people to stay in my hotels. All the other stuff about mitigating fire damage and weathering random draws should have been secondary. The other rules encroached on the “actual game,” and I felt like I wasn’t doing the core loop as much as just reacting to the chaos. So in the current beta version I’m refocusing on the core.

Which is all to say. Cold Comfort has a lot more work ahead. I played a couple of 4-player games yesterday and today, once by myself and once with Nora, and the current rules seem as good as anything we have had before. We both miss bits of the chaos, but I feel confident that we will find ways to bring it back.

And the new fire rule is pretty simple - you can light your own empty hotel on fire whenever you want, and even collect the insurance money for it. This was the rule a few versions ago, before OrcaCon, so it’s not entirely new. Collecting insurance on your own hotel is something like a mortgage in Monopoly, and it is useful even if you’re not trying to burn down your neighbors.

Things don’t burn much in this build, but the ever-present threat of fire makes people cautious about how and where they build, and that is part of the point. 

I hope you’ll play Cold Comfort, or at least give it a look, knowing that it’s still super-rough. And I hope to see you at the next convention, where I’ll be testing a better version of this one, and something else strange and new.

To find out where I’ll show up next, check the Schedule page.

Pictured Above: Arsonist (red) and Mayor (green). Not pictured: Chaos.