by Emily Vincent

I designed Knitting Circle during my first year as a game designer. I set the goal of designing as many games as possible so that I could build up my design skills quickly. I wanted to work on a lot of different types of games and mechanics, so I would give myself design briefs based on games that I liked, podcasts I was listening to, or random ideas that I had. Knitting Circle was born from a design brief that I gave myself. I’ve tried to capture that journey here for all of you, gamers and designers. I hope you enjoy this peek behind the scenes!

Knitting Circle Overview: How Does the Game Work

In order to understand the journey, I thought it might be useful to know the destination. Knitting Circle is played over 6 rounds, each of which has two primary phases: the drafting and the crafting phases. During the drafting phase each player takes turns moving their kitty pawn 1 or 2 spaces clockwise around the rondel, jumping over empty or occupied spaces. They then take one of the two tiles in the space and place it on their player mat, keeping it on the face that it was drafted (knit or purl). This continues until all players have drafted 4 tiles.

Then players enter the crafting phase. During this simultaneous phase, all players can place yarn tiles in sequence (starting top to bottom) onto their garment cards, purchase new garment cards with their yarn tiles, complete garments and place any corresponding buttons, pay to flip one of their yarn tiles, or use a grabby paw to take a tile of their choice from the bag. Players are trying to maximize their score across multiple dimensions: length of garment (longer garments score more), garment bonuses (end game conditions on each garment card), and their button objectives (using certain colors, making certain garments, or making certain patterns). The garment bonuses allow players to engage in a little bit of engine building as their early garments may set the path they take in their later garments.

Getting Started

I got the idea for Knitting Circle while listening to an episode of the Building the Game podcast about cozy games. I was driving to work and I paused the podcast to ask myself the question, “If I was going to design a cozy game, what would it be?” My first thought was of knitting, because winter hats and scarves are the coziest things I could think of.

Because I have dabbled in knitting, I was able to quickly come up with the thematic ideas that became the heart of the game. In knitting, there are two basic stitches, knit and purl. The cool thing about these two stitches is that they are essentially two sides of a single stitch – if you do a knit stitch, the back side of it is purl. If you do a purl, the back is a knit. So I knew I wanted tiles that were knit on one side and purl on the other. Building off of this idea, knitting patterns are specified by some number of alternating knit and purl stitches. By the time I had arrived at work, I had mentally sketched out the core concept of the game – to build color sequences that required certain patterns of knit and purl.

Inspiration

When I designed this game, I was inspired by the idea that cozy games leave you feeling proud of what you built even if you don’t win. So in a game like Creature Comforts (Roberta Taylor, Kids Table Board Games), you might feel happy with all of the cozy things you built into your tableau, such as a lovely bookshelf and rocking chair. I was also inspired by the puzzle games that I love playing, specifically Sagrada, Azul, and Calico. I wanted to create a game that lived in the same part of your game shelf as these amazing games.

Initial Prototype & Testing: The Good and the (Mostly) Bad

The very first prototype was made using index cards and resin gems. Because the gems didn’t have sides, I came up with a graphic design for the garment cards with two columns, one for knit and one for purl. This separation of yarn tiles ended up living far too long in the design, all based on the fact that my first prototype components couldn’t be sided. I thought the physical separation was helping players but this didn’t end up being the case. In retrospect, it was probably hurting most players more than it was helping.

First prototype with gems instead of yarn tiles

The primary goal of the game was to create different color patterns on your garments, specifically solids, stripes, symmetrical, colorblocks, rainbows, and abstract. The puzzle was to create those color patterns while alternating yarn from the knit and purl sides of the player’s board. This initial prototype allowed me to gain confidence that the knitting puzzle was actually fun.

I then made a set of circle tiles with knit on one side and purl on the other. This is when I discovered that the difference between the knit and purl stitch was difficult for a lot of people to see. Because of this, I continued to try to use columns on the player board and the garments as a second way to distinguish the knit and purl. I also just kind of liked the zigzag pattern that was created by the two columns.

Part of these early iterations included a passed draft that required players to take tiles in a knit/purl sequence determined by drawing a knitting pattern card. These knitting pattern cards had different lengths, up to five sequences of knit and purl. At the beginning of the round, five tiles were put into the four different bowls for the passed draft. When a bowl of tiles was passed to you, you had to take a tile from the bowl and flip it to the side designated by the current position in the sequence. So if the sequence was knit-knit-purl-purl-knit, you would take tiles on the knit side from the first, second, and fifth bowl you received but would take tiles on the purl side for the 3rd and 4th bowls. To help you keep track of them, you’d place the knit tiles into the basket on the left side of your player board and the purl tiles into the basket on the right side. You could only flip a tile if you discarded another tile to pay for the flip.

As I started playtesting publicly, I observed players struggling with the core puzzle, specifically with alternating the knit and purl sides of the tiles. They lost track of the requirement to play in sequence and they would also absent-mindedly flip tiles over. Since alternating between knit and purl was the heart of the puzzle, it was pretty disheartening to watch it be the biggest struggle for playtesters. I began to wonder if the puzzle might be too hard for many players.

The Eureka Moment

There were two breakthroughs that really turned the corner for this design: the yarn tile design and the simplified draft. One night in February 2023, I had an absolutely disastrous playtest. My players just could NOT keep track of which tiles were knit and purl. They kept placing them on the wrong side of their player board and using them in the wrong spots on the garments. I was so frustrated because it seemed very straightforward to me. But I came out of the playtest with a fire under me to make it so that they COULD NOT play their tiles in the wrong order.

That night after my playtest, I designed a custom tile shape that would only fit if played in an alternating knit/purl sequence. They were keyed with a positive and negative side. Before I went to bed, I sent a file to print 5 of these new tiles on my 3D printer while I slept. The next morning, the tiles worked exactly as I’d hoped. I immediately set out to print 8 different colors of tiles and redesigned all of my garments to fit the zigzag shape that these tiles created.

This was a huge turning point for the game. The bright colors, toyetic pieces, and intuitive puzzle were an instant hit when I debuted them at Unpub 2023. People immediately understood what they were trying to do and the common mistakes players had been making disappeared. Also, the table presence was AWESOME. It drew people in; they wanted to know what the game was and how to play.

Knitting Circle prototype at Unpub 2023

The other big breakthrough happened at Unpub. In the first playthrough, my friends Ashwin and David suggested that the knitting pattern cards were making the draft too hard. We immediately tried playing again with a much simpler approach – each player chose which side they wanted the tiles in their bowl to be and flipped them all to that side. When a bowl was passed to you, you had to take a tile on whatever side it was already flipped to.

With the change in the draft and the new tiles, the feel of the game changed completely. Suddenly people were talking and laughing around the table instead of quietly focusing on their own puzzle and/or frustration.

The follow-up from Unpub was exciting and a little surreal. Randy Flynn (designer of Cascadia) played the game and tweeted about it. Elizabeth Hargrave (designer of Wingspan) also tweeted about it, saying she’d heard about it at the show. A number of other designers at the show also posted about it on social media and suddenly the game had buzz. It was such an incredible act of generosity, seeing people in the industry using their platform to shine the spotlight on someone new. I’m so incredibly grateful to everyone who talked about the game and spread the word. As a result of this buzz, I ended up pitching the game to 16 publishers over the next few months and got connected to Flatout Games.

Feedback from Publishers

A lot of the feedback from publishers was that the game had a solid core but was in an “in between” place, where development could take it in a lot of different directions. It was a fun puzzle and the gameplay was fairly light. Some publishers wanted the game to be simpler, so that it was even lighter and more approachable. They saw a very toyetic form factor and wanted the gameplay to match that. Others thought the game needed to be a little bit more complex, to increase replayability and continue to challenge players as they increased their skills. It really helped me understand how much a game could change based on which publisher signed it and what their vision is.

The Development Process: Continuing the Journey with Flatout Games

Working with Flatout Games was an amazing experience. It was awesome to see how they approached development and I’m so glad I got to work closely with them. The high level goal of our development effort was always to develop Knitting Circle so that it felt like a Flatout Games’ game, especially Calico. The changes we made to achieve this goal really transformed the game.

The team wanted to add multiple paths to victory and increase the replayability of the puzzle. Multiple paths to victory is a hallmark of their games and particularly of Calico. In Calico, if you’re having trouble achieving your quilt goals, you can focus on scoring points with cats or with buttons. There’s always something to work on while you’re playing. Knitting Circle didn’t have those different paths to victory. It had two types of scoring, for garment length and also for patterns, but the pattern scoring wasn’t working well.

The first step was to separate the garments out into different cards. The version that had been signed had the player boards with one of each garment type on it, but this meant that the puzzle was basically the same every game. We worried that people would quickly “solve” the puzzle and wouldn’t want to play again. Separating the garments onto their own cards meant that the number and type of garment that you made each game was variable. This shift to cards immediately helped the replayability.

Another idea that Shawn Stankewich brought to the development early on was adding a deck of dual use cards that could be exchanged for additional tiles OR tucked under garments for end game scoring. This achieved two things. It increased the tile economy and provided another vector of scoring. Both parts of the card were working well, but the new deck of cards increased our component count and gave players another hand of components to manage. After much playtesting, we decided we needed to figure out how to keep the effects (bonus yarn tiles and end game scoring conditions) from the card but without actually adding new cards. We eventually moved these elements to the garment cards. We gave players bonus yarn tiles when they covered certain spaces on their garments and we put the end game bonuses on the bottom of each garment. While we lost a little bit of variability we simplified the gameplay and streamlined the components.

We ended up cutting two of the color patterns – rainbow and abstract. We found that the rainbow pattern was hard for many players to understand and the abstract pattern was too easy. We also cut the number of colors in the game from eight to six. This made it a little easier to get duplicate colors. This was important because all of the remaining color patterns required duplicate colors. Cutting to only six colors also enabled us to use the same six colors in Calico.

As our third vector of scoring, Joseph Z. Chen figured out that we could use buttons as mini-objectives that players can achieve throughout the game. Buttons also mirrored a scoring vector in Calico, which made Knitting Circle feel more like a Calico game. Once we had the length, button, and end-game scoring, we felt confident that the game really felt like a Flatout Games game. On top of that we added the advanced scoring objectives, another hallmark of Flatout Games games to enable variable setup, which in turn increased replayability.

As a team, we also had our eye on the passed draft – it just felt like we could come up with something more interesting. Shawn proposed the rondel concept when he was working on the solo mode and we quickly realized that it would work at all player counts. Prior to the use of the rondel, there was basically no player interaction in the game. The rondel added the opportunity to block other players from getting the tiles they wanted and leap over other players to move further around the basket to the tiles you might want to draft. It was an immediate winner. Later in the process, Molly Johnson wondered if the rondel draft could be used as a mini-expansion in Calico – to provide a tile-drafting variant! This was another step in bringing Knitting Circle into the Calico universe.

Lastly, my favorite development task was to add cats! If this was going to be a Calico game, it definitely needed cats. We hadn’t managed to add them while we were working on the scoring vectors but serendipity struck while I was at Unpub in 2024. I was testing the new rondel but had forgotten to bring any meeples with me. I did have cat pawns though, from another game I was working on. Players were delighted to imagine frisky kitties tumbling over each other in the yarn basket and retrieving yarn for their humans. I immediately messaged the team and we loved the idea of having kitty helpers. It was also a ton of fun to write the little cat backstories on all of the player mats. If you read those, you’ll see how all four of the kitties in the game live in the broader Flatout universe. We also decided that a couple of other mechanics could be themed as cats as well, specifically the cat grabby paws that let you take a tile of your choice from the bag and the “ugly sweater” button became the “angry cat” token.

Knitting Circle at Unpub 2024, with the garment cards, rondel, and buttons.

Final Thoughts

Overall, it has been an incredible experience to go from a cozy idea to a game on tables around the world. I’m so grateful to Flatout Games for taking the game on and for involving me in the development process. It was both fun and educational to be a part of the Knitting Circle development team. For designers who are interested in seeing how a game goes from signed prototype to finished product, I’d recommend participating in a development effort.

And the best part is that I think we created a pretty cool game! The sequencing puzzle based on color and tile side is fairly unique. I also think the close thematic connection to knitting helps the game shine – players who knit say that they can see that someone who has knit was involved in the design. Lastly, it really hits the highlights of games that I love. There is fun player interaction around the rondel, where I can lightly interfere with other players while setting up for my puzzle. And then there is my personal tableau where I can challenge myself against the puzzle without worrying about disruption from other players. At the end of the day, I think we delivered on the vision of a cozy and puzzley game night.