Arctic Roll is  a novel roll and write game which sees players competing on the same sheet of paper. It is being released through Kickstarter as a Print’n’Play — players will receive a series of PDFs to download and print at home.
https://www.kickstarter.com/projects/rollingrhinogames/arcticroll

The first part of this Designer Diary considered how the game reached the point where it was pitchable to publishers.
This, the second part looks at the decisions made once a publisher signed the game.

Pitch or Publish?

Something that many designers ask themselves is whether they want to self-publish or pitch. I have always been in the pitch camp (DYSWIDT?).  I am more confident that I would be able to find a publisher for a game deserving of publication, than of my ability to choose which of all the games I have designed, were deserving of wider attention.

Sellsheet

The game was beginning to take the form that can be recognised today.  I am pretty confident that Arctic Roll is the best sell sheet I have produced for any of my games. The most common critique of sellsheets is “too much text”. This sellsheet does have some text, but everything a publisher required to understand the game is conveyed in the pictures. The draft, the bonuses, the scoring and the map are all clearly shown, and it also conveys several ’hooks’ which set it apart from the slew of roll and write games that have appeared over the past few years. Players compete for areas as they move around an ice sheet, and the map records the entire history of the game.

While pitching the game to numerous publishers at Essen, I typically opened with a recognition that people were becoming a bit less enthusiastic about another roll and write. If it has done something that we haven’t seen before then we are still looking,  was something I heard from multiple publishers, and so this ‘hook’ was therefore important.

Arctic Roll is almost unique in the extent to which at the end of the game the entire narrative from start to finish is laid out in front of players. How many times have you asked  other players at the table, “So why did you get so many more points than I did?” At the end of a game of Arctic Roll, you are not also able to compare scores, but if you want to, you can also see exactly where and when the winning decisions were made.     

A True Print and Play

This was the form in which Arctic Roll was first noticed and signed by Martin Van Rossum, from Rolling Rhino Games, who specialise in print at home, roll and write games. 
My first discussion with Martin, touched on how we might remove some of the additional components, which could be easily included in a small box edition, but limited the suitability as a true Print’n’Play. Could we remove the separate draft board, and player pawns? 

It wasn’t too difficult to move the draft board to the side of the map adding columns for each round to record player choices, and the player pawn on the map which was there to make it easier to keep track of each players location, was not a requirement. Players can, if they choose, draw around the hexes that they stop on at the end of each turn, but this is not required.  

As we began to consider the potential for multiple different maps—this decision made even more sense. It would be easy to fit the specific bonuses available to players to the map on which they were currently playing. Although it would have certainly been possible to include draft boards designed for different player counts in a boxed edition, this constraint has actually allowed a greater freedom, both in the current edition and in the additional maps planned for the future.

Development

So began the process of development. It is not unusual for this to be something of a subjective process as the publisher tries to match a signed game to their target audience.
Designers often ask how complete does a game need to be before pitching. My opinion is that while a designer should make their game the best game they can, they typically reach a point where any further changes become change for changes sake — and could just as easily move the game away from a particular publishers preference as towards it.

In the case of Arctic Roll, my initial discussions with Martin Van Rossum from Rolling Rhino, were primarily around the desired level of complexity. How many scoring objectives did we want? How much should they each be worth?
It is not uncommon for Roll and write games particularly, to have numerous scoring methods with combinations triggering off other combinations which I turn trigger further bonuses, and these were all explored. Obviously the more of these, the longer the required ruleset and it was considered important that we produce short concise rules which could be easily digested — something that is important if the game is to be gateway or perhaps gateway+ in terms of complexity.

So Occam’s razor was consistently applied, reducing both the scoring objectives and combinations.   
One of the design principles was that each of the different objectives, could be the focus of a strategy which could lead to victory.

New Maps, New Mechanics

One of the nice things about the development of Arctic Roll was that the route to publication as a Print and Play allows for any idea to be put on the (ice)shelf for later.  The maps we include in the initial Kickstarter release, all have differences in the optimum player count  and minor differences in scoring, but we have already planned more with various different mechanics. Were we to be pursuing a traditional route to market, then short of a runaway success like Ticket  to Ride where there is a seemingly unquenchable desire for new maps , the number of new maps and therefore mechanics would be necessarily limited.

When all that is required to offer a new map to players, is to produce another PDF for download, then I am confident that new maps will be released as long as there is demand from players.

I playtested a 5-6 player map played over 2 sheets of A4, just a couple of days ago, which is a significant step up in terms of complexity. It felt closer to a big box game, played through the medium of pen and paper, than a roll-and-write filler.  I am looking forward to really pushing the print and play format to create a series of maps, which each have different mechanics and new challenges.

Arctic Roll—Solo mode

However, a map that was not shared could be used by solo gamers, but could also be used in a similar way to many roll and writes, where any number of players can compete in a form of multi-player solitaire. We were looking for a way to recreate the challenge of drafting without the pressure of other players taking the dice that you wanted.

As the game was now primarily focussed on shared maps designed for specific player counts, there was a need to reconsider how the game might be played solo. David Digby wrote a 3 part article in this blog about his approach to designing solo modes ( https://www.thegamespeople.co.uk/designing-solo-modes-part-1/ ) and I discussed with him an approach that would see a player compete against a virtual player who would take dice and block the player from placing his hunters in the best spots.

However, I am actually really pleased with the solution that I found.

4 dice—3 of one colour and a 4th of a different colour are rolled. The odd coloured dice is fixed, and all players must take that dice. The 3 other dice are assigned to 3 different bonuses which are available in that round and each player chooses one of them.
As it is always ‘your turn’ I added extra rounds to make the game 9 rounds long, thus creating the late game congestion which is a feature of the other multiplayer maps.

I also added additional bonuses so that there are 9 in total, with each bonus appearing 3 times. Players mark which bonus they have used, and score additional points for each row or column of 3 bonuses they use. There is a final bonus if players manage to use each bonus over the 9 rounds. I like the conflict created by the fact that players can see which bonuses will be available to them in future rounds but not which dice value they will correspond to. To add a final twist, players can actually use bonuses ahead of the round they are currently on, but they are also sacrificing the dice that would be on that spot before they see what value that dice will be.

Enough (Interaction) is Enough!

Generally, multiplayer roll and writes try and create some interaction between players. Cartographers for instance, actually allows you to add monsters to mess up your neighbours map but this had already been discarded from early versions of Arctic Roll  as a ‘negative player experience’.   More commonly games empower players on ‘their turn’ to make a specific choice, that then influences what every other player is able to do. However this does tend to limit the number of players, as everyone has to get the same number of ‘special’ turns where they control the game.
The lack of interaction on this solo mode is apparent, but as every other map forces high levels of interaction between players at every point in the game, it is actually a benefit to offer a multiplayer solitaire version which can be played by any player count, and without that interaction is a bit more ‘zen’.

Kickstarter Campaign and the Lipstick Effect

The term “the lipstick effect” was coined to describe the fact that in periods of economic downturn, people will purchase less costly luxury products. 

While Kickstarter backers are used to backing huge games that can take many months or even years to fulfil, the platform is surprisingly suited to low cost Print’n’Play games which offer almost immediate electronic fulfilment as soon as the campaign finishes. This model offers backers the opportunity to purchase a new game for a fraction of  the cost and with minimal risk.  

Arctic Roll will be Rolling Rhino’s third Kickstarter campaign to follow this model — and Postmark Games and No Box Games are amongst the other publishers who have a similar offering. It is very common for new publishers to be encouraged to start with a small Kickstarter campaign before attempting to publish something larger. While this model completely avoids the need to get your head around logistics and fulfilment, there is certainly a case to be made for encouraging new publishers to consider a print and play model as they start out.

Sustainability

Despite the fact that backers are being asked to print a new sheet of paper each time they play the game, we would assert that this remains the most sustainable way to deliver a new game to players.

By delivering electronically, we avoid the manufacturing and logistics costs, and do away with the need for additional packaging. Everyone has a few dice at home, and can gather together a few coloured pens or pencils, so by expecting backers to provide these, unnecessary production is avoided. 

In view of the typical lifespan of dry wipe markers we would also encourage players to avoid laminating the maps, but simply print a new sheet each time you wish to play. If you are the sort of player who ever goes back over the score pad to compare your games then keeping the maps might appeal.

Arctic Roll is currently on Kickstarter.
https://www.kickstarter.com/projects/rollingrhinogames/arcticroll

They say you never forget your first (don’t they?).
In which case this is a significant milestone in my attempts to become a game designer, as Arctic Roll is my first published game and has just gone live on Kickstarter. As I didn’t have very much to do for the campaign itself, I was compelled to write a ‘designer diary’ to coincide with the release. This is far too long for a single post so I will be breaking this down into somewhat manageable chunks.

Arctic Roll Designer Diary

Arctic Roll is  a novel roll and write game which sees players competing on the same sheet of paper. It is being released through Kickstarter as a Print’n’Play — players will receive a series of PDFs to download and print at home.
https://www.kickstarter.com/projects/rollingrhinogames/arcticroll

The Idea

Many years ago 5 dice were thrown and I was told “There are 12 Polar Bears around 3 holes”. The challenge was to work out through repeated rolls and a process of elimination, what was the secret code.
Essentially the central pips on a 1, 3 and 5  are holes, and all the surrounding pips represent polar bears. It stuck with me. So when, in January 2020, I decided to design a roll and write, the basic mechanic was there—as was the theme.  

What’s in a Name?

I used to do some kayaking, so when I came to look for a name, Eskimo Roll was the first to come to mind. As a name, this was dismissed as culturally insensitive so I switched to Arctic Roll. Of course only the Brits will remember this seminal dessert from my childhood, but I can’t resist a good pun.
Making the most of the “dice” / “ice” pun potential the earliest version of the game, saw each player stock a “dice-flow” with 1 more dice than there are players, then pass these dice flows round the table drafting a dice from each. These dice were used to drop hunters, dig holes and move. 

What’s the (Power)Point?

I created a physical prototype in time for UKGE and playtested several times during that event.
Around this time I also entered Arctic Roll into a couple of design competitions. One was specifically for a print and play Roll and Write.

Another required a video. Now I would simply record a screen capture from Tabletop Simulator, but at that time I resorted to PowerPoint.

We are much more familiar with high quality editing of videos nowadays, but at the time I knew I could do what was required in a program with which I was already familiar.
Even now I see designers using PowerPoint for online pitches. Presentations can be much easier to script and control than a live virtual setup. Each slide can be a new viewpoint with a new arrangement of components. You can make text appear and pieces move around. You can add sound. With animation you can make everything do what you need it to do.
When finished, you can convert the presentation to a video format. Don’t bother watching the embedded video, unless you have a genuine interest in trying to use PowerPoint to produce playthroughs.

Virtual Playtesting

Shortly after, I began playtesting online. The vast majority of my design, development and testing is now carried out virtually in Tabletop Simulator, with my playtest group Virtual Playtesting.
I have written previously in this blog about the benefits of playtesting online.
https://www.thegamespeople.co.uk/playtesting-in-pyjamas-why-havent-we-all-been-doing-this-sooner/
https://www.thegamespeople.co.uk/getting-the-most-from-virtual-playtesting/

Perhaps one of the greatest benefits was the access to other designers across the world. I have lost track of who gave me input on which iteration of which game so I can really only express my indebtedness to the Virtual Playtesting community as a whole. You guys are great!

Ditch the Dice

One of the first significant changes that my playtesting community all agreed on was to reduce the number of dice. I wanted the game to be able to accommodate up to 6 players, and my first version, used 7 dice per player.  6 dice were sat on a ‘dice flow’ which floated round the table from player to player while they each chose a dice, before passing the flow to the next player.
This was essentially a card drafting mechanic from Sushi Go translated to dice, but 42 dice were not an expense that the game was likely to bear.

It is well worth considering production costs at the earliest stage of game design. Certainly games existed with 50+n dice—but they were generally big box games, not humble roll and writes. So I went back to the imaginary drawing board to find a new way to draft dice.

Dice Drafting

Looking back, with the benefit of hindsight, and aware of how the game has since developed, I can see that a Sushi Go type draft was not the best choice to use with dice.  The value of each Sushi Go card changes, as each hand plays out, and tension is created when you have 2 of the 3 cards required to complete a set but don’t know if that card is out there, or whether the player ahead of you will take the card that you need. 

With all the dice visible on the table, it was easier to see what might be available 2 or 3 turns ahead, but as some values  were inevitably better than others the tendency would be that the dice left at the end would be lower values that would simply not be as useful or as interesting. I think that would have created a cadence in the game which would have reduced interest just at the point where you want the game to be ramping up.   

So, I am very happy with the alternate draft which found its way into the game.

Increasing Interaction

Laying the dice out in ascending order, players make a choice which determines both the dice they want and their turn order for the next roll at the same time. Higher value dice would naturally mean that players would have a later choice in the next round. At this time, every player was still playing on their own ice sheet and this increased the sense of interaction between players, which is often difficult to establish in multiplayer solo games. Once this link to draft order was established it became so clear that this was an improvement that an alternative system was never considered.

The draft has of course continued to evolve. Adding bonuses to make low value dice better, or make it easier to use high value dice for movement, was so well received that it was a natural step to add a different bonus to each dice on the draft board. Playtesters often comment on the fact that this or that bonus is powerful and should therefore be further up or further down the track, but while there is certainly no science to it, and there certainly remains scope to rearrange the bonuses on different maps, there is a consideration of the value of the dice that are typically at the top or bottom of the drafting track. Bonus movement is typically available alongside dice of value 1 or 2, whereas the ability to change direction during straight line movement becomes much more useful when players have drafted a 5 or 6.   

Rolling Doubles

The final piece of the puzzle was to give players 2 dice (and therefore 2 actions) each turn rather than just one dice which they could use for either movement or placement.  Placement was fun, whereas movement was necessary and always felt a little bit of a down turn. The obvious strategy was to collect 2 dice adding up to 7, which always allowed 1 hole with 6 hunters around it, before using another dice to move on, rinse and repeat.

Giving players 2 dice with a requirement that they place and move each turn, allowed the game to progress more quickly, but also created additional decision points. “Which of the dice available do I use for placement, and which for movement?”

More importantly it broke the obvious sequence. It wasn’t always possible to fully populate each ice hole, creating an imperfect placement that players were required to optimise. This was further enhanced by creating a fish stealing mechanic (the Publisher wouldn’t let me call it “What an Icehole!”) which gives an attractive bonus if players are able to place hunters beside a hole that an opponent has already dug.

But where to get the second dice? The tried and tested Roll and Write solution is to make it a shared dice that every player uses. So the final player to choose their own dice also chooses a second dice—assigning this for all players to use.

A Cold Draft

So in Arctic Roll, when players make their choice, they are now in a position to consider;

1. the value they need for movement;

2. the dice face they want to place;

3. the dice they expect to be chosen as the common dice;

4. the bonuses associated with each dice;

5. the dice that opponents may want;

6. the affect it has on turn order – which can be critical when players are in close proximity to each other;

7. and the drafting order – the order they will choose dice in the next round.

All these factors create multiple micro-decisions and significantly ramp up the opportunities for interaction between players.

A cold draft indeed.

Area Cont-Roll

Around this time, one of my playtesters suggested playing the game on a shared ice sheet. This was such an obvious way to play the game that I tried this during the next playtest. It elevated Arctic Roll from ’just another roll and write’ with some novel mechanics, to a game which seemed to be somewhat unique. While I would be hesitant to claim to be the first roll and write to allow multiple players to play on a single sheet of paper, I am not aware of another one, and certainly this is unusual enough to justify becoming the primary way to play the game. This obviously dramatically increases the level of player interaction, elevating a game which revolved around an efficiency mechanic to a genuine area control game. I enjoyed the elegance of playing a game with a single sheet of A4 paper and a handful of dice, and never one to run away from a bad pun, particularly enjoyed the chance to brand Arctic Roll the first ‘Area Cont-Roll and Write’. 

Kill your Darlings but not your Polar Bears

The earliest prototypes began with a series of Calamities represented by a series of black dice on a shared ice flow in the middle of the table. There was a genuine drafting dilemma where players would have to forego one of the good white dice in order to take one of the black dice. The black dice all allowed a players neighbour to place something bad on their ice sheet (a little like the monsters added to the maps in Cartographers) but as the calamities were not all equally bad, players would have to choose how long they could wait taking the best ‘good dice’ without being left with the worst ‘bad’ ones.  It was a sort of reverse hate draft, or push your luck draft which I would be keen to reintroduce into another game, but it didn’t really work with the new drafting mechanics.

Polar Bears and Walruses survived the move onto a shared map. They started in the middle of the map, but could be moved around chasing away hunters or fish, thereby creating a further opportunity to mess with opponents. One of the drafting bonuses allowed players to harpoon the walrus and another to shoot the polar bear (sniping across the map if players were able to position themselves along the same line of hexes as the polar bears current position).  Some playtesters commented on the fact that they did not like the feel of hunting large mammals (fish were apparently fine) so bear spray and a beach ball to distract the walrus made a brief appearance, but overall the negative player experience of one player destroying something that another player had created was unpopular, so it was removed.    

By this time I was actively pitching Arctic Roll to publishers at UKGE and Essen…. But more of that in Part 2.

Where do you buy your shoes? Are they handmade by a cobbler or are they produced in a factory? Does it not matter to you that cobblers spent years honing their craft, so as to make boots that last a lifetime? Or do you just need something cheap (or perhaps ‘fashionable’) to put on your feet? 

What about bread? Do you care that bakers are few and far between now? Or Greengrocers? Or what about farmers? The families of farmers may have been caring for their land for generations – and now are finding it hard to do so with any sense of financial security.
Do they not all have the right to feed their families? Surely at least as much as someone who spent 3 years at art school?

Every day you go to a supermarket to buy food as cheaply as possible, you are telling the ‘market’ that price is more important to you as a consumer, than the livelihoods of other people. And while it may be easier to make ethical choices when you are buying luxuries like board games, rather than literal bread and butter, I would have far more respect for anyone taking the moral high ground who was consistent in making all their purchasing choices.

I’ll admit there are some people making noise about Chinese labour practices – but that hasn’t made much of an impact on the board game industry as a whole, or on the buying decisions of the majority of consumers. Are those that work within manufacturing not worthy of our concern as much as ‘artists’?

Are AI images ‘Art’?
Is it because artists produce ‘art’ that make them the exception?
I have seen lots of people declaring AI art ‘not good’. People love to point out weird anatomy, or bodies that  are out of proportion. The human eye is really good at picking up anything that doesn’t quite look right in those areas. But I have also seen plenty of artists get this wrong too. And no-one calls them out and tells them they aren’t artists. They just might not be very good at their job.   

Are all creative endeavors ‘art’? Art that is inherently worth more than making hand made shoes, or cabinetmaking a bookcase that doesn’t look like a Kallax?
While I would not diminish the skill of those able to produce digital images for use within games I am not convinced that a 2″ square image on a playing card really speaks to the human condition. It is a pretty picture that fulfils a function.

What if a publisher doesn’t want or require ‘art’ for a game, are they obligated to pay a class of labour that call themselves ‘artists’ to produce the pictures that they want on their components?
Should ‘artists’ be able to claim a right to produce all graphical representation? Is that in every context – or just in board games?
I didn’t go to art school. If I draw my own stick figures and stick them on my own game, is that OK? Am I thereby denying an ‘artist’ their right to earn an income? Which ‘artist’ exactly? All artists?  Or just a specific artist that I might have employed but chose not to? 

Do Board Games need ‘Art’?
I went to Mars University. Well actually I went to Leeds University, where I spent a good amount of time in the Roger Stevens building.

 

It didn’t take an artist to turn this into Mars University. It took someone moderately skilled with a graphic design filter.

The Terraforming Mars base set included 208 Project cards.
I am not surprised that they produced the pictures on the cards as cheaply as possible.
Prepublication, the publisher did not know whether the game would be a success or not, and it would have been potentially foolish to have spent large amounts of money on a game that might have been a flop.

The fact that the game was a success despite the 2″ square pictures not being terribly good, tells us that most of us don’t really need ‘art’ in our games. We just need a picture. Are we perhaps getting a bit above ourselves if we claim that all pictures within board games are ‘art’?

I am not suggesting that there is no artistry in board games. Or that the images used are never ‘art’. But most of the time they probably aren’t. People have long debated whether art that is produced commercially can truly be art. I don’t see anyone suggesting that images within board games are not commercial. 
Would a publisher that stopped using the word art and just talked about AI generated images face less of a backlash?  

Are AI images theft?
I’ll admit that this argument – of all the arguments against using AI to produce images – does give me some pause. 
The only examples I have seen, proving ‘theft’, appear to have been generated with prompts that were designed to generate that evidence. They present indicators or shadows of logos, or text, or signatures. But it does feel like a form of entrapment. If I ask an algorithm to produce an image in the style of a named artist then it is unsurprising if their personal hallmarks appear in the images. I have not yet seen anyone point to an image and declare that it has been generated directly from an image that they created. And I have certainly never seen an image that has been used commercially within a game, that has been attributed as theft.

Actually that’s not true.
There was the discovery that the artist, Jakub Rosalski, whose images created the world in which the board game Scythe is set (and in no small way contributed to the success of that game) was ‘guilty’ of tracing photos and the work of other artists. The furore died down pretty quickly.
https://www.reddit.com/media?url=https%3A%2F%2Fi.redd.it%2F16j5ltsmmbm01.jpg

The Genie is out of the bottle
It seems to me that we cannot go back in time to a world in which AI algorithms will not radically change the way we work and what we work on. In much the same way that automation has always done.
I don’t even know that I would want to.
I know that we are reaching a point where AI is better and quicker at reading a blood film or an X-ray despite the years of training that Haematologists and Radiologists have undergone. I don’t feel a great desire to allow them to make a living when my cancer diagnosis is on the line — but I would welcome a future where health care professionals have the opportunity to do more ‘care’ because they aren’t having to do mundane stuff that a computer can do for them.

I don’t know whether this tweet is the original source of this quote but I certainly appreciate the sentiment.

Humans doing the hard jobs on minimum wage while the robots write poetry and paint is not the future I wanted

— Karl Sharro (@KarlreMarks) May 15, 2023

But how is it to be avoided? I don’t think it will be by making the hill that we choose to die on, that of defending the right of artists to produce content for board game publishers who do not want them to.  

“When will AI take Your Job?”
I warmly recommend the writer, Tomas Pueyo. I discovered him when he wrote about how to manage the Covid crisis – and I suspect that his seminal article “Coronavirus: The Hammer and the Dance” (https://tomaspueyo.medium.com/coronavirus-the-hammer-and-the-dance-be9337092b56) genuinely saved thousands of lives.
More recently he wrote about “When will AI take Your Job?”, and once again he seems to talk a lot of sense.
https://unchartedterritories.tomaspueyo.com/p/when-will-ai-take-your-job

It seems to me that artists will continue to produce art. Good art and bad art. Art that inspires, and art that transcends the human experience. That tells us something about ourselves, or perhaps just looks great on our wall. Maybe it will even attempt to address the value we place on human creativity.

Other artists will continue to work in commercial settings. But they will likely find that those who are producing images for commercial applications, whether that be board games or product packaging, advertising or publication, can produce images that are good enough, quicker and cheaper if they use AI applications. And if artists are using it – and it seems inevitable that they will (indeed some already are) then it also seems wrong that anyone who is not an artist should be vilified for doing the same.

Vote with your Wallet or with your Conscience
“You pays your money and you takes your choice”. You are free to choose whether to spend your hard earned money on a board game that has been brought to you with the help of an artist; or a graphic designer or game designer, or an accountant or a person loading pallets on a dockside in China. But if your conscience is genuinely pricked by this debate then perhaps you should consider how many other purchases you make on a daily basis could be made with greater ethical consideration. There are a tiny number of artists that will be affected by this particular change, but there are very many more people whose livelihoods have been affected and or even eliminated as a result of our pursuit of cheaper food and goods and an ever increasing GDP.

This blog post explains how forcing players to take longer on their decisions (yes; increasing down-time) helped to remove randomness and apparent complexity.  

Fight, Flight Bite is a ‘dudes on a map’, area control game where you send your tribe of critters into the forest to fight over food resources. The ‘hook’ is that combat is resolved through an asymmetric application of Rock Paper Scissors.

There are 3 types of pieces on the board. Your own pieces, opponents’ pieces and food resources. In any duel, if the attacking player wins with Fight (Rock) they may move opposing pieces, if they win with Flight (Paper) they may move their own pieces and if they win with Bite (Scissors) they may move food resources. Thus, the design intent is that before any duel, players engage in the mind game where the attacker decides what they want to achieve, and the defender decides what they need to do to try and stop the attacker. But, of course, the attacker knows what the defender is thinking so chooses an alternative attack…and the defender knows that the attacker knows…etc

This was inspired by the combat in a miniatures game called Freebooters Fate. It was more complicated than RPS, using attack cards to decide which parts of the body you were aiming at, matched by cards played by the defender to determine which parts were protected. It was by far and away the most fun I have ever had with toy soldiers, as every ‘battle’ became a battle of wits against the opponent.

Following a test of the concept of an asymmetric RPS game at an UnPub event, FFB was developed virtually in Tabletop Simulator (TTS). It was a bit of a test of faith. It was hard to playtest virtually. Mechanically, I used custom dice, simultaneously turned to the desired gesture, but as is often the case in TTS it took much longer and bounced off a lot of playtesters.
However, when I played face to face, the game took a comfortable 30 minutes and players liked the use of the familiar in a novel way.

So, I started pitching the game at UKGE and Essen. Several publishers took an interest and asked for either a Print and Play or a prototype. As feedback from publishers began to come back, I also continued to playtest. Feedback from each was different but common themes began to emerge.

“Too random.” “Too complicated.” “Too chaotic.” “It was too hard to process so I just chose at random.” “The use of RPS suggests a light game, but this is too complex.”

Players who played strategically would consistently win against those who gave up trying and just played randomly, so the strategy was there, but some players were finding it difficult to find it. And the fact that some players did just give up, is never a good thing!
I began to take Occam’s Razor to my design, removing everything that was extraneous.

An interesting experience was just going for a long walk with my wife, when she indulged my enthusiasm and asked me about the game. As I explained the game to her, I realised a lot of the things that I still feel the game does well. But I also realised that the split-second decision as to whether to choose rock or paper, needed to be clearer.

There were various reasons why one or other might be incentivised.

First there was the immediate tactical position – what did the attacker want at any particular point in the game.

Then there was the terrain where the duel was taking place. Unsurprisingly, fighting on rocky ground gave a bonus to ‘rock’.

Finally, there was the attacker’s special ability – the intent was that different tribes of critters would have a different play style. Bullying bears would be good at Fighting. Birds would be Flighty. And sneaky Snakes would be great at stealing food from opponents.

As a player, I knew that a winning strategy was to find in-game situations where whatever the attacker wins with grants advantage, but as a designer I began to recognise that this defeated the design intent. During that walk, I realised that if the defender was to feel that their choice was relevant, it needed to be really clear what the attacker wanted. So, I removed all these additional layers.   
By the time I came to UKGE 2023 I had a lean mean version of Fight Flight Bite, which I was excited to playtest.

I met with a publisher who I had pitched the game to a year ago, and who had expressed some interest. I explained what I had taken out, and how I was trying to reduce the level of complexity.
“It is still too complex.” was their response.

So it was with some disappointment that I moved to another playtest with Jacob Jaskov, the designer of Fog of Love – a game I was very familiar with. I ran a stand to demo Fog of Love at several smaller conventions, pre-Covid. It is a truly unique 2-player game, part roleplay, part board game, in which players enact a couple’s relationship. Shut Up & Sit Down covered it in their podcast. (https://www.shutupandsitdown.com/podcastle/podcast-75-stress-testing-your-tarot/)

The relevant point however is that Jacob is an experiential designer. He focuses on what he wants players to feel, and chooses game mechanics which create these feelings. His BGG profile states, “Jacob Jaskov works with behaviour design: How do you affect people’s behaviour through interfaces, objects, physical environments and social structures?”

I happened to be staying with Jacob (thanks to our host Chris Backe), so on Saturday afternoon Jacob and I met up to playtest Fight, Flight, Bite.  

We played a 3-player game – doublehanding the 3^rd player – agreeing what we thought they would do, and alternating whoever wasn’t attacking to play the 3^rd player. Jacob’s feedback wasn’t very different from others. He found the interactions on the map interesting but didn’t like the RPS over the table. “Ditch it” he suggested, “and focus on the bits of the game that are more interesting.”
I pushed back;
“RPS over the table IS the game. It is ‘the hook’ and table presence that you don’t have to pay for. Both publishers and players have been attracted to this aspect, if I remove it, then I don’t think there is anything unique about my game.”

“I believe in the use of RPS. I have to find a way to make ‘that’ work. If I can’t then I don’t think I have a game.”

And this is where inspiration struck.

I can think back to specific occasions, playtesting several of my games, where suggestions of very minor changes, have immediately and significantly impacted how a game plays – where another designer was able to remove a roadblock or to ‘fix’ a game. For instance, allowing multiple meeples on a single space instantly fixed a 2-player worker placement game that I was struggling to make work with 3 or 4.

“Well, in that case, do RPS beneath the table.”

“?”

“Make the sign, beneath the table and then reveal simultaneously. This will give players the opportunity to look at the ‘map’. Look at the other player. Look at the game state. And then make their decision. System 2 thinking. Not system 1 thinking.”

The whole concept of System 1 and System 2 thinking comes from a book by the psychologist Daniel Kahneman. System 1 thinking is the stuff that we can do – almost without thinking (it is fast, automatic and unconscious). Whereas System 2 thinking is engaged when we make plans, or think about consequences.
https://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow

I didn’t even have to playtest this suggestion to know that this just made sense. We immediately tried it a couple of times, but it was clear that while it would change nothing mechanically, this would have a very significant impact on how the game would ‘feel’.

It was also an obvious addition to replace the cards played by the attacking player, with a gesture made with their other hand. So, they raise one hand as a ‘tell’ or a ‘bluff’ or a ‘double-bluff’, before both players then reveal Rock Paper or Scissors. If the ‘bluff’ turns out to be true then the attack’s strength is doubled.

With marvellous serendipity, Jacob happened to know the publisher I had just met with. By the time I went back past their stand, he was animatedly explaining the change to the same person who an hour earlier had told me to reduce the complexity.

“When can we play it again?” he said.   

I’m not sure what the takeaway from this story might be. Or whether there is anything here that other designers will be able to learn from.  
Using RPS is a pretty unusual mechanic, in that it relies more on ‘muscle memory’ and system 1 thinking than most decisions taken within board games. What I hadn’t realised was that this would bypass the strategic decisions that a game presents to players. I can’t think of another of my designs – or even of another game – where this is an issue.

Most board games – by design – engage system 2 thinking. We talk about meaningful decisions and the decision space in our games, and this is what we design to include.
One aspect of games such as Dobble (Spot It!), Ghost Blitz or Jungle Speed is the juxtaposition of requiring an instant system 1 response, to something that actually requires system 2 consideration from most players. The handful of players for whom these games require only System 1, who can recognise these patterns almost unconsciously, invariably win without effort. Since she was about 6 years old, I have never seen my daughter lose a game of SET – even when playing against everyone else in the room.

However, this does encourage me to think more about the experience, of those playing my games.

To be truly honest, I am just excited to have made this step, and grateful to those who helped me make it. I just wanted to share it.  

I have started another blog series, over on the Virtual Playtesting website.
To quote myself,
These are intended to be a series of short posts about playtests. They aren’t intended to be full reports, or extensive critiques of the games played. What I will attempt to do is pull out one or two things I liked about the games tested, and then one or two issues that were brought up in the feedback at the end of the game.
Ideally, I will try and draw attention to lessons of general relevance to a range of different games so designers can apply these points to their own designs. 

This is perhaps a design brief – or mission statement, but there are other reasons.

I think blogging is potentially a great way to showcase the playtesting we are doing within the Virtual Playtesting group. I am proud of the quality of games that we are seeing playtested and the quality of feedback that designers are receiving.

I know that writing down my thoughts is a good way to force myself to think more deeply about something. I think it will help me to become a better playtester and to provide more useful input in future.
Although these posts aren’t intended to be a designer diary, as I am usually present during playtests of my own games, and I do not need to worry about hurting the feelings of others, I expect my own games will feature disproportionately.

https://virtualplaytesting.com/2020/07/pocket-playtests/

Having been involved in a large number of online playtesting sessions over the past couple of months I have a few thoughts on how we can all get the most out of virtual playtesting sessions.

We all (well, many of us) have a lot more time on our hands; that is one of the reasons virtual playtesting is on the increase. But time remains a limited commodity. If 4 designers meet to playtest each other’s 4 player games, and each game takes 2 hours that is potentially a mammoth 8 hour playtest session. I appreciate that many of us have more time on our hands at the moment, but anything we can do to cut down wasted time, will benefit all of us, and will in fact lead to better feedback. I have lost count of the number of playtests I have been involved in, where the feedback has been dominated by how long the game took, or how slowly it played.

So, a bit like my recent blog post on how to give good feedback, this post is first as a reminder to myself, about how to respect the time I am asking of fellow playtesters. However, I think that some of these suggestions are based around smoothing the overall process and trying to reduce time being unnecessarily wasted. They certainly are not intended to be rules – but I have framed them as questions that I might want to have thought about, going into a virtual playtesting session.

1. Are Steam and Tabletop Simulator up to date?
2. Have I set the table up already? And for which player count(s)?
3. Have I been realistic with the requested time slot?
4. What player count do I need?
5. What is the purpose for this playtest?
6. How am I going to teach the game?
7. Am I committed to managing the allotted time well?
8. Is the ruleset fixed? What should I do if the game is broken?
9. How will I be managing discussion?
10. How do I want to hear feedback?
11. After Feedback

Before the Playtest

Are Steam and Tabletop Simulator up to date?

As online traffic has increased dramatically over the past couple of months, in order to save bandwidth, many programs on Steam have changed the way they offer updates. Rather than filtering those down continuously, they are now only downloaded when the program is opened. I have lost count of the number of times I have tried to open TTS only to find that it has taken 10 minutes or longer to update and open.

I would suggest anyone joining an online playtest group, opens Steam and Tabletop Simulator (and even Discord) at least 15 minutes before the scheduled time, and if you have not opened either recently, maybe allow even longer.

Have I set the table up already? And for which player count(s)?

I have joined some playtests where playtesters arrive to see neatly stacked decks of cards and tiles and the first 10 or 15 minutes is taken up by the designer setting up the game to play.
Tabletop Simulator offers unlimited save points such that it is easy to save a different set up for each player count in the game. You can create a huge play table and have the game set up for different player counts at each end. You can create tutorial areas with duplicated components so you can show players what the game might look like.

You do not need to assume that players are sat around a table. Think about it. If it is easier to read player boards a certain way up, then everything can be the same way up for every player.
Don’t treat the virtual space the same way as your kitchen table and don’t pretend that you only have one virtual copy of the game which needs to be put away at the end of each session. 

Have I been realistic with the requested time slot?

Playing virtually always takes longer. I’d estimate at least 30-50% longer than playing physically. If we add on time to explain the rules and time for feedback at the end then you could easily expect a playtest to take twice as long as playing a physical game. We all tend to underestimate, but we should all try and get better at this.

An indicator that you have taken the limitations of virtual environment seriously is to compare the length of an ‘in-person’ playthrough with the time requested. “My game plays physically in about 45 minutes. I’d expect the playtest (with feedback) to be over in about 90 minutes.” 

What player count do I need?

This should depend on where a game is in development.
I think we all want to get as many people as possible to play our games – that is why we design games. And we also want as much feedback as possible – that is why we playtest games. But more players take longer and during the earliest stages of playtesting it is often the case that feedback tends to reach consensus.

At this point I can’t help but look at the maths…
If a game takes about 20 minutes per player then a 3-player game will take 60 minutes and a 4-player game 80 minutes. But in terms of the committed resource of playtester time, the former takes 180 player minutes and the latter 320 player minutes. And a 5-player game, 500 player minutes! This means that a 4-player playtest adds 140 player minutes to a 3-player playtest and the 5-player game adds 320 minutes (i.e. a whole new 4-player playtest!).
If you are going to try and put as much into the community as you take out, then you could ‘owe’ 6 to 8 hours after a 5-player playtest.    
Unless you are really looking at scaling, do you need the full complement of players? As a designer you might legitimately have a reason for playtesting at a particular count, but if you are at the point where you need to test with 5 or more players, I would expect the gameplay to be very solid at lower player counts, and the focus should be on minor tweaks to adjust for the number of players not basic mechanics.

What is the purpose for this playtest?

It might be the case that I just want to get another playtest under my belt, or have other designers cast a critical eye over it, but depending on the stage of development, it can also be helpful to focus on a particular aspect of the game. Do I have a specific part of my game that I want to test?
Have a look at what Ignacy Trzewiczek says about the differences between playing and playtesting a game. 
He also distinguishes between the objectives that he will share with playtesters and those he will keep to himself.
Or similar thoughts from our own David Digby here.

You might want to ask one of the playtesters to fulfil a particular role “Can you please try and get as many points as possible from the corn strategy?”

“I’d like it if you can just focus on combat and preventing other players from trading.”

This can make your playtest immensely more valuable, but it also has the additional benefit of giving your playtesters (who may well be designers in their own right) something valuable to do. No longer are they just making up the numbers, but they are taking a role in the development process. They have a job to do and something to report back on.
As a playtester given such a task I would not feel like agency was removed, I would feel valued and empowered.
In addition, players tasked with a specific objective might not need to fully understand the entire ruleset to make meaningful decisions. I do not need to understand the nuances of the trading system if my role is to try and stop other players from trading by blowing them up.

During the Playtest

How am I going to teach the game?

This is sometimes the biggest time sink in a playtest. I have spent 45 minutes listening to rules before I ever touched a card.
I have shared before, that I feel the most useful piece of feedback – something that a designer can immediately learn from and put into practice – is how well did they teach the game. And frankly most of us aren’t great at it.
One reason for this is that we are trying to give other designers as much information as possible to play the game.
I think that we need to approach early playtests, especially those conducted in a virtual environment, with the mindset of a demo rather than a ‘teach’.

First however, I should admit, I am bad at learning the rules of a game. Really bad!
Actually, I think I do myself a disservice. I am really bad at learning the rules of a game the way most people teach. When people just point at different parts of the board, explaining what goes here or what that does I involuntarily switch off. In the virtual world this is even more the case. I cannot always see where the ‘teacher’ is pointing, or even looking. I have learned that if I wait until the game actually starts, I can usually learn the rules during turn 1 when I start to see things moving round the board. I usually accept that my first play is a bit of a learning game. Although I am fully aware that many players want to know everything before they start I think the best way to teach everyone at the table (whatever their learning style) is to control a turn or two, taking people’s actions for them and explaining why and how.
This tends to be the way the pros teach, or the way publishers will run a good demo at a convention. I’d even suggest the fact that so many people gravitate towards the pro teachers on YouTube who generally tend towards my preferred methods, suggests that I am not in the minority.

I think one of the reasons for not wanting to do this is to give players their own agency. We don’t want to take their turns for them.

But;

• Done well, I could demo 2 turns each of a 4 player game, in a few minutes. If players did not like the position in the game they have arrived at, we could reset the game and start again, in less time than most rules explanations take.   
• This takes us through the less interesting parts of a game very quickly. Many games tend to get more interesting from turn 3 or beyond. Unless your game is an exception, or you really need a playtest to focus on the first turn or two, then you might find this gets you to the best parts of the game, and the most useful part of a playtest, more quickly.
• Finally – but perhaps most importantly; Playtesters are not here to play your game. They are here to playtest your game. If the two are not different in your mind, then you might want to reconsider.   

Am I committed to managing the allotted time well?

Many playtests overrun. It’s expected. But should we collectively aim to be better at managing our time?

Knowing when to call a game – either because the playtest has yielded enough information or because the game end has become predictable is a useful discipline to exercise. And it is far better to call a playtest early than wait until playtesters have become frustrated and bored.   

Also, managing the turn structure can be helpful. Encouraging players to say “ turn” when they have finished, or encouraging the next player to start, while the previous player is still dealing with the steps of moving components around, which become more time consuming in a virtual interface, can help the game move at a clip.
This does two things – both of which are to your benefit as a designer. First it avoids wasting the limited time you have – both yours and your fellow designer/playtesters. Secondly, it helps the game to play almost as fast as it would in the real world. This means that comments about the pace of the game and how long it feels are less likely; but if they are made, they are more likely to be relevant.

Is the ruleset fixed? What should you do if the game is broken?

I love changing the rules of a game during a playtest. It makes me feel that the playtest is achieving something – especially if the new version works better. Obviously, this depends on the game, where you are in development and whether the needed changes are clear, but if there is an obvious problem, pursuing the playtest to the end might not be the best use of time, and making a direct A B test within a game can generate immediate and useful feedback.

After the Playtest

How will I be managing discussion?

I have already written a blog post about how I try and give feedback. As the designer, and host of a playtest session, you are in a position to control the way playtesters respond to some extent.
Some discussions can start during the game – especially if they give players something to do when they are not playing. Asking about game play choices and decisions (often revealing the secret player information that players would usually keep to themselves) seems to be perfectly valid during a playtest. John Brieger refers to this as a “Think-Aloud Protocol” in his playtesting talk to the Ignite Conference.

However, if discussions begin to slow down gameplay then they should probably be shut down until the end. That’s not to say that the end won’t be brought forward; but if players have to wait 5 minutes for their turn because the player preceding them is carefully dismantling your action point mechanism that is probably not a good idea. 

How do I want to hear feedback?

When the play finishes can be when the playtest ‘proper’ starts, and how the designer manages the post-match analysis can be important.
Some designers may just sit back and listen to the discussion at this point, but that is to give up the opportunity to control the way this discussion takes place, and even the feedback that may be received. It should not be seen as a negative if the designer wants to impose a certain order on proceedings, rather than opening the floodgates to whatever the loudest voices have to say.
Developing a local culture within any playtest group will be helpful to all.

If the designer is trying to control which feedback comes when, they should outline this at the start – playtesters will be more likely to sit on their thoughts, if they know that they will have an opportunity in due course.

A general introductory question can help start the ball rolling, if players aren’t immediately forthcoming. I like to start with “How could I have taught the game better?” or “Was there anything in the rules explanation which didn’t make sense at first?”
This can also help set a tempo – moving round the table and giving each playtester the opportunity to speak; thereby avoiding one playtester dominating the discussion and leaving very little for others to say.

If the discussion allows, then I will continue to work through specific questions, managing the input from playtesters and asking for input from those who have not spoken.
If you have given playtesters a focus at the start of the playtest this may be easier – if they have been asked to run a particular strategy then they will naturally comment on how that part of the game felt, and may in fact be the only person able to give feedback to that specific aspect.

I don’t think I have been involved in a playtest where at some point, the discussion does not become more free form, as participants unload all their ideas about how the game could be improved. While it is important for the designer to listen to all input, it is not necessary to be defensive or justify design decisions you have already made. It is well known that while it is important to try and understand how your playtesters feel about the game, and why they may feel that way, there is no obligation to accept the solutions they offer.

After Feedback

It goes without saying that you should have written notes during the feedback process.
Playtesting should not be about just ticking off another play of the game, but about improving it. I’m sure I am not alone in having playtested a game, and then forgotten the feedback received until it was raised again in a subsequent playtest. You might not iterate after each playtest, but if a playtest has revealed nothing that you want to reconsider then you probably aren’t playtesting hard enough!

Finally, don’t forget to thank your playtesters.
Processing the advice given on the spot, suggesting what you might try in a subsequent iteration of the game, is one way to show your appreciation.
But perhaps the best way to really show your appreciation is to fully engage with the community where you are playtesting and immediately jump into another designer’s game. It doesn’t happen very often – but you don’t want to be that playtester who shows up, plays their game, and then vanishes.  

This is the first in a series of blog posts about virtual playtesting. Others here;
Playtesting in Pyjamas
Giving Playtest Feedback
Getting the Most from Virtual Playtesting

Since writing this I have been involved in establishing the Virtual Playtesting Group. We meet on Discord on Thursday evenings (6pm UK time) until late, and playtest in Tabletop Simulator or Tabletopia.
With moderators in time zones from Eastern Europe to the Pacific, we welcome you to bring your online prototypes to playtest with other designers.

Here is an invite to the Discord Server.
https://discord.gg/Ze9mBWc