qI was one of the 1,600 laid off from EBay/PayPal (I was on the PayPal side), so I’m on the job trail again. My resume is in the standard place (HTML, PDF).

The kula model is interestingly incestuous.

Changelog:

• Added track markers
• Added boat marker
• Simplified delivery rules due to boat marker
• Added missing text that when delivering through another player’s kahuna the other player gets VPS
• Clarified 8th proa/prestige text to indicate that it may apply to both delivery rounds
• One extra prestige awarded for every two gifts given at an island
• 1 prestige point for each two kula given when a market is delivered

New rules for ‘Ohana Proa.

New Player Aid for ‘Ohana Proa.

Another playtest of ‘Ohana Proa last night using the knocked back rules (no reciprocal giving, shorter prestige track, single prestige for extra explore or extra proa etc). This time I got to sit out and watch them play rather than participating directly. Game-wise it worked well.

Notes:

• Several requests to add a boat marker so that players can move the boat along the routes during their deliveries and thus more easily visualise their positions and potentials. This seems a fine idea.
• Just like the last playtest, many comments to the effect of the game being mechanically simple and yet quite un-obvious.
• Many exclamations on how well balanced the game was. While one player fell far far behind in prestige, he was able to catch up through clever route building and bidding in the late game. Scores ended up with a ~25% spread from first to last (the eliminated).
• Single prestige discard for extra explore or extra proa worked well and was regularly used. Player who used it the most later cursed this as he ended one turn in the lead with 29 prestige and was then second when the game ended. I was delighted.
• Complaints that the board is too busy and difficult to visualise. Primary problem seemed to be that the route markers are round and easily visually confused with the round markets. The result was that there were many Oh dear that route is/isn’t claimed yet! during the game. Discussion suggested that making the route markers rectangular and visibly directional so that they clearly indicate the claimed routes and strongly visually suggest the connected network.
• Only one kahuna was placed during the entire game and it was central to that player’s success. Early kahuna are clearly worth more than late kahuna. The players were not convinced that kahuna were really useful, but most indicated that they’d look hard at putting down early kahuna if they played again.
• Remembering whether kula were fresh or stale when given was found a problem. The kula tiles are too small, easily flipped, easily lost track of etc. Some of the current problem is simply due to the tokens being too small. Cards may be a better option. -The game again ran long: ~200 minutes. These latest games have run far longer than previous playtests, despite the accelerated start, simpler end-game, faster economy and other game-shortening measures. I suspect this is due to the majority of prior playtests being conducted with experienced Age of Steam players who quickly grokked the delivery patterns of the game.
• The game nearly ran out of markets before players broke 30 prestige. This is concerning. Prestige production rates are slightly lower now that prestige is not rewarded for making new kula. There are a few obvious addresses: 1) Lower the end-game prestige bar, 2) Increase kula production rates, 3) Increase kula production opportunities. It feels like the kula pool needs to be sweetened by 15%-2o%. The problem with sweetening the pool is that it then also becomes more volatile. A little more volatility would be nice, but 20% would be far too much. I don’t want to drop the bar. I don’t want to re-introduce prestige for new gifts. Thematically prestige for stale gifts works better but is also richer than I want. So far the most tempting idea is to award an additional prestige for giving away more than N gifts at a single island where N is probably PlayerCount-1. Another idea is to multiply the points received by the recipient by the number of gifts given at the island. This would be a much more indirect approach but have faintly similar outcomes(?). Interesting…

Good stuff. Yep, gifts for gift-giving parties seem a fine idea. Now to run some models.

We did a semi-blind playtest on Monday. I was there to answer direct questions but otherwise intended to be silent. As happens we also lost a player at the last minute so I also participated in the game which was regrettable.

They taught themselves the game from the rules, pretty much just reading it aloud in somewhat backwards order. This took roughly an hour. I can easily teach the game in under 15 minutes, but I also know it well. I’m a little unsurprised at the length involved as none of the players were prepared; they simply sat down, picked up the rules and attempted to learn the game from scratch A repeated complaint was the large number of forward and backward references in the rules. I’m not sure what, if anything to do about that. More distressing was that they did not use the introduction section to gain an overview of the game and thus provide context for the rest of the rules to fit into. Conversely I was pleased that there were no questions left unanswered by the rules and that all questions they did have were answered by the rules as written and roughly about where they thought that data would be.

The game also developed unusually. All initially claimed routes were adjacent in the initial exploration with many shared islands. Kahuna and a gift were purchased on the first turn of the game (first time ever for that). The game ended explosively with all players earning more than 30 prestige points in the last round. Final prestige scores ranged from ~56-75, which is a little ridiculous.

A few of the more specific complaints:

• Calling exploration costs bids, while accurate, was confusing. They understood the costs as a bid towards turn order but felt that calling it a bid suggested an auction for the route explored. I’ve changed that language.
• They missed the entire Game Start section (setting initial turn order and initial route explores). They suggested I either fold that into Setup or provide a link to Game Start (the immediately next section from the Setup section.
• They missed the ability to trade VPs for resources ability. The text was there but they skipped over it for some reason. As a result they were confused over how anyone accumulated shells during the game. When I pointed this out the paragraph they’d missed there were no surprises as to where the text was or confusions over its contents.
• Requests for simpler language in the (long) Delivery section. Done.
• They noticed that the rules did not specify that delivery resources were taken from supply. Fixed.
• Wayfinder was called out as a clearly visible antecedent design. They’re right. Added.
• Kahuna were considered confusing and likely unnecessary.
• Reciprocal giving caused an overly exponentially explosive end-game. This was considered a big problem. I agree. Reverted.
• Playtime was over 3 hours. That’s far longer than any other recent playtest. I’m not clear on why.

On the reaction level the summary roughly summates to:

• Very unclear what to do, what to head for, what to attempt from reading the rules
• Mechanically simple, surprisingly mechanically simple
• Too long
• A (ver)y good game that still needs rough edges knocked off

Changelog for the new rules:

• Bids are now costs
• Corrected later/lowest language for turn order
• One prestige for 3rd explore
• Simplified delivery language
• Specified that delivery payments are from the supply
• Removed market colour game ending.
• One prestige for an additional proa
• Kahuna moved at cost on delivery
• Added credit for Wayfinder
• Moved Prestige multiplier boundaries
• Reciprocal giving is gone (it was exponential in the end-game). Old-style gift/points are back.

I also reduced the prestige costs for extra explores and proas so as to make those choices more viable and interesting.

• New rules for ‘Ohana Proa.

• New Player Aid for ‘Ohana Proa.

With Hippodice drawing near it is time to dust and prune about the edges.

• New rules for ‘Ohana Proa.

• New Player Aid for ‘Ohana Proa.

• New Rules for Muck & Brass

The changes aren’t large. I’ve shortened the end-game in “Ohana Proa a bit, hopefully lopping off 10 minutes or so, and allowed a pass action and end-game qualifier for Muck & Brass. I’m not convinced the latter is necessary but it is at least consistent with the rest of the pattern.

Gulf, Mobile & Ohio by Eddie Robin is a member of the Winsome Games 2008 Essen Collection and it is an odd game, a curiously odd game. Every player I’ve taught it to has exclaimed what a strange game it is!

On the face of it the game is relatively simple. Across player turns a variety of companies start and grow, worm-like from their start locations to connect to other cities and each other. There are 25 companies, each represented by two shares, a founder’s share and a secondary share. Initially only 8 companies are available, scattered about the edges of the board. During the course of the game additional companies from the rest of the 25 become available based on the activity of earlier companies. Victory points are earned by connecting companies to cities and to each other. Money is used only as a funding source for winning shares which give the right to build with a company.

Physically the companies are represented by coloured cubes placed in hexes. The restrictions on cube placement are simple:

1. A railway company must start with a cube in its home city
2. Company cubes are placed in hexes adjacent to one or more cubes of the same company
3. One one cube may ever be placed in an empty/clear hex
4. No cubes may be placed in the gray mountains
5. Any number of cubes may be placed in cities
6. No cube may be placed such that its ownership cannot be traced unambiguously back to a single railroad company.

Given that the colour of cube used by a given company is dictated by which cube colour is most plentifully available at the time the company was founded, that last placement rule is a doozie. In short it means that every worm-pattern of cubes of a given colour for a given railway company is surrounded by a one-hex penumbra of hexes not containing that colour – and the player does not get to chose what colour that is (and thus what other colours it can connect to.

Structurally the game is a process of iteratively managing timing, opportunity and positional advantage. This is not a game of building up companies, establishing an economic powerhouse, carefully assembling synergistic systems or running faster. This is a game of ensuring that other player’s choices are minimally profitable for them and that your choices sum to be (a little) better timed and a little more profitable than their’s. The game is a minuet dance, a delicate series of fencing moves, parries, ripostes, lunges and recoveries, each one a tactical dance move carefully gauged to give little ground to others while grabbing every advantage possible. Advantages are often measured in single dollar differences.

While the shares are made available via player-selected auctions, winning a share has little long term implication. What is being auctioned is the opportunity to build for a given railway company. Building is done both to gain victory points¹ (building is the only source of victory points) and to minimise or constrain the opportunity afforded to later players by that build. The bid money is spent directly on building track for the company with any unwanted or unspendable excess discarded. As the mesh of railway lines extend to new cities, railway companies that start in those cities become available for auction. Thus each build can also extend the set of companies available for auction by later players and so builds are carefully gauged, often with extra cubes unprofitably placed simply to prevent or discourage victory-point-generating connectivity by other colours. Sometimes shares will be won and no track built at all (all the bid money simply discarded) as building would create too much victory point opportunity for later players, either by exposing the colour for easier connection by other companies and players, or by making more companies available in the game which offer too much positional opportunity to other players. Measuring opportunity cost, both for yourself and others is a constant challenge.

In short, often, usually, later players are able to get more victory points for less cash than earlier players as they can build to connect to the colours already present on the board. However someone has to go first and if you never go first you’ll never win either. Remember that bit about how company colours are determined? Carefully tracking, controlling and predicting what colours will be available when and to whom, and thus what companies could build to connect where and to how many other colours is the logical centre of the game. It all depends on what shares are auctioned when and by which player in turn order considering the balance of cash holdings across the set of players. This is the core of the dance: watching cash holdings, opportunities, turn order, and income and selecting shares to auction and bidding and forcing others to bid exquisitely close to the line. Single dollar differences can make a huge difference – even if those extra dollars in a bid are discarded as opportunity cost in order to not build track or are used for blocking builds to prevent other future connections.

Ultimately shares also pay dividends at a fixed rate of either $5 or $3 per share. Shares are the income source the players use to fund future auction bids. Establishing a good income source is unsurprisingly important but is also not as critical as maintaining a tight control of opportunity and timing. A player low on cash often has better potential advantage to manage opportunity than a richer player who must balance their choices across more players and a wider bid opportunity variance. Thus a low cash player can re-establish themselves by very careful management of what is auctioned and what is built when and where as the richer player waits to swoop in with their cash behemouth. They can effectively force the richer player into ungraceful diseconomies simply because the rich player must spend their money or lose their positional advantage. This is not easy and is near impossible if the gap grows large, but it can be done and it is a wonderous thing to behold when done well.

Passing, a player passing on their entire turn, is a common action choice as part of the dance of opportunity control. Much like in King of Siam, passing is a way forwarding the onus for an unwanted decision to a player who cannot afford not to make a decision². Of course they will attempt to make the decision that offers the east opportunity to the other players while also preserving their own victory opportunity, but careful play will limit their available choices while moving turn order forward to something more attractive to you. In this way the dance is lurchingly moved forward to the next unwilling player.

It is hard to describe Gulf, Mobile & Ohio as a strategic game. It is also hard to describe it as a tactical game. The decisions made each turn seem entirely tactical, but rest on an analysis that should extend out 2, 3 or even 4 turns into the future as the opportunity implications are assessed and weighed. This process seems tactical as that analysis needs to be performed on every turn given the current play state but also seems strategic as the look-ahead is fairly deep and there are core patterns in the game which can be built and leveraged.

Expect your early games to be filled with runaway winners. It can take a while to comprehend how to use the tools that the game provides. I’ve played about half a dozen games now, all with either 3 or 4 players³ and do not yet feel I’ve a good grasp of the game’s depths. I’ve merely seen a few small patterns and the hints of many more in the wings⁴.

The picture below shows the end of a 4 player teaching game. The player closest to the camera (green) won in the last two turns of the game despite being behind in points but marginally ahead in cash and income for the entire game.

Perhaps oddly I find 4 player games far easier to teach than 3 player games. The edges are a bit softer and the timing controls are a little less fragilely unstable with 4 players than 3. There is enough to digest and dance with here that the extra ease granted by a 4 player game is welcome learning space. However for those same reasons I find the game noticeably improved with 3 players. That’s where all the safeties and guard rails are off and every decision is filled with knife edges.

This weekend we visited the Northern California Renaissance Faire:

Picture gallery

Having fought through much of the three day weekend with the Neuland-style action track for Muck & Brass, I’m calling it quits. It is not a good fit for the design. The mechanism has a number of interesting properties but they will be better used elsewhere.

I am haltingly, gingerly, considering the idea of setting up a small vanity press for my games. The initial target would be my Age of Steam expansions. Once past that only the gods know if it would fitfully stagger onward or die on the vine. But what to call such an ad-hoc publishing company? Suggestions for names are welcome.

In the prior discussion I defaulted without examination to moving forward in the event of a collision¹. Why?

If collisions do not trigger slides then either it must always be possible to avoid a collision (easily proven impossible in 4+ player games) or there must be an ordering function for players resident at the same action slot, perhaps a FIFO or LIFO. Both queue forms have their interesting aspects with LIFOs perhaps being the more interesting as later players can get an instant second turn by ending atop a prior player, perhaps to the earlier player’s discomfort (making turn prediction more difficult and yet more valuable).

Stacking seems inevitably less interesting. Collision sliding offers a few clear interest-benefits:

1. A slide moves the subject player further around the ring thus making their next dividend closer/sooner
   • Situations can be crafted which allow multiple positions to be collision-slid
2. Collision sliding encourages players to chose actions which use/absorb the lost action point. Because this value cuts in both directions, Development gets an extra boost encouraging a slightly more Development-centric game than Expansion-centric
3. Unpreventable slides (can) force players to waste action points in non-income producing movement

Meanwhile almost the same multiple-turn opportunity remains, just trimmed by one action point to not include the collision point.

More simply collision-sliding offers an additional decision: To slide or not to slide. Each side of that decision has significant effects on turn order – effects which are interesting for the subject player and interesting for other players to predict and arrange afore-hand.