Collusion: elections

I mentioned two minorly worrisome things in the last post, and the second of those is voting for baronial control.  

Now I should be worried about this in general because of the potential for a kingmaker effect.  Say that it's the final baronial election of the game, players B and D are candidates, and the barony is worth 5 VP.  To whom are you, player A, going to gift with your votes, and 5 VP?  In a game with 1 VP victory margins this may seem like a swingy decision.

I'm not actually worried about the elections themselves, for a few reasons.  First, because players have the ability to cut deals.  If your votes will give someone the barony, you can try to get something in return, and there are many kinds of deals to strike.  Second, because at least part of your scoring potential must remain secret and so no one exactly knows how well everyone else is doing unless they're paying really close attention and making several good inferences.  Most players will find the outcome of this decision uncertain.  Third, because you know from the beginning of the game the order in which baronial actions are going to take place for the whole game, and have ample opportunity to shape the board state in such a way that that final election doesn't "swing" the game at all.  You play a role in deciding, not just who gets the points, but how many points the barony is worth in the first place.

My concern is actually with the configurational aspect of the elections.  Each player gets votes from the factions they influence that are present in the barony.  Players cast votes in clockwise order, and the candidate who receives the most votes gets the barony.  If there's a tie, the player who received the most votes earlier gets the barony.  This introduces a turn order effect that has a weird consequence.  Let's say that B and D are candidates, and voters have the following votes to cast:  A-3, B-0, C-2, D-1, E-0.  A will vote first.  

Now this means, essentially, that only A's votes matter.  If A votes for B, B will have gotten to 3 votes first, and D can't exceed 3 votes, so it's a done deal.  If A votes for D, again, it's a done deal.  This means that C's votes don't really matter.

My only thought for a change is that if there's a tie, instead of resolving in order, resolve based on the votes of the players who are not candidates.  But that doesn't change things in this case.

I think it just needs more study at this point.  Getting influence is supposed to mean having power in elections, either to vote yourself in or sell your votes to an opponent, but if it's often the case that only one person's votes matter, then that's less fun for the other players.

On the other hand, getting lots of votes consistently usually means that you've gained influence over a bunch of factions, or a bunch of buildings for factions you influence have been built.  Thus if other players don't want you to get more votes than they, they should stop supporting your influence actions.  

This is one of the fun subtleties I've been discovering about the game, actually.  Compared to board-altering actions like "switch barony" or "build a faction", which interact directly with schemes, the influence action seems indirect and harmless by comparison, and sometimes you'll support an influence action for that reason alone.  But do that often enough and one player can become very powerful indeed come election time, and can sway every election.

It's interesting to watch what that power over elections looks like in practice.  During setup, each player is given one barony.  Although the baronies grow and shrink, usually maintaining parity seems like a wise course: giving someone a second barony seems to make them very powerful, unless you're somehow sure that one of their valuable goals is definitely going to go bust.  Thus the elections aren't always about big swings of power; rather, they're often about, how much are you going to pay me so that I keep you in power and maintain parity?  It's as much about what you can get out of the current baron as anything else.

This is why one of my other concerns about elections may be allayed.  Necessarily, an election needs two candidates to be interesting, but additional candidates only come through successful use of the estate action.  I've biased that action in a few ways to make estates easier to add.  And the aforementioned considerations make it advantageous to support other players' estate actions: if you can get a second candidate in a barony you don't care about, that just means that one more person to potentially bribe you for your votes.  But even if there's only one candidate, there's still an incentive for that person to cut some deals, to help secure his/her position in later elections.  For example, maybe I'll take one of your heirs as a courtier in the baronial court, because if you later depose me, that heir leaves the board, costing us both 1 VP.  Thus, my accepting your heir early biases you slightly toward wanting to support me later.  And this is important because the ability for territories to change ownership from one barony to another sometimes means that candidates pop up 'unexpectedly' in different baronies as the game unfolds.

I think there are all sorts of little considerations and heuristics like this that come into play in the game, but because it's so interactive the ones you rely on should change every game.  At least I hope that's the case.

Collusion: three or five worrisome things

I've been solo-testing Collusion furiously lately, and while it's not easy to play multiple seats in a negotiation-heavy game, I feel like I'm able to mostly make reasonable assumptions about what deals would be offered and accepted.  But in a live test a few weeks back, I was interested to find that the deals players were making were much more elaborate than those that my imaginary players were making.  On the one hand, this was cool, because one wants a negotiation game to give some scope for player creativity.  On the other hand, that group only covered 1/3 of a game in 1.5 hours of play.  I don't think this game has the legs to be a 4 hour game.  The problem, it seems, was that players were haggling to death points that probably weren't worth the level of mental energy they invested. 

Part of the problem, then, is on-boarding.  We need a simple way to encourage players not to go too crazy on the negotiation until they've gotten their feet wet.  I think one solution for this is simply to require that, in one's first playing, every deal must involve the transfer of some physical commodity; no promises extending into the future.  And, to end those endless back-and-forth negotiations, we can introduce a "wrap it up" timer, which another player can activate to instruct the two involved players to finish their negotiations within (30 seconds, 1 minute, whatever).

Game length, then, is one of my three concerns. 

A second is the way that the game state can change so much, that progress doesn't always feel unidirectional.  All that matters for scoring is the final board state, but if in one turn I add, say, a red building to territory 15, well, in some future turn you can just propose to add a blue building there and if it gets enough support, poof, the red building is gone.  So, what was the point of the red building having been built in the first place?  It is true that there are short-term impacts that such actions make.  For example, that red building will give votes to whoever influences the red faction, so while it was standing, influencing it might have enabled me to get some leverage over my opponents and extract some concessions.  But it does seem like late actions matter much more in the final reckoning.  

One solution to this might be to just reduce the total number of actions in the game.   There are at most 45 actions that will happen during the game, and there are 25 territories, which means that on average, two things will happen in each territory during the game, and since there are four kinds of things that can happen (change barony, build a faction building, influence a faction, build an estate), maybe the amount of flux per territory isn't so bad.  Another solution might be to just watch and see whether players can use the tools available to them in the game to mitigate this.  For example, if I don't want anyone to disturb that red building I built, I can put a different action into that territory -- perhaps "influence" or "estate" -- thus protecting the territory from disruption.  But it means I have to get to that territory before anyone else.  It may be a little weird to have these sorts of configurational matters influence the game so heavily.

But that leads to the third concern, which is the turn structure.  There's a row of barony cards, and on your turn you skip ahead to a different barony and place an action in one of that barony's territories, then provide support to other players' actions.  This creates a seat order effect.  If you're an early player, you'll have more freedom of choice among the territories within that barony; if you're a late player, you'll have more control over which actions actually come off, since you provide support later.  This is asymmetric but I don't know if it's imbalanced.

There's also the effect whereby players can skip ahead to get into an important barony at an earlier time, and I think this is important and advantageous, but if it turns out that it's always strictly better to propose an action in every single barony, as opposed to skipping ahead, then there's a seat order problem that needs to be balanced in some way.  Plenty of ways to do this but realistically it would call into question whether this turn mechanic is the right one to use.

There are two other littler things that I'm not yet worried about, but I'm worried that maybe I should be worried about them, which itself worries me.

The first of these stems from something that happened in our live test last week.  Based on the deals a player was trying to strike, the other players were able to surmise what one of his schemes was, and more importantly, that it was fairly important to him.  The intent of the game, of course, is that when you figure out what someone else wants, to try to make deals that bring that about but that reward you richly for your assistance.  Instead, another player took the approach that "this goal gets him points.  Thus we should band together and deny him those points, and not approve any action that would let that goal be fulfilled". 

If the players take that view collectively -- that it's a game mostly of obstruction -- the game may have a low energy state and not much will happen.  Now I think this is permissible -- the game is called 'collusion', it isn't specified how you're supposed to collude -- but at the same time I want this kind of approach to feel like an uphill climb.  The game should trend toward action, and the focus should be on steering and shaping the actions that happen, not trying to prevent actions from happening at all.  And, I think the game does do this; players have the ability to make offers and deals, they have natural alignment with each others' goals, and the rules require that some actions have to happen each year.  Players hopefully would find, then, that it's better to help and get paid than to deny and obstruct, because if we make a pact to obstruct but someone then dissents from that pact, that someone will be the one to get paid for the help you could have provided.  But it's a slight concern that needs more observation.

The second little thing deserves its own post.  

Collusion in a nutshell

The last three posts may not make a ton of sense since I haven't yet explained how the game works.  Here is a quick synopsis.

Collusion is played on a board containing 25 territories.  These are clustered into 'baronies', and wooden pieces belonging to 'factions' are added to those territories.  Additionally, players place discs into territories as 'estates', or associate them with the factions to 'influence' those factions.

During setup, each player receives four 'schemes', representing the end game state you're trying to achieve -- things like "X barony is largest", "Y faction controls 2 cities".  You arrange your schemes in a row, and they stay in that order.  At game's end, the leftmost scheme is worth the most points, if you meet its condition, and the cards to the right are each worth a little less.  

You also get points by controlling baronies, and the bigger a barony you control, the more points you get.  You also have three heirs that are worth negative points unless you get them out onto the board.  And finally, those power discs you have count as points for other players if you give them away.  

The turn mechanic is very simple.  There's a row of five 'barony cards'.  You advance your marker to the right on this row, to a different card from the one you're on.  Then, you propose an action within that barony, placing an action tile into one of its territories.  Finally, you place three support tokens onto action tiles previously proposed by other players.

When everyone has had a change to propose and support within a barony, that barony is evaluated, and the three actions within that barony that received the most support are the ones that happen.  Then there's an election.  Anyone with an estate in the barony is a candidate, and anyone who influences a faction gets votes, if that faction is present in the barony.  The person with the most votes becomes baron.

There are four types of actions and they're all quite simple: expand/shrink the barony, add a building for a faction, influence a faction, or place an estate in the territory.

But of course, because your actions only happen if they gain enough support, the trick is to propose actions that you think your opponents will support.  There's a clever way of dealing out the schemes such that, in a 5p game, each of your four schemes will "rhyme" with a scheme held by exactly one opponent.  Thus you're aligned with every other player in exactly one way.  But of course, you can also try to bribe them for their support, and that's where giving out power discs and trading heirs come into play.

There's a bit more to it but that's the basic gist.  I think it's unique in that everything that scores you points -- achieving your schemes, securing baronial control, getting heirs onto the board, acquiring power discs -- involves your opponents.  Thus it's no exaggeration to say that at the end of the game, the person who wins will have been gifted the victory by the collective action of everyone else.  

Collusion: a bolt of lightning strikes

In 2017, my final attempt at making the cost matrix approach to the game fell flat with three of my most trustworthy playtesters.  It just wasn't going to work, and I had to try something else.

I reduced the number of factions from 9 to 6, kept the idea that each faction has an action type, but did away with the idea that each faction's actions are limited each year.  And, I kept the concept of loyalty and some of the ideas about cost discounts and such.

But I made two really sweeping changes that I think were wise.  First, I had just gotten a copy of In the Year of the Dragon and Broom Service, and was thinking about alea games and how they all get great  depth out of simple turn mechanics.  I needed a similarly simple turn system, but I wanted something original as well.  I settled on an idea inspired by the old 'king's tour' concepts: each year, the king was going to visit several of the baronies all in succession.  In each barony, each player would propose one of the six available actions in one of that barony's territories.  And each action had a 'power track', which you boosted with your intrinsic power based on the factions you influence, the alliances they're in (loyal/disloyal), and with some of those power discs.

Second, in the previous version I had ordered little aspirin pieces to use for tracking your influence in each of the 9 factions, and now I had no further use for them.  I decided that they could be used to provide 'support' to the actions that were proposed, advancing them further on the 'power track'.  And, whichever actions exceeded the (faction-specific) power threshold would happen.  And then we'd move on to the next barony and do the same thing.  But you only had a fixed amount of support to spend for the year, so, use it wisely.

We played -- yes, you guessed it -- half of a game of this, and once again it didn't work.  Except, parts of it really worked.  The support idea was great: it led to all sorts of interaction and deal-making and horse-trading.  There was finally something that looked like actual collusion between the players:  we're proposing actions but need to work together to make those actions actually come off.  I'll support your action if you support mine.  It was fun, even if the deals we were striking weren't especially creative or interesting.

The parts that didn't work were the same things that always didn't work.  The connection between factions as entities on the board and factions as the source of actions was confusing.  The idea of intrinsic power in a faction was confusing.  The tracking of alliances and loyalty was confusing.  And some of the actions, which manipulated these things, still seemed disconnected from the actual goals that you as a player were trying to advance.

Thus, in a final act of merciless whittling, I decided to cut the number of actions down to 4, and to have dedicated tiles for each action type that were disconnected from the factions, also cutting out the idea that influence over a faction gave you some kind of discount in that faction's action type.  

So, to propose an action, you just place its tile into a territory, and then there's a track that shows who proposed what action and how close that action is to success.  But I also decided that instead of a threshold, it could just be that the actions that received the most support were the ones to happen: no intrinsic power, no loyalty, just the support of your opponents would decide whether your action comes off. 

When I did that, I realized that we could do away with the power board entirely, and put the action tiles onto the map board and the support discs right onto those tiles.  Now it would be easy to see who proposed what action, and which actions are going to happen in the present configuration of things.  This had led to an enormous improvement in the game's readability, which in hindsight was always the game's biggest problem.  It's not always easy to keep track of all of the ramifications of all of the proposed actions -- it's still a highly intricate game -- but at least seeing what can happen and how close it is to happening is now much more accessible.

Those power discs are still in the game, but they're used a bit differently.  They can provide support to your own actions, or can be given to your opponents as a bribe.  Discs you hold belonging to other players are worth 1 VP at game's end, and this seemed to make good sense:  a disc is worthless to you at game's end, so you want to use it and get something out of it, but if you give it to an opponent, you'd better get the equivalent of 1 VP of value out of it.  Yet, these calculations are by no means easy -- there's little that can be reduced to a strict VP valuation because so much depends on which actions get the most support -- so you have to strike deals judiciously.

The result is something that's many miles away from where the game started or where I had any intention of taking it, which is ironic considering that it was called 'collusion' from the jump.  It's a game where you can't power your way through anything; literally everything you want to accomplish in the game depends on your opponents.  It's not a trading game, though: we're not exchanging commodities, we're exchanging support, and whether I support you depends not merely on what you offer me but also whether what you're asking me to support is something I'm inclined to like myself, based on my own goals.  And this is also why it's not a typical alliance game: in this game, there aren't stable alliances that persist for big chunks of the game (Diplomacy, Struggle of Empires), instead there are alignments.  So, in this barony, you and I may have compatible ideas about what we want to happen, whereas over in this other barony, our intentions are diametrically opposed.  You'll be aligned and misaligned with all of your opponents in all sorts of different ways, and skillful play isn't about browbeating them into doing what you want them to, so much as it is incorporating them into what you want to do, and finding ways to be useful in the things they want to do.

I'll give two simple examples.  First, one of your goals says "Barony X is the largest barony".  The obvious thing to do would be to seek to become the baron of barony X, and expand X.  In that way, you get points both for the goal card and for the value of the barony itself.  Except that, if you instead install one of your opponents as the baron of X, then that player is helping you achieve your goal, because it's in your mutual interest for the barony to grow.  It's fewer points but has a greater likelihood of success.

Or say that Barony Y has estates belonging to purple and white.  If you, as player yellow, can influence one of the factions that's present in Y, or build buildings in Y for a faction that you already influence, then you gain votes in Y, and can use this to extract concessions from purple and/or white.  You have made yourself indispensable in their rivalry, and this is usually much more profitable than if you tried to install an estate and join in the rivalry yourself.

These three posts, taken together, hopefully show what Collusion's journey to its present form has looked like.  In future posts I'll comment on further design and playtesting progress, and some of the issues that pop up along the way.  The game is still a little fiddly, and there may be some balance tweaks needed, so there may yet be room for improvement, but I think it's coming along really nicely and I think, much to my delight, that it's turning out to be something that's completely different from any other game that's currently out there.  It's fairly simple but it's so interactive, and the ramifications of those interactions are so far-reaching, that I think it will have a lot of appeal, both at the level of a fun deal-making game and also at the deeper level of a knife fight in a phone booth where everything is hotly contested and haggled over.  Games seem to build to an exciting climax every time, but players definitely need a brain break after tallying up the final scores!

Collusion: glimmers of collusion

Collusion spent most of its life on the design backburner, by which I mean that I'd take it out, work on it, find that what I tried didn't work, and put it back onto the backburner for a couple of years until my next great idea.

In reasonably short order from the early stages documented in the last post, game-development-wise (not calendar-wise), some core design ideas emerged.  The game needed a proper scoring system, and it was mostly built around the core concepts that animate the current scoring system:  players are given some "schemes" representing game states they want to achieve when the game ends, and then you get points for controlling baronies and for forming marriage alliances.  

In these early stages, most of the actions snapped into place as well.  There were 9 factions, and each had an action you could take: grow a barony, add a faction to the board, move the king, form a marriage alliance, and so on.  Some of these actions persist into the final version.

What took many years to figure out was a sensible way to put interesting obstacles in the player's path.  For too long, that obstacle was cost.  It was supposed to be hard to do what you wanted, because actions have a cost, and you have to be able to pay that cost.  In fairness this is how almost every other game works so it's an understandable mistake.  

At first this cost was explicit.  There were resources, and you had to pay resources to use a faction's power.  The more it was used, the more expensive it became, and each faction had a preferred resource.  Except! As players constructed alliances between the factions, you could pay the resource that faction A accepts to use the ability of faction B, but you have to pay extra based on the number of 'hops' required to get from A to B.  There were a few ways to do this, but the idea was largely about having a sort of power network that you could hop into and out of en route to paying for actions.  Additionally I introduced an idea of 'loyalty'; factions were either loyal or disloyal to the crown.  Early on this was helpful because the idea of factions being in alliances quickly merged with the idea of factions being loyal or disloyal, such that there was a loyal faction alliance and a disloyal faction alliance.  

Thus, one of the things you could do was manipulate the alliance landscape, to switch a faction from one alliance to the other.  This had some benefits to you, since you yourself were loyal or disloyal, but mostly it was a ways-and-means thing, it let you use factions less expensively if you could get the alliance landscape right.  And then in addition, you could influence factions yourself, and using a faction you influenced, or one of its allies, also gave you a discount.

A new wave of playtesters bravely braved this wave of changes, and once again we played a few games that didn't get halfway to completion before the players gave up.  They were good sports, but again, it was just too indirect to go between resources I have access to to actions I want to take to schemes I want to complete.  

The solution to this, I decided, was not to streamline the action and cost system, but rather to try to find ways to make visualization of the costs easier.  Of course this led to some streamlining for the nonce, and the first to fall were the resources themselves.  Instead, paying for an action required 'power' (ooh!), with all the previous discount concepts being ported over to the power system.  

But we also added a system that I truly liked and that, though it's since been cut, may find use in another game someday.  The idea was this: you have a rondel with power discs on it, with each rondel space representing a season.  When you pay 'power' from the current season, your power boost is given by the number of discs in the stack, and then you discard the topmost disc on the stack.  Any unused discs are added to the next season's stack.

Now as you can readily see, this does two nice things.  First, it means that the longer you wait to act -- i.e. the more seasons you let elapse -- the more impactful your discs become.  If you have, say, three discs total, and you use one action per season, you'll take three actions of power 1, but if you wait to take those actions till season three, you take three actions of power 3, 2, and 1, a big advantage.  (But, there's also a limited number of actions of each type available, so waiting has consequences).  Second, it means that power discs now become a tradable currency, and if you give me a disc for my use, it goes to the bottom of my stack, and comes back to you when it floats back to the top.  Thus there's a positive-sum interaction in making deals -- colluding, I daresay -- but also time ramifications.

Now this system was actually pretty good.  The trouble was all of the other stuff, the nine actions, the discounts, and so on.  The idea was that a faction's use cost depended on the target of its action, but factions targeted different things, and so each faction had a different cost structure.  For example, a faction that lets you build a building has its cost determined by the presence of the faction whose building you're going to build, whereas a faction that is going to grow a barony has its cost determined by the size of the barony it will expand.  Now this makes logical sense:  a strong faction should cost more to build, a big barony should cost more to expand.  But in practice, it meant that there's a pretty big matrix of costs to consider (what faction will I use, and what is its target?) and then the discount rules manipulate those costs, and it's just a lot to parse.

Undeterred, I kept charging ahead with different board layouts to permit easier visualization of the cost landscape, and when even these failed, I did away with the subjective action costs and just gave each faction its own cost, irrespective of its target.  Even this didn't help, because trying to visualize the discount structure was still a huge pain.  It depended on alliances and various things, but every time something changed, your entire discount matrix changed with it.  A real mess, and more importantly, it just wasn't fun.  Players couldn't connect their actions to their goals because they were so preoccupied with trying to connect their discounts to their actions.  And too many actions revolved around manipulating those costs, too few around manipulating the goals.

Collusion: origin story

I guess I'll start using this blog to document ongoing progress with my latest "big" game, Collusion, which I think/hope is nearing completion.  I've thought/hoped that about other projects in the past and had those thoughts/hopes dashed, so it's possible that will happen here as well!

I mentioned a year or so ago that Collusion had just turned 10, which means that now it's about 11, but a lot has happened in this last year.  It's almost night-and-day different in many respects.  I'll spend a couple of posts talking about those first 10 years -- its childhood, I guess you could say -- and then in a separate post I'll say a bit more about this adolescent phase it has just gone through.

As I have said previously, Collusion was my attempt at a really heavy game, which in hindsight I realize to have been silly and misguided.  At the point that I started Collusion, Sands of Time had not yet reached its apex of complexity, and so I was not quite aware that I was already working on a very complex game.  Moreover, really complex games were still something of a rare breed at that point, but 11 years later, the complexity creep in the hobby has given way to a complexity flood, and games are incredibly complex, cluttered with ideas and systems and icons and components and all sorts of things.  I don't think this has been an entirely good thing and have disavowed complexity as an animating design philosophy in my own projects, which has had interesting ramifications for Collusion as we'll discuss.  But let's look at how it got started.

At around this time, I was an active participant at the Board Game Designer's Forum, having spent most of my formative years as a designer hashing out ideas about design in that community.  One of the designers there was Rene Wiersma, designer of Gheos, and he was talking at that time about a new design of his, his "magnum opus", which was going to pull together all of the things he had learned about design into a single game.  (I don't know for sure whether he ever finished the project though!)  This got me to thinking about embarking on a similar project, or at least a project of a similar scope, and I started with, more or less, two main ideas.  First, the title, "Collusion", and second, that it would have area control but not area majority.

The premise was that we were nobles in some fictitious realm.  Fictitious realms in games always seem to have kings on death's door, but in this game's case, we weren't trying to succeed the king or even impress him; rather, we were trying to use him as a pawn in the great game we were all playing for the supremacy in our behind-the-scenes scheming.  The kingdom had a bunch of territories, each with a characteristic resource that it produced.  The game had seasons, and each type of resource only produced in one or two of the three (?) seasons.  The realm had roads, which were only passable in some of the seasons.  It had estates, which we could claim.  And it had baronies, groups of territories that could grow and shrink. 

Most of all it had factions.  Factions each had a characteristic building, which could be placed into certain special spaces on the board.  And they had cards representing their members.  This was important because we each had a mat, which specified our heirs and with spaces for advisors.  Thus we could claim cards from certain factions and marry them to our heirs (actually, this was a two-step process, betrothal and then marriage) and we could hire them as advisors, by placing the cards onto the corresponding space on the mat. 

But each card also had a special action it could perform.  Thus the turn mechanic was actually quite simple: take a card, and either do its action or attach it to your mat as a betrothal or advisor.  Each faction's actions were associated with the kinds of things that you'd think thematically that faction could do.  The church performed marriages, the merchants converted resources, and so on. 

Each year, the king took a royal tour around the realm, following the paths that were open in whatever season it happened to be.  You wanted to steer the king to your estates so you could entertain him, by throwing a festival in which you expended resources.  Each season had a set of appropriate festival types -- i.e. a resource-to-points conversion -- and we tracked the quality of the festivals we each threw, with the person who threw the best festival getting some reward at year's end.  And then there were also baronial elections, whereby you gained control of the baronies. 

You'll notice I haven't said anything about collusion, because at that stage there wasn't any.  It would have been more accurate to call the game "Scheming"; I toyed with a name switch to "Levers of Power", but always stuck with Collusion, and now it's a more fitting title for the game.

We played the game once.  Or rather, I taught the game over the course of an hour to two patient souls (one of them was my brother) who attempted to discern just what in the world I was talking about.  Then they heroically attempted to play the thing, we may have made it through a year, and then ended in confusion.  Ending the game early is going to be a common theme in this.  And for comparison, the latest version, I can teach in about 10 minutes. 

I've talked about how in the process of designing Sands I came to realize that a game should have a core idea, but that realization came late in Sands' development and I certainly hadn't made it by the early years of Collusion, although I now know what the idea of Collusion is.  At the early stage, it turns out it was just a kind of generic and not especially focused power-acquisition game, with the intent that there were a lot of different things you had to think about.  I guess that my intent, though I never fully realized it, was that there was a 'levers of power' construct, in that there were different systems in the game, and you couldn't interact with all of them, so the point was to choose how to compartmentalize your participation into just those systems that you were going to emphasize in your strategy.  It wasn't that the game's turn mechanic was complicated, but just that there were so many different layers to what went on during the game. 

As I said, the game was hardly even playable early on, much less memorable.  I think in its current form it may be something special.  I'll talk in the next post about the intervening years and how it made this journey.