food, bed, games

Playing to win, the only way to play

…Or The Spike Trap

When I play a game, I love to explore.  I like to see what’s possible, to try out strategies others think are bad to see for myself, and to generally play in a way I think is fun.  I have a sort of “achievement unlocked” mentality.  I don’t often win, but I usually have fun.

Mark Rosewater, the lead designer of Magic: The Gathering, wrote an article once about different types of players.  In his classification, I would be a Johnny because I like to “win with style”.  So given that, I clearly do not always play with the sole intent of winning, why did I use such a click-baiting title?

Extra-Game Objectives

Although I do not play with the sole intent of winning, I realize that often the “achievements” I’m unlocking, such as trying to win in a particular way or with a particular strategy, are extra-game objectives.  All the actions I take are allowable in the game system, it’s just that they aren’t necessarily the stated goals of the game.  For example, sometimes I like playing a theme strategy, or playing a particular faction.  I’m usually more drawn to the evil ones, and so if I play a new game set in a fantasy world, I’ll pick the necromancer character or the undead if possible.  I saw the movie Army of Darkness as a teenager and thought it was awesome, and my love of commanding an army of damned souls grew from there.  It’s not because I know anything about the game or think they are the best, it’s simply because I like the theme.  Liking the theme isn’t usually part of the game though.  It’s not a requirement in the rules to only play factions you like.  That’s what I mean by an extra-game objective; it’s as if I’m adding that objective, play a faction I find interesting/cool, to the normal rules for picking a side.

Playing to win is the only way to play for definitional reasons.  The rules of the game are about how to win, not how to have fun roleplaying as a necromancer.  Playing to win isn’t the only way to have fun, just the only way to play clearly given in the rules.  I need to state an important caveat that I am only talking about a subset of what people consider “games”.  There is no way to win playing with a yo-yo, but to me, I don’t think just doing tricks with a yo-yo is really a “game”.  The discussion of what is and is not a game is a bigger issue than I can fit in here, and I fully acknowledge that what I’m saying might not apply to toys like yo-yos, roleplaying games like Dungeons and Dragons, “games” where there are no decisions like Candyland, or other things sometimes considered “games”.  What I say “game” I mean games of strategy, where players can make decisions.  This would include classic games like Chess, older board games like Sorry! or Clue, or more modern games like Settlers of Catan, Agricola, or Through the Ages.

Some people have countered my argument by saying that the rules don’t cover how to win or the best strategy to play.  Therefore, playing to win is an extrapolation, and so is playing thematically or for some other reason.  In other words, the argument is that the rules put bounds on players actions who are playing to win as well as the player playing thematically and don’t suggest any particular way to play.  I am sympathetic to this argument, and I can understand people seeing it this way.  I accept this view as equally valid as my own, even if I don’t agree with it.

My response to the argument that the rules don’t necessarily imply that playing to win is the main way to play involves an analogy.  Imagine I say, “I go to my job to earn money.”  That makes sense because it’s generally the accepted purpose of a job; most of us think of a job as a way to earn money primarily.  Now imagine I say, “I don’t care about the money, I go to my job for another reason.”  There are many possible other reasons I could go to my job;  maybe I enjoy the work in and of itself, maybe I made a promise to someone I’d go, maybe I’m preparing for a role in a movie, etc.  You don’t know.  That’s because earning money is an outcome we can both observe.  It’s objective.  If I’m doing things for subjective reasons, things you can’t observe, it’s harder to understand.  In a game, the rules define the objective goal, the analog to “making money”, in the section on how to win.  If I’m not pursuing that goal, then it’s not as clear what I’m doing.  It is true that the rules lay the foundation for playing thematically, but they (usually) don’t imply what makes for good subjective decisions within that system.  On the other hand, the rules do imply what makes for good decisions in relation to winning.

Playing the Same Game

Players get to do what they want in a game.  Playing the undead is fun for me, and so I do it when I can.  My point isn’t that playing for reasons other than winning is bad or a less valid way to play.  It’s only potentially a problem when I’m playing for theme and my opponent is not.  Part of what makes a game work is that both players agree to the same rules.  If I’m not playing to win, and the other player is, then we’re playing different games.  This mismatch can create problems for both players.  My playing-to-win opponent, Spike, might become frustrated his flavorless knights keep soundly defeating my undead, or I might become frustrated that Spike doesn’t don an appropriately thematic costume or talk in an accent.  Most of the time, this kind of mismatch isn’t a big problem because designers create systems that are robust.  A game where all factions have a chance to win helps keep the game competitive for Spike, even if I’m secretly playing so that I can wear my dark robe with the gold-embroidered hood and chant incantations while playing my turn.  In other words, I can pursue my extra-game objective of playing thematically and not ruin the competition.  Ideally, playing thematically is also a good strategy.  If playing the undead well in a strategic sense simulates or captures some of what a player imagines commanding a horde of zombies and skeletons would be like then it works even better.

Advertisements

Ameritocracy: Evolution

Ameritocracy is a non-collectible, strategic card game I’ve been working on for the past two years with a friend of mine.  In the game, you are running for President of the United States.  You have to get supporters, money, political operatives, zealous but perhaps naive volunteers, and out maneuver your opponent to get enough votes to win.

The first version of Ameritocracy wasn’t politics-themed.  Its beginning was a game you could play with Tarot cards.  We liked the idea of what the four suits in a tarot deck could represent, each could work as a type of power a government or institution might use; swords were military power, coins were economic power, cups were religious power, and rods (turned to crowns) were political power.  We talked a lot about what those types of power represented and eventually turned our game into one about building a city.  Each player built a deck, representing all the potential resources in their kingdom that could be used to build their city.  Each turn, they drew cards and constructed buildings to make their city more awesome or to attack their opponent’s city.

There were two main resources in the game, gold and population.  Gold could be generated each turn, but it didn’t stick around.  Also, if you ever needed to pay a gold, but didn’t have any “floating”, you could discard a card for one gold.  Population stayed from turn to turn.  The goal of the game was either to amass enough culture, represented by a boxy icon on the bottom of a card, or to use military cards to attack your opponent.  When you attacked your opponent, you did damage, and for every 1 point of damage, the opponent put a card from their deck into their discard pile.  One cool twist was that each card could be played right-side up, as whatever it was named, or upside-down as a generic population.  Population could be turned sideways to indicate they were “working” on producing something.

Screen Shot 2015-09-05 at 3.42.36 PM

Bank was one of the first cards I designed.  It takes an investment of 3 gold and 0 population, and returns 1 gold per turn.  It provides 0 culture, but if you turn it into a worker, it has a population of 3.  The population numbers originally ran from 1-3, with a few 4s.  Each turn, you could choose to either draw a card, or “populate” by flipping the top card of your deck over upside-down and using it as population.

This card seemed to work ok.  We thought the best cards should all be 1s and the worst cards should be 3s.  We thought Bank was a fair deal, but nothing special.  Part of why you would play this card was that it was worth 3 population.  The problem was that the military cards were way to good.  Our mechanics around combat sucked, and the game was really unfun.  Still, the basic resource mechanics seemed to be ok.

We play tested quite a bit at this stage.  We decided to first focus on just the “coin” cards and try to make those work.  We envisioned that these cards would “ramp” up their production, produce a lot of culture, and win the game that way.  We thought it was more straightforward than the military costs, and we could do some math to figure out how much stuff should cost.  Since it was relatively simple to figure out how much culture was needed to win and how many resources a player could generate each turn, we could just do the math based on how many turns we wanted the game to go.  After about five major iterations of the game where we completely revised everything, we arrived at a new version.

Screen Shot 2015-09-05 at 3.37.03 PM

In this version of the game, buildings required you to pay costs to use them.  We also removed the idea that gold could “float”.  In this new version, gold costs were always paid by discarding cards.  We decided we wanted everything in the game to be represented by a physical object, a card, a deck, a counter, etc.  We didn’t want anything that players had to actively remember.  Bank now required 2 population to work to produce it and 2 gold.  Then, once it was built, you could spend 1 gold (discard a card) and have 2 population work in order to draw 2 cards.

This version of the game seemed to be more stable, but it was lacking something.  Although a game could be played, it wasn’t much fun because there wasn’t any player interaction.  We imagined that some of the other suits, like swords and crowns, would be more interactive, but we didn’t know how.  We moved on the designing the cup cards, and since those cards were based around religion, we wanted religions to be represented somehow in the game.  It felt weird to build them like buildings, so we created a new card type, Movements.

 Screen Shot 2015-09-05 at 4.04.41 PM

Movements went into your deck and you played them like other cards.  The difference was that instead of being just in your city, they went into the middle and both players could use them.  The Meek Shall Inherit the Earth is a good example of an early movement card we had.  The three boxes affected players differently.  The first box was general text about the movement itself, the second box was just for the person who played it, and the third box was activated abilities any player could use.

We really liked how each player could use the movements.  It added some interaction to the game because now one player’s cards could really affect the other player’s, but there were problems.  First, these cards were pretty confusing.  We could have maybe solved that with better graphic design or smarter text, but they weren’t just hard to understand.  Play testers wondered why they would use a card that that opponent could use.

Furthermore, we were beginning to feel dissatisfied with the resource system.  With so many cards on the table, it was hard to remember which population was working where.  We wanted to try some new things, to try to push ourselves past our comfort zone.  We’re both Magic: The Gathering players, and so having our population “tap” to play things felt a lot like lands.  Since the beginning, like many other designers, we had been tempted by trying to improve Magic’s resource system.  I think it’s more difficult than it seems at first.  Still, we had some ideas.  First, we changed the way activated abilities worked.  In Magic, and many other card games, a player pays a cost to get an effect.  This is usually written like “<cost> : <effect>”.  Although we were still in that mode of thinking, instead of paying work, our equivalent of mana, we instead required players to move a population to the building, which we called “specializing”.  You could only specialize a building so many times, and we noted that by having “S(X): <effect>” where X is the number of times you could specialize a building.  Now, each turn, you either drew a card or populated, then could either specialize or despecialize, and then built buildings.  You also didn’t tap population to play cards.  If you had enough population, you just played the card.  Theoretically, you could play infinite cards with the same population if you could draw them and pay any other costs.

Screen Shot 2015-09-05 at 4.12.01 PM The second major thing we did was change the way movements worked.  Movements were removed from the kingdom deck and placed in their own deck, and instead of having to pay costs to use them, you simple specialized them.  They did different things either when they were specialized or when they hit a certain threshold.  This change was huge.  They created a neat tug-of-war mechanic where players battled over the cards in the center while trying to build up their cities.  However, the game was still pretty complex.  There were often cascades of triggered effects that were hard to remember and sort out.  There was a ton of board complexity.  The biggest culprit were buildings that either did stuff automatically or triggered off of something else.  A seemingly straightforward card that read “When you populate, draw a card” would combine with a card that said “When you draw, populate” for infinite loops and other problems.

Around this time, we were play testing the game with some friends, and one of their roommates said, “Hey, you know what your game kind of reminds me of?  A political campaign.”  We looked at the cards, looked at each other, and started brainstorming about this new theme.  Although the medieval town/city theme fit some stuff, it was pretty cliched.  This new theme was inspiring, and wasn’t territory any similar card game had gone to our memory.  It also helped us finally abandon some things that we thought were important about the game, but really were not.  First, the suits were extraneous.  Although they had been a useful guide in the beginning to get some stuff down on paper, they had no actual effect in the game.  We had talked about ways to make them more like colors in Magic, or limit the number of suits you could have in a deck, but we never got a solution we liked.  In the new politics theme, there were no suits.

Also, the game had been moving farther and farther from the original play test, where the knights would ride over, destroy parts of the kingdom, and put some cards in the discard pile.  Counting your deck to see how much more damage you can take was a chore.  Although I loved the idea thematically, because it felt right that when your kingdom was attacked you lost potential future resources, it never played well.  So we sent the knights home.  In a city building game, there is a real expectation of warfare.  Most fantasy games have some sort of combat mechanic, and it’s part of what makes the fantasy genre exciting.  Moving to a political theme fit our new direction.  It matched our desire to have a game based more around building an economy and using subterfuge, dirty tricks, and soft power to outmatch your opponent.  The political campaign fit that idea quite well.

Screen Shot 2015-09-05 at 3.38.40 PM

So Bank became Political Action Committee.  A lot of other mechanical changes happened too.  First, the upper left now has a number that shows how much support (the new word for population) can be staffed (the new word for specialize) to a team (new word for building).  This was a great change.  Instead of limiting the number of times you could staff a team, we limited the total value of supporters.  This created a cool dynamic where supporters with a value of 1 became a lot better.  You could activate a teams ability no matter what value supporter you staffed to it, and so you didn’t want to spend a 3 on it.

We also cut all the cards that triggered automatically and removed the option to despecialize/unstaff supporters during your staffing phase.  This simplified the game greatly.  In this new version, teams only had an effect when you staffed them, and you only could staff to teams, never unstaff them (with a few exceptions).  In addition, movements were now headlines, and they all worked similarly.  You staffed them and when either you or your opponent got enough support on them, you won a certain number of votes.

Playing this version was the point when it started to feel like a real game.  It had been playable before, and we had enjoyed it, but when we watched other people play it, we saw them get confused and/or annoyed and/or sad and/or quit.  This new version seemed fun for people to play.  We play tested it a lot with each other and with our friends making tweaks, clarifying text, adding/cutting cards to balance the game.

Screen Shot 2015-09-05 at 3.39.46 PMScreen Shot 2015-09-05 at 4.39.59 PM
Here is Political Action Committee and an example headline now.  We have printed cards for our first blind play tests (where we mail cards and the rules to someone who won’t have us to explain it) and are going to expand our play testing more broadly.  The game feels pretty fun to us and most people who’ve played it seem to enjoy themselves.

It’s interesting to me to look back and see what never changed.  If you look back at the first version of Bank, it has the same mechanic of the upside-down number in the lower right.  In a lot of ways, the evolution of this game was about peeling back layers and finding what was actually interesting about that idea.  In a follow-up blog post I’m going to expand on this idea more, explain our recruiting and staffing mechanics, and this idea of refining the core mechanic of our game.

How we play tested Ameritocracy, and how you can test your card game

In this post, I’m going to be talking about how my partner-in-design and I play tested Ameritocracy.  Ameritocracy is a strategy card game where you are running for president.  You can read a bit more about it on our WIP thread at Board Game Geek (https://boardgamegeek.com/thread/1403479/ameritocracy). We’ve been play testing it for more than a year.

In the beginning, we followed the guidelines I previously wrote about (https://foodbedgames.com/2015/04/19/how-to-create-resource-costs-for-effects/).  Basically, we decided how long we wanted the game to go, how much power it took to win the game, how many resources a player generated each turn, and then did the math. Total time broke down into the expected number of turns we wanted, and then we costed stuff so that the games would end when we wanted.

From that base, we played the game, which sucked, but it gave us some ideas for making it fun. I think after the initial guess at costing, you want to focus on your core mechanic. If your game is a “dudes attack” game like Magic or Hearthstone, this is easier because you have a lot of examples. Our game isn’t, it’s kind of a worker placement game, so we had to do a lot of work understanding our core mechanic. When you have what you think is your core mechanic, I think you want to try to create a deck that is the simplest implementation of that mechanic. For example, if I am making a Magic-like game, I might make an aggressive deck with a bunch of creatures with no abilities. Just make up some cards and play that deck. See how it stacks up against your assumptions of how long the game should take, and whether there’s any fun.

I believe that the core mechanic of the game should be fun. It doesn’t have to be great, but there should be something interesting. In Magic, the decision to attack or hold back a creature, and combat in general, is the core mechanic. It’s not the most fun, but you can play a game with just creatures with no abilities and attacking and blocking. Simplifying the game lets you work out things like the resource system and the nuts and bolts of the mechanic. Once things work at the simplest level, add a bit more complexity. In the example of Magic, maybe add in a few creatures with larger costs and power/toughness.  See if the game is still fun.

At this point, you can start to add in mechanics to add more strategy to the game.  In Magic, maybe you add in some instants and sorceries.  After you do, maybe you realize the bluffing element of instants is something you like, you expand on that.  In Ameritocracy, we created what turned into the headlines cards.  We liked that type of player interaction, and so we keep iterating on it.  Hopefully, you will see things you like in your core mechanic and you can add in elements to expand it and make it more interesting.

From there, it’s about playing things a lot.  You can use data-driven approaches to track what are the best strategies, but ultimately, you will have to decide what you want to incentivize in your game.  In early Magic, either through design or misunderstanding how many cards a player would purchase (or something else), the best strategies often were not focused on creatures.  More recently, Magic has put creatures into the focus.  I think this is smart because it keeps the game grounded in the core mechanic.  It’s still possible to play a creature-less deck and win, but those decks aren’t dominant.  In Ameritocracy, we were worried about a deck with all 3s, the most powerful workers, winning too quickly and thus making it pointless to play other cards.  A lot of our balancing has been to make 3s good, but not overwhelmingly so.

It’s also been important to us to not get to a point of strategic collapse.  Strategic collapse happens when there is one (or a few) best strategy.  Usually some collapse happens, there just are naturally powerful combinations of cards.  So I think the objective is “controlled collapse”.  The goal for Ameritocracy was to have 4-5 decks that each played differently and either were similarly competitive and/or existed in a rock-paper-scissors type arrangement.  We are still working on this one.  It’s certainly possible right now to build multiple, similarly competitive decks, if one is focused on doing so, but we aren’t sure if it’s possible to do so if someone is focused on making the best deck.  In other words, we can create similarly powerful casual decks, but we don’t know if the game would stand up to tournament play.

Tempo: a helpful CCG concept

Competitive players of games like Magic have a concept they call tempo that deserves wider consideration.  Although definitions differ, I will define tempo as “the speed of winning the game.”  In Chess, people sometimes say “mate in two” meaning that they expect to achieve checkmate in two moves.  In this case, the number of moves is tempo, as the game progresses, the player who will checkmate in less moves has more tempo.

In many strategy games, having a faster tempo doesn’t automatically mean a player will win.  There is often an interplay between players using a fast tempo strategy attempting to win, and players using a slower, but ultimately more powerful, strategy, attempting to first disrupt the faster tempo players before overpowering them.  This dynamic is a fundamental dynamic in gaming, on the level of rock-paper-scissors.  In Mario Kart, there are small, weak characters with fast acceleration, and bigger characters with slower acceleration, but faster top speeds.  The bigger characters can bump the smaller characters to disrupt them.  In the board game Puerto Rico, one strategy to win is to quickly build enough buildings to fill up one’s city, while a slower, but ultimately higher scoring strategy, is to ship goods and end the game by depleting the victory point pool.  There are countless examples of this dynamic between fragile, fast acceleration, and consistent, high top speed.  It’s even a dynamic we observe in stories, the mouse is fragile, clever, and quick, the cat is powerful, but is out-maneuvered.

The emotional resonance of this dynamic makes it fun and engaging for players.  Players like to inhabit the role of the mouse while another player is the cat.  Tempo is the key to understanding this dynamic and to building it into your games.  To create this strategic diversity, you want the early actions in the game to be more efficient, but ultimately less rewarding than the later actions.  How you define early and late is up to you, but the basic principle remains the same.  To continue to use the Puerto Rico example, gold is quite valuable in the beginning of the game, but ultimately converts to less points as the game goes on.  So the fast tempo strategy spends their role choice on generating gold and building, while the slow tempo strategy tries to disrupt and build shipping and production.  As I wrote about in an earlier post on costing effects in games, it can be helpful to model this using an exponential function with some tweaking.  In my experience, the exponential function does a decent job at modeling higher cost effects, but the early effects tend to be weak. The remedy is to boost the early effects by a bit.

Noble Deceit

I’ve been working on a game tentatively titled Noble Deceit.  The game is a trick-taking game like Hearts meets strategy game like Magic: The Gathering.  It’s fast, easy to learn, and rewards strategy.  To win, you must become the monarch by currying favor with the nobles, however, the nobles are fickle and their allegiances shift.  In the game, this is represented by winning hands, if you win three hands, you win the game. Each player begins with seven noble cards, number one to seven, and three deceit cards, which are all unique.  Players take turns passing or playing any number of cards face down.  These cards can be either noble or deceit.  Once all players have passed, the cards are revealed.  Any duplicate nobles are then discarded.  Then each player adds up their remaining cards.  The player with the highest sum wins the hand, and all cards are discarded.  Players should keep their discarded cards face up in front of them so other players can see what they’ve played. Most of the time, when a card is played, it’s gone for good, so players must use their limited resources well if they want to rule the land. The main thing I’m trying to figure out now is how to determine the order of resolving the deceit cards.  Right now, there are icons on the cards, one for the reveal (arrow flipping up), one for discarding duplicates (two cards with X’s), and one for total (two cards with a plus sign).  These icons are meant to indicate the phase where the cards resolve, either when cards are revealed, duplicates discarded, or totals determined.  If two cards are resolved in the same phase, then resolve them in clockwise order, beginning with the card played by the player who went first in the round.  Resolve the deceit card for everyone, then move to the next card. Example: Alejandro went first this round, and plays “We’ll Meet Again”, a reveal phase card.  Emma plays “Bound by Honor”, another reveal phase card.  Since Alejandro went first, he returns a noble, then Emma’s card takes effect, meaning no other deceit cards are active.

Printable version of the game: (link removed as game has been updated substantially)

I-can Icon

Iconography is an important part of game design, and I wanted to make a brief pitch to game designers to dabble in graphic design.  While you engage with your game abstractly, most players will first encounter your game visually.  You are the best judge of whether players are understanding your vision, and consequently, I think it’s important for game designers to understand something of the visual design.  I don’t think it’s important for you to make things look really pretty, but it is important to be able to test ideas.  Should this icon be at the top or the bottom?  Should there be icons for this idea or should it be written out as text?  I think these are the types of questions you should be asking.  Then, when you likely work with an artist/graphic designer, you can “speak their language” more effectively, be able to show them prototypes and sketches.

For me, it was important to find a tool that was a good balance of easy to use and powerful.  I found Omnigraffle to be that tool for me.  It’s Mac OS only, and it’s mainly designed for making flowcharts, but I’ve found it to be a great tool for me to interact with my graphic design.  I can make quick little icons, layout cards, and so forth.  It has no “data merge” capability, the ability to generate a pile of cards from a spreadsheet, but that’s it’s main weakness.  I use InDesign for data merge, but find it difficult to use for other stuff.  Eventually I may move to InDesign, but I find Omnigraffle is just so quick and easy.  I’ve heard great things about Sketch and I imagine PowerPoint might be a good alternative too.

Is it drafty in here?

Card drafting is an interesting mechanic that has a lot of variation. It’s a strong enough mechanic to be the basis of an entire game like 7 Wonders.  Many deck building games such as Dominion, Ascension, Star Realms, etc. could even be thought of as variations of card drafting. There’s just this added step of playing the cards.  Worker placements games or role-choosing games, such as Agricola or Puerto Rico, can function like drafting games as well, although typically drafting occurs without replacement.  The grand-daddy of drafting games is probably the limited format of Magic: The Gathering.  Although many of the drafting methods described here were developed for drafting Magic cards, they may be adapted to other games.

Drafting is a great game mechanic because it’s fundamentally interactive, but allows for interesting limitations on interaction.  By using hidden information or differing incentives, it is often resistant to problems such as kingmaking and lame-ducks.  It may be possible that players are in kingmaking or lame-duck roles, but in many drafting games, players are unaware of their relative positions.  Even though they may be a kingmaker or a lame-duck, since they don’t know it, the game doesn’t have the same feel-bad consequences.

Although one could theoretically draft any game object, for example, Chess pieces, cards have a number of interesting properties that make them great for drafting.  First, because they have a front and back, they elegantly allow for hidden information.  Second, they can have text or symbols on them that convey game rules or information, and so the rule book can be smaller and you don’t have to define how game objects work ahead of time (e.g., how a Chess piece moves).  I mentioned before that there are games with draft-like elements that don’t use cards, like worker placement games such as Agricola.  In that game, players take turns placing their marker on an action, which allows them to take that action and block other players from taking it.  The next player then chooses from the remaining actions.  Although this functions a lot like a draft, I’m going to be focusing on card drafting.  Many mechanics here could be adapted for worker placement or other types of games, drafting is much more common in card games.

Please note that I’ve maintained and collected this information for a while before I even thought of posting it here.  Most of it comes from other sources, such as various Magic: The Gathering  and card game rules websites, Wikipedia, or people I’ve met.  I don’t remember where I found many of these ideas, but most of them are not original.

One last note: I try to edit and expand this list to make it as comprehensive as possible, and so I view this as a living document.  If you know of a drafting format, for cards, sports teams, or anything that you think think would be a good fit here, add it in the comments or message me.

Pack
Each player is given some number of individual “packs” of cards to draft. To begin, each player takes his or her first pack, selects a card, and then passes the pack to the next player. Players continue picking and passing the pack until it is depleted and then proceed to the next pack. The draft concludes when all packs have been drafted.

Solomon
Decide who will be “Player A” and who will be “Player B.” Player A drafts first, Player B may get some other advantage later, e.g., to play first in the game. A communal stack of cards is used to draft. To draft, Player A flips over some number of cards from the stack and then divides them into two piles. They can be divided evenly or disproportionately into two piles.

Once Player A has split the cards, Player B picks one of those piles and adds it to his or her drafted cards. The other pile goes to Player A. The process is then repeated, this time with Player B separating cards from the stack into two piles and Player A selecting a pile.

Solomon n
As a variant, more than two players can play, with each player splitting into n piles, where n is the number of players. Crucially, the player who splits always takes the last pile, which helps to prevent unfair splits.

Rochester
Reveal four cards. Player A picks one card first, player B picks two, and then player A picks the last one. Next, reverse the order with player B picking first. Keep drafting, alternating order until all cards have been selected.

Winston
Decide who will be “Player A” and who will be “Player B.” Player A drafts first, Player B may get some other advantage later, e.g., to play first in the game. A communal stack of cards is used to draft. To begin drafting, Player A sets the top three cards from the main pile in three stacks face down on the table, without looking at them.
Player A then looks at the card in the first small pile—without showing it to Player B—and chooses to either draft it or leave it. If Player A drafts the pile, it’s replaced with the top card of the main stack (also face down). If Player A leaves the card, the top card of the main stack is put on top of it, creating a two-card pile. Player A repeats the process with the second pile and then moves on to the third pile. If Player A didn’t choose any of the three piles, that player must take the top card from the main stack, no matter what it is. Once Player A drafts any pile or the top card of the main stack, it’s Player B’s turn to go through the whole thing again.

Remember: Each time a player takes a pile, it’s replaced by the top card of the main stack to form a new one-card pile. And each time a player puts a pile back, the top card of the main stack is added to it. There’s no limit to how large a pile can get!

Rotisserie Draft
All cards in the draft are laid out face up in front of the players. The players then roll dice to determine who will get to pick first, second, third, and so forth. The first drafter selects a card and places it face down in front of his or her seat. Picks continue until the last player has picked a card. The draft order then reverses: the last player selects a second card, then the second-to-last player, and so on. When it gets back to the first player, that player takes two cards, and so on. For example, if there were four players (A, B, C, and D) in the draft, the pick order would be A, B, C, D, then D, C, B, A, then A, B, C, D.

Winchester Draft
A combination of Winston and Rotisserie or Rochester Draft. Each player is given a random stack of cards. Determine a player A and B. Each player flips over two cards from his or her stack and creates two mini-piles. Player A selects one of these piles and adds it to his or her pool. Both players then add cards from their stack to each of their piles. Repeat this process for player B, and then alternate picks until all cards have been drafted.

Grid
Create some number of packs (commonly 16-20.) Each pack contains nine cards. Lay out one pack in a three-by-three grid. All the cards are placed face-up. Player A picks a row or column of three cards. Then, Player B picks a row or column from what is left. The remaining cards are set aside. Those cards are no longer part of the Draft. Repeat with all the remaining packs, switching the first-pick choice back and forth between players.

Glimpse/Burn
This is a variant of pack drafting that may work better for smaller numbers of players.  In this format, players begin with some number of packs.  Each pack has N cards in it.  In the draft, players will look at each card in a pack, choose one, and then eliminate (“burn”) a fixed number of cards from the pack.  For example, in a burn two draft, a player would select one card and then burn two.  Cards that were burned are removed face-down from the draft so that only the player who burned them knows what they were.  After a player has selected a card and burned others, they pass the pack to another player.  It may work best if the number of cards in a pack, N, is equal to the final number of cards you want a player to have per pack multiplied by the number that are burned.  For example, if you would like a player to draft five cards per pack and burn two, then each pack should have fifteen cards (each pick removes three cards from the pack).

Purchase
Players begin with some number of points or money. A row of cards is placed face up. Each card has a fixed cost to buy.

Optionally, after a player buys a card, a coin/money/points are placed on all remaining cards and then the empty slot is replaced with a new card. If a player drafts a card with coins on it, he or she adds them to his or her pool.

Communal cards
In any drafting variant where players alternate picks, communal cards can be added. These cards can be hidden or public information. Instead of picking their usual pick, a player may select one (or more) of the communal cards. Essentially, this just adds an extra pile to the draft.

Auctions
Similar to the purchase option, but without a fixed price.  Since there are many formats of auctions, I am working on a separate post that describes the various types and formats.