Kelp: Shark vs Octopus takes place just off the coast of South Africa in an area called False Bay (or Valsbaai in Afrikaans) in one of the world’s most enchanting and mysterious habitats. The kelp forests of False Bay are home to a myriad of species, including a beautiful and ingenious creature — the charismatic common octopus (octopus vulgaris) — but also her mortal enemy: the cunning pyjama shark (Poroderma africanum).
Her struggle for life is constant. She finds herself between the shark’s jaws and it seems like game-over. However, she has an unexpected plan. By inserting her tentacles into the shark’s gills, she’s able to suffocate it until it releases her and she escapes!
If that’s not incredible enough, in a second encounter, she does something so clever, witty, and ludic that it begs to be included in a game design. Again, she is being hunted by the pyjama shark, but this time she’s exposed in a part of the kelp forest with fewer places to hide. She has to improvise, so using her tentacles she grabs the shells that litter the sea bed and then by rolling herself up, she creates a makeshift ball of armor, giving her some protection and confounding the shark.
It circles her, clearly able to smell her but unable to figure out that she’s right under its nose. As the shark finally deduces where she’s hiding, she drops the shells, darts off in the other direction, leaving the shark completely baffled and still very hungry, chomping at the debris as it floats slowly back to the bottom.
The Birth of the Idea
It was April 2021 when I saw the Netflix documentary My Octopus Teacher for the first time. I was immediately struck by the interactions between the two animals and wanted to capture this fascinating encounter from nature in a game. After a few attempts at creating my own board game in the years before, ideas were now starting to form in my mind of some mechanisms that could convey this. I was reminded of some of my favorite games that feature hunter/hunted dynamics: Mr. Jack, Escape from the Aliens in Outer Space, and Android: Netrunner.
I particularly enjoy the thrill of playing as the Corporation in Netrunner, being under mounting pressure but with some control over the flow of information and with plenty of tricks up my sleeve. (In fact, if you know Netrunner at all, my favorite faction is Jinteki because of all the shenanigans they have available in their arsenal.)
But Netrunner is a game that neither my partner nor most of my friends is interested in playing with me. It’s very involved and often a steep challenge to learn, due in part to the asymmetrical aspect of the game. Nevertheless, it is this asymmetry that works so well for the hunter/hunted dynamic, so I wanted to see whether I could create a similar experience to Netrunner with the same emotional journey that I love, but in a more approachable game. With that in mind, I began the work on Kelp.
Hide-and-Seek and Hand-Building?
After a couple of months of incubation, my ideas began to develop into something that I could prototype and playtest. By June 2021, version 1 was ready.
To simulate the tension of the octopus having to balance its need to eat yet stay hidden from the pyjama shark, I knew I wanted a system like the agenda cards in Netrunner.
My idea for Kelp was for the octopus to be represented by a single card that would keep moving from place to place on the board, sometimes appearing, then disappearing again, always in the search for food. It would have tricks it could employ and ways to bluff, trying to deter and deceive the shark from killing and eating the octopus. The octopus would be playing a kind of three-card monte or “shell game“.
At this stage, my idea was that the octopus would play a kind of deck-builder that actually functioned more as a “hand-builder”. In games like Dominion, players draw a hand of five cards each turn, but I thought it would be cool to have those hands already set at the start of the game, spread out on the board.
I tried laying out four hands of five cards, with the octopus card hidden in one of the hands. Each turn, the octopus would manipulate and improve one of those hands of cards by moving cards around, upgrading them to better ones, and flipping hidden cards over to activate them. To use the card effects, the octopus would have to reveal them, slowly reducing the options of where the octopus could be hiding. cards had abilities like “Swap one face-down card in this quadrant with this card“, “Shuffle three adjacent cards, then place them face down on the board”, and “Reduce an adjacent die by 2”.
I love games with dice because of the highs you can get when you are able to overcome the odds. I wanted the shark to be mitigating and managing their dice by improving their dice-pool and by using special abilities (inspired by my favorites, Dice Masters and the very underrated Shanghaien).
I also took inspiration from the worker-placement system in Spyrium, which gave me the idea of placing dice (instead of workers) at the intersections between cards, with this functioning within a dice-pool-building system for the shark — a LEGO shark from my childhood collection! They would circle around the board, place their dice next to the octopus’ cards and, if they met the required value for that area of the board, reveal cards and hopefully attack the octopus.
Both players had ways to improve their options, but each had a different victory condition: The octopus won the game by surviving until the end, while the shark won by finding and killing the octopus.
This early version had fun concepts, but they were overwhelmed by the problems created with my hand-builder idea. I had dramatically exceeded my complexity budget for the game. The octopus had twenty cards in play at the same time, and even though they could interact with only 5-7 of them per turn, they had to keep track of too much hidden information. In addition to tracking where they were hiding, they had to either remember all of their hidden cards or check them every turn — and if they didn’t check a particular card frequently, it was often a clear sign that it was the octopus, leading to unnecessary loops of checking cards to bluff, which frustrated players further. I went through a couple of iterations of this version, but ultimately moved on in order to reduce the memory element and simplify the amount of information present on the board.
Mahjong Blocks Change Everything
In August 2021, I playtested Kelp with a friend who’s also a game designer and it was a great playtest — which means the game sucked and my friend told me so.
In the debrief after the game, he suggested reducing the amount of information on the hidden components by adding a 3×3 grid of tokens and using the cards to manipulate them instead. It was one of those lightbulb moments that seems so clear in hindsight, but I’d been unable to see it for myself.
I knew that tokens still wouldn’t work well as the octopus would need to remember what was on the face-down side of them just the same, but the suggestion launched me down a new path, eventually leading me to using blocks. Blocks with information on only one side could be placed upright so the octopus could always see them, but also laid on their back so they could be revealed to the shark as well.
Suddenly, I had a Stratego-like concept, similar to Lord of the Rings: The Confrontation, but with only one of the players having hidden blocks. I just needed to find a suitable component.
I had only Jenga blocks on hand and tried them, but they were too narrow. Dominoes wouldn’t really stand, and there was nothing suitable in my LEGO collection either.
For a while, I scratched my head and searched second-hand stores for a game from which I could salvage some suitable parts. Fortunately, I struck gold in the form of an unused Mahjong set for €1, an absolute bargain! With some blank stickers, it was perfect for my needs.
Once I came up with the idea for the octopus to “pay” for their card effects by revealing blocks to the shark, this system of hidden/revealed blocks materialized into something captivating, challenging, and fun. There was a satisfying tension for the octopus in finding a balance between hidden and revealed information, and for the shark there were chances for “A-ha!” moments in tracking the octopus’ movements by reading a bluff correctly or taking a chance on attacking a block. This was when playtesters started saying things like “Oh, that’s cool!” as I explained the rule, and I knew I was onto something.
Initially, however, the shell-game idea was overpowered as the octopus was able to shuffle their blocks too frequently, making it difficult for the shark to keep track. After I introduced another way for the octopus to move without randomness, the combination of the two forms of movement led to a hidden movement mechanism that created lots of interesting gameplay opportunities. There was plenty of bluffing and mind games for the octopus to play with and just the right amount of deduction for the shark.
While I was working on the octopus, the mechanisms of the shark were also developing as I streamlined and refined what I started out with.
Hunger, Energy, and Dice!
One of my favorite games is Abyss, and what I love the most about it is the way it harnesses “opportunity costs” when players take actions in the game. (I first heard the microeconomic-theory term “opportunity costs” described by Tom Lehmann in the <a href="https://justingarydesign.substack.com/s/think-like-a-game-designer-podcast/archive?sort=new” target=”_blank” class=”postlink” rel=”nofollow noreferrer noopener”>Think Like A game Designer podcast — it’s an enlightening listen!) In Abyss, you have to make compromises to get what you want and often the action that you take improves the options available for your opponents, or sometimes you push your luck too far and a great opportunity passes you by. I wanted Kelp to have those elements so that players had tough choices to make each turn.
An early idea I had for the shark side was to allow dice to have multiple uses, and what I finally settled on was two interlocking systems through which dice cycled. The first was that dice could be placed on the main board for the shark to move more easily and to be able to reveal and attack the octopus’ blocks. Then most of those dice would get added to “growth tiles” that gradually unlocked permanent abilities for the shark.
The second was that dice could be stored for their value to eventually purchase more dice and one-time effects that did all sorts of different things. High-value dice were often useful in both systems, making the choice of where to place them each turn a satisfying decision point.
That in itself was quite fun, but I wanted the shark’s compromises to more closely mirror the octopus’. Therefore, I made it so that whenever the shark spent their stored dice or whenever they used a rare “strike” die and missed, they would have to place a die onto their “hunger track”, which would count down to the end of the game. Now they had to be efficient and accurate in their hunt, and there was satisfaction in that, too. This ratcheting up of tension for the shark equaled the tension experienced by the octopus, and Kelp started to be really enjoyed by playtesters.
A Last Chance to Escape
The next year saw further balancing and streamlining changes take place, though the endgame still needed refining. In earlier versions, the octopus won the game by surviving until the end and the shark won by killing the octopus. This was a clear, simple victory condition for each side, but I was searching for a way for the octopus to have a chance of escape if they were caught. Without this, the game can be anticlimactic as the octopus can be discovered suddenly, attacked, and killed without a chance to react.
Returning to that moment in My Octopus Teacher when the octopus suffocates the shark to escape, I implemented a mini-game that simulates the tricks both animals employ in this final confrontation. The octopus chooses an escape strategy of fight, flight, or flinch, and the shark tries to correctly counter it.
The octopus now has a last-ditch attempt to escape if they get caught, but rather than the cards being equally beneficial, the effect of each card is circumstantial and may have pros and cons if it triggers. There is often a better or worse choice to make, which adds to the mind game of what your opponent will play, so now you have an interesting dilemma.
Most importantly, the mechanism is in place to lead to an exciting and tense climax of the game. With the confrontation, both players know the end of the game could be imminent, and they have to make a decision that impacts which way it goes. This leads to a spike in tension: If the octopus does manage to escape, they must discard the card they used, with the shark discarding their matching counter card. Now if the octopus is caught again, that spike of tension will be higher and their chances of escape lower. There are more nuances to the endgame as well, but those are to be discovered in gameplay.
Finally, I wanted one last wrinkle to encourage the octopus to take more risks and not to just turtle and wait out the game. That incentive came in the form of food blocks that give the octopus a one-time special power but at the cost of revealing themselves in order to eat the food. This move was a big risk and the payoff was good, but I knew I needed to increase the stakes even higher. Now, if the octopus manages to successfully eat all of the food in the game, they win immediately — a very enticing, but challenging prospect.
With this alternate win condition for the octopus and the final confrontation giving a little bit of hope, tension peaks in the endgame. As the octopus comes closer and closer to winning outright with food — and also closer and closer to death — the shark is growing in power and looking for the perfect chance to attack…
It is with all of this that Kelp simulates the relationship between common octopus and pyjama shark, and hopefully delivers a tense, thrilling experience at the same time.
That wraps up my designer diary. Thanks for reading! I’d love to hear your thoughts in the comments below.
Carl Robinson
<a href="https://boardgamegeek.com/boardgamepublisher/49697/wonderbow-games“>Wonderbow games
Leave feedback about this