Master of Spies
An interview with SpyNet designer Richard Garfield
Can you give us a bit of the history of SpyNet? How did the game come to be?
In the 90s I was playing around with a lot of different draft styles for Magic: The Gathering. One of my favorites started by laying down a line of face down cards, as players alternated peeking at them, and either choosing to take them or going to the next card. Another card would be added to a pile of cards that had been looked at, so that piles of rejected cards would grow until the collection became valuable enough to take. The method of drafting was fun – involving pressing your luck, and a bit of hidden information.
Soon, I made a game which was played with a regular deck of playing cards and involved that style of drafting. The deck I was using had a Winston cigarette logo on the back, and so the game became known as Winston, and the draft was known as Winston drafting. That game is the great-great-granddaddy of SpyNet.
Was the game always about spies or was that something that was decided during the development process?
When it was Winston, the game was abstract. One early iteration of the game was called Hollywood Boulevard, and players were drafting movie roles and trying to get the most impressive filmography. It wasn’t until I had been playing with the concept for about 17 years that I started using the spy flavor. The flavor really went well with the style of play, and the game felt much more complete than the many iterations that preceded it.
SpyNet has a very distinctive art style. Can you tell us a bit about how the game got its look?
The prototype I was working with used black and white art from graphic novels, including Sin City. I liked how the black and white made the color in the cards stand out, and I talked to Z-Man about working with that look. They did that, and Ross Curtis, the artist, really used it magnificently.
What they did was use color elements on the cards that corresponded with the suit – so it was black and white with a green border and a few green elements, for example. This worked better than I could have hoped – because it allowed for a lot of fun detail in the look without interfering with the game play at all – since the image did not interfere at all with the game play elements. And they did have a lot of fun with the art – for me it has the perfect blend of seriousness and humor – I have laughed out loud more than once when I look at some of the details in this weird spy organization they have created.
What are some key concepts central to SpyNet and how are they implemented in the game?
The main mechanical idea is the Winston Draft, and the idea that each turn you can draft or play. To make the decision to draft or play interesting, it is important that players get an advantage for having more cards when they choose to play – which means they should draft a lot first – but also an advantage for playing first – which means they can’t just draft in the first half and play in the second half.
Also, drafting is much more interesting when cards have varying values. There is a knee jerk reaction these days to making cards the same-ish in value, and we had to fight that instinct – cards have to have a range in value such that players are correct to sometimes go fishing for better cards even if it means their opponents are going to get multiple cards.
Were any spy novels or movies a particular influence on the game?
I don’t think there were any particular influences. I was picturing the game taking place in a city like Vienna where historically many different powers had spy representatives. In fact – the original name of the game was City of Spies.
What are some of the major differences between the two-to-three player game and the team game with four players?
Two-player is a fast, tactical experience, you get lots of action, and likely every branch will be in play for both players. With three players, it is unlikely you will be competitive in every branch – so you will be trying to control the ones that the other two players are putting less pressure on – let your opponents spend their effort outdoing each other, if possible!
When you have an ally – in 2 vs. 2, you will have to rely on your partner to cover the branches you can’t handle. I think it plays best where strategy is not discussed during play – where players have to figure out the best plays on their own – reading partner’s intent through their plays and passes.
Did you run into any problems while you were designing SpyNet? How did you resolve these issues?
Not really. My biggest problem was finding a publisher willing to try a game that was best played with teams of 2. Once Z-Man was on board they had some development ideas that improved the game in many ways. One example of that was removing the missions which had reveal effects; which players often forgot about, and instead focusing all the special effects on the spy cards. Another is the addition of the multicolor missions.
I guess one thing we went back and forth on was the first player advantage. There were a number of suggestions for fixing that – which all failed in some way or another. Eventually adding on a single card reverse draft at the start seemed to satisfy everyone, and had a certain elegance, so we went with that.
Can you tell us a bit more about the unique card drafting system at the heart of SpyNet?
There are 3 face down piles of cards - at the start of the game there is one card in each pile. When players want to add cards to their hands they can look at the first pile, and take it or move on to the second. They can take the second pile or move on to the last pile. They can then take the last pile or take the top card of the deck. Then all the piles that the player looked at are augmented by one card, and the pile they took – if any – is replaced with one card.
That is the entire draft system. Drafting feels spy-like – as a player can look further and further and find hidden secrets the opponent doesn’t know about. There are some fun tactics you can come up with, for example, if you look at the first pile and it has an ok card but you decide to go further, you might find a good card. You can choose to take that good card, or you could gamble that your opponent will take the first pile and the good card will be there on the following turn for you with a bonus extra random card.
One of the interesting features of how the drafting interfaces with the play is that the value of the cards changes a lot during the course of the game. For example – at the beginning a player could reasonably take any card in the first pile rather than give someone two cards – even a level 1 agent with no special powers is enough to give you dominance in that branch. As the game progresses, that level one agent might become worthless or might increase in value depending on how the brnaches are contested. There are games where even midway through I might pass up a level 4 agent, which is usually an automatic pick for me, because I dominate the branch – or am losing the branch so badly – or I have no intel left to play in that branch and don’t think my opponent has time to use it effectively.
What did you have in mind when you were creating the situation cards?
The situation cards were one of the improvements added by Z-Man, probably (Head of Studio) Steven Kimball. I am generally supportive of incorporating simple rules changes that can be revealed to give variety to a game, and the ones they settled on do that well, they are flavorful and really mix things up.