This final post covers how the game
design for Lost Island was created. A
direct transcript of the notes in the book used to plan the game is
supplemented with some annotations for clarity. It is probably only of interest
to you if you want to look at the design process behind the game, but I largely
suspect this is the interesting part of the process for many readers.
Adventure board game
Comments: Usually my premise statement says something more tangible than
this… I suspect my mind was already skipping ahead to the mechanics. I already
knew that I wanted to make something with the feel of a Ray Harryhausen or cheesy Doug
McLure movie to it, and the premise should probably have stated this. But since
the notes are really for just my own benefit, it scarcely matters.
Originally, I had been planning to make
a fantasy game with my specially cut hexes (the ones I used previously to make Black Sun), and I had presumed that ‘fantasy’ would mean
‘sword-and-sorcery’. However, after thinking about it idly for a while, it
occurred to me that ‘fantasy’ was a much broader term and could describe many
different settings. I think this was the point that I thought a fantasy monster
B-movie was perfect material for a fun little boardgame – and my infinite respect
for the work of Ray Harryhausen was certainly an inspiration.
Start with a cluster of 3 hexes.
[There is a sketch here of the initial
setup for the game which shows three hexes, one of which is the crash site.]
Each turn, explore the island by discovering
new hexes. (Can also move back through hexes faster when exploring).
Monsters (3) roam the island and must be
One resource: Wood. Comes in little sticks.
Get one when you find a forest and travel
through a forest and chance on beach.
Comments: The setup of the hexes didn’t change, but this point about moving
back through the hexes never amounted to anything. An extra monster was later
added, but the resource mechanic stayed. The rule about getting a wood when you
travel through a forest was later changed, as we will see below. The idea of
having a single resource was to stimulate a resource economy without excessive
complexity – for this game, which has plenty of spatial play, this transpired
to be a solid design choice.
Volcano (1) (1)
Centre: 1, Edge: 0
Mountain (m) (5) Centre: 2, Edge: 2
Forest (2 x m) (10)
Centre: 6, Edge: 4
Beach (2 x m) (10) Centre: 0, Edge: 10
Comments: It seems like I already had an idea of which different hexes I was
going to want, and the only issue was the distribution. The process of deciding
how many hexes of each type would be required began with some speculation: if
there are m mountain pieces, let’s assume there are 2m Forest and Beach hexes.
Then, I looked at how this would play out if m = 5, and split the hexes between
centre and edge hexes. Sometime around this point, I got out some blank hexes
and started experimenting with how they might fit together. I was testing how
the island might come together if about half of the hexes were ‘edge pieces’ –
that is, half water and half land – and this actually worked out rather well,
producing hex patterns that felt very island-like. After 15 minutes or so of
dealing hex distributions I was sufficiently convinced to proceed on this
The final distributions were 1 Volcano,
8 Mountain, 16 Forest (6 edge), and 16 Beach (all edges), so the original
algebraic pattern held, and then I specified approximately half as edges – all
the Beaches plus 6 Forest edges. This should have been 5, strictly speaking, but I decided to
over estimate since my practice runs with the hexes had shown me that the edges
were key to how the shaping of the island would work.
Next, I started looking at the internal
game economy: what would Wood be used for?
Hut: 5, Fence: 1, Fire: 2
Raft: 10, Spear: 1
Comments: The numbers I produced here stuck for the game – it just made sense
that a Wood counter could become either a fence or a spear, and that you would
put two together to make a fire. 5 and 10 for the Hut and Raft were chosen
because these were internal game goals – the first would want to be at least
twice its prior value (2), and so again for the next one – hence an implied
game sequence of 2, 5, 10. Clearly, I had already decided the players would be
using a Raft to escape, but I don’t think I had decided on a purpose for the
Hut yet – other than as a stage goal (which in some respects was the more
mountain = 0, beach = 1, forest = 2, starts on largest beach, remains on beach
unless beach with crab symbol comes up à moves towards nearest survivor. Can go 1 hex inland only i.e. must
always be on hex next to beach.
forest = 2, beach = 2, mountain = 0, starts on largest forest, moves towards
mountain = 1, all; moves 3 but requires all 3 to strike in 1 hex.
Ants/Termites: eat Wood!
Comments: I had no preconceived ideas here, so I was just jamming off the
possibilities. It looks like I was originally thinking about different movement
rates over the different terrains for each type of Monster, but later it
occurred to me that the game would be improved if all the Monsters followed the
same general rule which would just be adjusted to match a specific terrain
type. (We’ll get to this). ‘Starts on the largest beach’ or ‘largest forest’
transpired to be a meaningless term, but the inclination was sound – it became
placing the monster when there were 3 or more contiguous hexes of the given
type. Oh and the types of Monster at this stage were just suggestions, and
became a more general specification later so that when we would get to make the
Fimo models we could make what we wanted to. In fact, I’m sure part of the
motivation for this game was the idea that we would get to make little Fimo
Points = Stories
Kill monster = 1 Story
Build Hut = 1 Story
Sequence: Fire --> Hut --> Raft
Hut must be sealed on all sides by a hedge
and fence before Hut can be built.
Monster killed only when it can’t flee.
If you can’t stop the monster, it ‘defeats
you’ and you miss your turn.
Comments: The basic idea here held – and the sequence of play did derive from
the sequence above. This idea of ‘Points = Stories’ fell away; points were just
scored for achieving goals, with nothing especially fancy going on. The idea of
the Monster being killed when it couldn’t flee eventually fell away once I
decided there would be a scoring sheet. A health mechanic, with one point per
wound caused, eventually became the rule.
Spear (1 Wood): drive Monster back 1 square
Fence (1 Wood): put on board. Monster must
Fire (2 Wood): can include fence.
Hut (5 Wood): Hut on board… safe at Hut and
can store Wood safely
Raft (10 Wood): end game
Hut: must be ‘safe’ i.e. Wood/edge lined.
Raft: must already have Hut.
Comments: Just developing the internal economy. Most of this stands, except
the idea of a Monster that would destroy Wood was dropped, so the idea of the
Hut protecting your Wood supplies dropped naturally out. I decided that the
internal economy would be fragile enough from player competition without
something destroying the basic resource.
Move or Build
Fire or Spear
3 Actions Monster Attack
Take 3 Actions Monster Moves
- Monster Move
- Monster Attack
Spear = 1w, Fire = 2 w, & 1
- Take 3 Actions
Spear = 1w, Fire = 2 w, Fence = 1w, Hut = 5 w
Comments: watch me go around in circles on the turn sequence. The final
sequence – not listed in my notes – was Monster Move, Take Actions, then Monster
Attack, but this did not become apparent until the early play testing.
Beach = 1 Action
Forest = 2 Actions
Mountain = 3 Actions
(Swim = 2 Actions)
Scout hex = 1 Action (place anywhere)
Spear = 1 Wood & 1 Action
Fire = 2 Wood & 1 Action
Fence = 1 Wood & 1 Action
Hut = 5 Wood & 3 Actions
Raft requires 10 Wood contributed by all
players & requires two players to be on the same beach spot.
Comments: Apart from dropping the ‘swim’ action, and changing ‘scout hex’ to
being called ‘explore’, this is pretty much how it turned out. As the different
actions became apparent, they gradually split into two different verbs Make
(Fence, Spear, Fire) which uses 1 Action, and Build (Hut, Tool) which uses 3
Actions. You can see the roots of that here, but it didn’t become finalised
until the Tools were specified.
A. Explore: ends when there is a
fire for each person covering unique 7 hexes.
B. Build: build Hut – requires hex
which is bounded on all sides by fences or edges
C. Escape: build Raft – requires 2 players to be in same beach hex and
contribute wood, then requires 10 Wood to finish.
Comments: This is how it worked out in the final game, except some of the
wordings were cleaned up for the rules.
- When new forest hex appears, place Wood on it.
- When dinosaur passes through Forest hex put Wood on it.
- 1 Action in Forest = 2/3 chance of Wood [3, 4, 5, 6]
- 1 Action on Beach = 1/3 chance of Wood [5, 6]
Carrying: can carry 3 Wood (therefore to
make Hut need to leave 2 on spot)
Comments: I have a liking for adding a little alea into board games (helps make it more accessible), and the
patterns here were designed as “in a forest, it should be easy to get Wood, but
you could fail, on the beach, finding Wood (drift wood) should be more of an
achievement”. Rolling 5 or 6 on one die is a way of getting fiero out of a die
roll, but it must be said since Forests are the better source of Wood, I knew
that most of the time these die rolls would be about avoiding failure rather
than treasuring success. Still, all these mechanics remained in the final game,
and worked well in practice.
Crab – Beach; miss turn & lose half
Ant – Forest; eats Wood
Eagle – Mountain; miss turn and lose half
Dinosaur – Volcano; as above + requires 2
Spears to kill
Hex for each challenge Scoreboard
<Fire> <one for each monster>
<Hut> <Raft> = 7
Comments: Shaking down the rules, here… The cost of being biffed by a Monster
eventually became the same: you get pushed back to the Crash Site and you lose
all your Wood (which is left behind). Moving across the island is expensive in
this game, so losing your spatial position is a significant penalty – there was
no need to penalise the player with a lost turn as well. (Lost turns are an
overused penalty in boardgames in my opinion – they create frustrations but no
rewards). At the end here you can see me finally decided to get a Scoreboard,
and then looking at how many different ways to score there would be. The
answer, as you can see, would be 7 – one for each monster, plus the goals of
each stage of the game.
3 Wood & 3 Actions. Can carry 5 Wood.
Crossbow: spears now hit on 4,5,6 (not just 5&6). Costs 3 Wood and 3
moving through Woods generates 1 wood. Costs 3 Wood and 3 Actions. Forest move becomes 1 + search
for Wood in forests always succeed.
10 Spear, hit 3 1/3 times.
Buy Crossbow, 7 spears hit 3 ½ times.
Comments: This is the last page of my notes. The final addition was the Tools,
called here ‘Power Ups’ because, well, that’s what they are. Note how they all
developed the same cost – this led to the split of the ‘Act’ verb into a ‘Make’
(1 Action) and ‘Build’ (3 Actions) verb. The calculation at the bottom is seeing how effective the Crossbow would be in practice by checking it probabilitically. The cart was stupidly underpowered
here! Carry 5 Wood? Not worth the price. Almost immediately this became ‘carry
as much Wood as you like’.
After some play testing, there were two
apparent problems with the Tools. The first was that the Machete was clearly
the most useful Tool, as being able to move faster across the island is mighty,
and getting Wood automatically is a sweet bonus.
Another problem that came out in play
was that to make the Tools you would just sit in a Forest hex, farm Wood until
you had enough to make a Machete, then a Cart then a Crossbow.
The eventual solution was to make each
Tool belong to a particular hex type, so that the player would have to move to
a Beach, Forest or Mountain to make the different things. The Machete was placed on
the Mountain (the hardest to access) for balance. The Cart was still too weak,
even with an infinite ability to carry, so this was granted faster movement
through Mountains as well. All these changes left the Crossbow seeming weak, so
that was given the ability to hit Monsters in neighbouring hexes. After these
changes, everything seems to be pretty neatly balanced.
Overall Comments: Not much changed between the end of the design notes and the
implementation of the game, except perhaps for the addition of the ‘Eruption!’
rule which ends the game with the threat of encroaching lava. This is so
perfect for the pulp adventure feel of the game that I can scarcely believe it
wasn’t thought of sooner, and I so wish we had sprayed the underside of all the
hexes Red to show the lava. We might still do so, but I’m paranoid about
damaging the painted side of the hexes.
This was a surprisingly straightforward
design task for me. Most of the rules were made in a single night, and the vast
majority of the draft rules worked fine in practice. There were problems with
hex placements until we made “Mountains wild” so that they could fill in the
spaces that occurred when hexes were positioned awkwardly, but this was the
only real oversight; everything else was just game balancing issues.
On the whole, I feel this game fulfilled
its original design goals perfectly, and I’m very pleased with how it turned
out. The cuteness of the Fimo monsters my friends made really adds to the fun
of the game, too – it wouldn’t have been the same without them!
If you have any questions about the game
or the design process behind it, feel free to ask in the comments. Have fun!