Removing variants

At Manorcon, back in July, I was talking to Ian D. Wilson, the designer of 1861, 1812 and 1858.  He advised me to reduce the number of optional variants, on the grounds that many 18xx players would spend time discussing which variants to play rather than getting on with the game.  In addition, the extra components add the possibility of confusion, as we have already seen in the game that we're playing.  18GB already has a set up that varies with the number of players, so adding extra complexity at this stage is probably a bad idea.

On reflection, I think this is good advice.  To an extent,  variants such as the extra tiles and the 8X trains simply reflect different design options that I explored during the development of the game.  I kept them partly to hedge my bets, as well as to allow some extra variety should players wish it.

I have already taken some steps in this regard.  In version 26, I have simplified the main choice into a standard game or an advanced game.  The advanced game adds the character cards and the rule for World War One.  These are reasonably well tested and proven, bar some minor tweaking to the fine details of the characters' powers.  The character cards are clearly separate from the other components of the game and cannot be easily confused with them.  All other variants are put separately at the back of the "Setup and Scenarios" booklet.

I now intend to take this further.  I will remove the extra tiles variant and the 8X train variant altogether.  These will remove the extra components which are easily confused with the standard components. I think this decision will simplify the game setup.  And I think the game plays better without these rules - otherwise I would have included them in the core design.

Catching a runaway leader

I hope the design of 18GB avoids the effect in which one player gets an early lead which nobody else can catch. I am fairly confident that if a company starts strong, it can face sufficient challenges in the middle and late games that other companies may be better investments.  The need to raise cash for new trains, the potential of routes being blocked, and the possibility of share price falling from hostile sales (particularly before the company converts) can all help to pull a leading company back, while the multiple jumps on the stock market may make other companies more attractive investments.

I am a little anxious that the same may not be true if a player manages to stake out an early lead.  With good play, someone who has earned more from revenues and the stock exchange will often be able to shift their investment into new companies.  If the early lead is too strong, I’m worried that the other players may not be able to catch them.

There are some ways to pull back a leader.  One is the classic tactic of depressing the price of stock they hold by selling shares in those companies.  It’s not clear whether this threat is sufficient, given that:

1. the sale of shares may give a director the chance to buy over the normal 60% limit, #
2. the depressed value might be regained by multiple stock jumps and
3. the price only drops for every 20% sold by a non-director 

I often see players hanging on to minority holdings in successful companies, for fear of falling even further behind; I’m not sure whether this is the right play but if it is, the consequence is that there is one fewer way for pulling back the leading player.

So one further change that I am considering is as follows:

• Until the Grey Phase, all sales of stock drop the Share Price one space for every share sold, instead of every 20%. 
• From the start of the Grey Phase until the first time that a 6X train is removed at the end of an Operating Round, a sale of a Company’s shares by a player who is not the Director of the Company will reduce the Share Price one space for every 20% of the Company’s shares sold - as in the current rules.
• After a 6X train is removed at the end of an Operating Round, sales of a Company’s shares by a player who is not the Director of the Company will not affect the Share Price.

v26 released to playtesters

I've just uploaded 18GB version 26 to the Google Drive where playtesters can access it.

As the change notes say, the changes in this version reflect feedback and observations from playtests with a broader group of players.

Rules

Insolvent Companies may no longer lay track or place Station Markers.  This makes the option of entering Insolvency incur some disadvantage.

The rules for the Orange zone of the Stock Exchange have been modified.  Shares valued in the Orange zone do not count against your certificate limit at all (instead of counting half).  If you buy shares from this zone above your share limit, you do not have to sell back down to your limit until the shares are worth more than £115.  (More precisely, you may ignore any shares worth £115 or less when determining whether you have to sell back down to the certificate limit).

The rules for Characters and for WW1 have been grouped together as the Advanced Game and are described in the main rule book.  The remaining optional rules are listed as variants in the Setup and Scenarios booklet; they are not recommended for normal play.

Scenarios

Grey tiles are no longer used in the four-player scenario; the number of equivalent brown tiles has been increased to match.  I found that the grey tiles were undoing the impact of station markers placed in the late game so that the markers became just a minor irritation instead of significant game players.

Companies

The Director’s Certificate of the LNWR costs an additional £20 when first purchased (i.e. from the Initial Offering).  I noticed that the LNWR was usually a better buy than the other companies at the start of the game; this £20 cost is an attempt to balance that advantage.

The Midland Railway has a free upgrade for its Home city. It is the only Company to have a special ability.

The GSWR has only three Station Markers, like the LYR and MSLR, for game balance.

Private Railways

The ability of the Taff Vale has been changed to give a free upgrade for Cardiff.  The station marker in Cardiff was rarely used or useful.  The ability of the Leicester & Swannington has also been changed.

In East Anglia, the NE has been replaced with the EU, to reflect the map changes.

Map

There are a number of minor changes throughout the map.   In Scotland, Aberdeen is one hex further away.  Trains may not run to Inverness until the Blue Phase (although track may be laid to connect to the Inverness red hex).  These changes stop the Caledonian from being too strong in the first pair of ORs.

London increases income in every phase (30/40/50/60/70).

In the Midlands, Warrington has been removed.   In the South-West, Exeter has been removed and Taunton added.  In South Wales, Merthyr Tydfil has been added in the TV’s hex.  Two extra brown hexsides have been added in Mid-Wales.

The representation of East Anglia has been changed.  This doesn't actually affect the game much; it just makes the routes more historically accurate.

Holyhead now has a fixed income of 20.  The Irish Mail bonus has been increased to 40 and may be scored for any one of Holyhead, Pembroke or Aberystwyth.

Characters

When setting up the game, randomly select a number of Characters to auction equal to the number of players, instead of having all Characters available in every game.  This adds more variety to the game setup.

Characters are no longer auctioned.  Instead, players choose one each after the Private Railways auction.  They choose in reverse player order and then commence the first Stock Round in normal player order.  There are a number of tweaks to the Characters' abilities, for game balance.

Minor Changes

4+2 trains now cost £300, the same as 4+1 trains.  5+2 trains cost £500 instead of £450 and 4X Trains cost £550 instead of £500.  Companies can now be started at £115 during the Brown Phase.

Station markers

Following on from the preceding post about Companies that start in single-station cities, another way such companies could conceivably be losing out is in the number of tokens they can deploy at the end of the game.  With one token required in their home city, they only have three to place in higher-value cities.  Does this disadvantage them in comparison with Companies that have higher-value cities as their homes?

Six of the twelve Companies in the full game start in single-station cities.  That’s half of them.  So even if there is a disadvantage, it is quite widely spread.  In the two-player game, the mix is rather different.  Only two Companies start in single-station cities, but this leaves the four others available for purchase between the two players, so neither player is necessarily disadvantaged.

Of the other six Companies in the full game, the LYR and MSLR start in the centre of the board in high-value cities. But these Companies only have three tokens in total instead of the usual four, so they don’t have a significant advantage over the single-station cities.

The LNWR and CR start in triple-station cities and have the usual number of tokens, so they might have an advantage.  They are slightly limited in that their route through their home city will never upgrade, so they have to commit early on between routes that generate immediate income and those that have longer-term potential.  Also, the CR, starting in Glasgow, is slightly removed from the action in the centre of the board.  The LNWR may be a bit stronger but I’m happy with this, given its historical significance.

This leaves the GWR and the NBR, which start in double-station cities and so have reasonably high income at the end of the game combined with flexible routes.  The GWR’s home is in Bristol, which is strategically important for the South-West of the board but not too central.  I think this is balanced (and the GWR was one of the larger companies historically, like the LNWR).  The NBR’s home is Edinburgh, which is a key station on the East Coast route to Scotland.  If any company is over-advantaged in tokens, it is the NBR, but it is also yields some of the lowest possible incomes in the early stages of the game.  So I think this is not too unbalanced.

One small change I may make is to give the MR an extra token when it converts to ten shares.  This would give in one more in total than any other company.  I’ve found that the MR tends to underachieve in 18GB compared to its historical reach and this extra token might be a small encouragement for players to expand its network.    The change would also add a little more asymmetry to the game, which I think is a good thing.  There is a risk that combining this change with the free upgrade suggested in the previous post would make the MR too strong; we shall have to keep an eye on it during test games.

Single-station city bias?

A couple of playtesters have commented that the fixed income for single-station city tiles puts the companies that start in such hexes at a disadvantage.  That is not my intention and I review this aspect of the design in this article and the following post.

Although the home cities of these companies yield lower income than larger cities, single-station tiles can be upgraded more easily and can have multiple connections, so it should be easier to build more varied networks and to run multiple trains without laying a second marker.  Some of these companies start near higher-scoring cities and are better placed to upgrade those cities than companies that actually start in the higher-scoring cities.

That is the choice I intended these companies to offer.  This design goal would be forfeit for some companies if it turned out that they were unable to build multiple routes while the game still allows companies to run three trains.  To check this, I have reviewed the possible early incomes of all the companies in the game and have investigated a couple of possible rule changes that might make building multiple routes easier.  Some of changes these would have been quite major ones to make at this stage in the development of the game, so it is fortunate that they turned out to be unnecessary. 

One such option was to ease the restriction on laying two tiles, replacing the current rule that only one may be a city, with a version that applies that limit only to double and triple station cities.  This variant would allow a company to upgrade its single-station home city in addition to another city build.  But an evaluation of all the companies shows that only three (the MR, GWR and SWR) would significantly gain from this rule and the primary advantage would arise from building longer routes rather than multiple routes, so this option didn’t help the stated goal.

Another possibility was to add an extra “prong” to the yellow one-city tiles.  This would allow companies to build three routes without having to upgrade their single-station home city.   But this change breaks the upgrade process.  Companies would have fewer reasons to upgrade their home cities to green, and other companies would have one fewer option for building new track in the hex to force an upgrade.  If the green tiles had two marker spaces instead of one, there would be less incentive to upgrade to brown and if the tile were tokened out it would be harder for another company to do so.

In the end, I have adopted a very limited change.  The Midland Railway now has the special ability of a free upgrade for its home city when laying a green tile there.  Other companies are not affected.  Most of them would not benefit from applying this rule and the GSWR would be too strong.  So a special ability for the Midland Railway is sufficient.

Orange paper tactics

An interesting point arose during a game at Tringcon last weekend.  When a stock falls into the lowest-valued part of the stock market, the shares of that company count as only half a certificate against the players' certificate limit.  This gives an incentive to buy lower-valued shares.  A similar feature exists in several other 18xx games (as discussed in a very old post on this blog).  In some of these games, it is a valid strategy to keep stock in this zone.  This is sometimes called a "yellow paper" strategy.  In 18GB, the zone is coloured orange, hence the title of this post.

The difficulty with this tactic in 18GB is that companies jump multiple spaces on the stock market when their income exceeds a multiple of their stock price (to a maximum of four spaces).  The price of stock in the orange zone is so low that a four-space jump is almost certain. A jump of this size will take the company out of the orange zone and so the shares will now count a normal against the player's certificate limit and he or she will have to sell some shares.  This isn't necessarily a bad thing but it does mean that an "orange paper" tactic is only temporary.

The playtesters at Tringcon felt that this was a lack in the game.  One suggested that I add an extra space on the stock market so that a share in the very lowest space would not exit the orange zone in one move.  But I think that would be tinkering at the edges of the design; it would only have an effect when a company's share price was in that one space on the market.

Instead, if we want to give more longevity to this tactic, a better change would be to make stock prices in the orange zone only ever increase one space per move.  Then players could buy such stock in the knowledge that although the price wouldn't increase immediately, they would get payout for potentially longer.

The main reason I'm not rushing to implement this is just that it doesn't feel right.  Why should low-valued stock not increase in price once it starts paying decent dividends?

One player also noted that in other 18xx games, stock in this zone doesn't count against certificate limits at all, and that our game was so close that this difference would have let him win.  The argument about winning works both ways, of course; the actual winner might feel aggrieved if the rules were changed for that reason alone.

From my point of view, I'm happy to explore changes that remove a difference between 18xx and other games.  The reason for the current rule is discussed in this post but I can think of a different way of avoiding the problem described therein: I can rule that sales of stock never drop a company's share price into the orange zone. Given this protection against the old problem, I may well change the rules back so that stock in the orange zone does not count towards certificate limits.

Putting the sparkle back

After more plays with the new finances, I have decided that the reduced income from cities has had a deleterious effect from the mid-game onwards.  It has tended to keep company income just a little too low for any companies to jump two spaces on the stock market, which in turn has made the companies slightly too similar and has slightly reduced the amount of cash going in company treasuries if they convert.  The cumulative effect was to make the game a bit flat, like drinking champagne without the bubbles.

So I have produced a new version that reinstates the previous income for the major cities (OO and OOO tiles) from the green phase onwards.  Initial tests seem to indicate that this has put the zing back into the game.  You'll need new versions of the map and tiles but the end result is worth it.

In other changes, I have changed some of the graphics to better distinguish the different types of tile.  I've also made most of the certificates the same size rathrer than the hotch-potch they were before.
  I've also removed a couple of minor rules that weren't adding much to the game, in order to keep the focus on the important aspects of the game.  These are all small changes which I think they will make the game easier to play.  But the important change is that those champagne bubbles are fizzing again.

Tuning the new playtest release

I released a playtest version of 18GB with revised finances at the beginning of the year.

So far, the tests imply to me that the share prices of companies are too low, both when they start and when they convert, especially in games with four players.  This means that between them they have slightly less cash to buy trains.  The combination of the two (lower price, fewer trains) will mean they progress slightly slower up the stock market and thus the game will be a little too slow and too long.  Fortunately, it's not too hard to tune the factors that give rise to the underlying causes.

One such factor is that I set the highest opening share price at £100, where previously it had been set to (the equivalent of) £120, one space higher.  I did have a reason for this, to do with the possibility of a new company in the brown phase immediately buying two permanent trains, but I can address that problem in a different way.  So I will undo this change to the range of opening values.

At the lower end of the opening prices, I will return to requiring a choice between £70 and £80. Allowing the intermediate value of £75 seemed like a good idea but it gives players a false impression.  The extra £25 starting cash in the company is not enough to buy a better train or place a station marker and so is pretty pointless. 

The most important fix for the four-player game will be to push the players's starting capital back up a bit.  The new version required a complicated re-vamp of the private railways in order to mesh with the new finance system and starting capital was one factor in this mix.  The total cost of the private railways is slightly lower in the new version but I overcompensated by cutting the starting capital too much.

Once again, I did have another reason for reducing the starting cash.  In 18GB, it is sometimes possible for a player to start two companies in the first stock round.  I wanted to limit the chance of two players choosing this strategy, in case it accelerated the train progression too much.  But it is questionable whether it would be a sensible strategy for players to adopt, so by cutting too much I was addressing a edge case at the expense of the normal cases.  If the edge case does prove to be problematic, I can probably address it in other ways instead.

I may also need to make some additional tweaks to offer more chances for companies to double-jump on the stock market in the early game, but any changes in this regard would likely be isolated and subtle and are just part of normal tuning.

Testing also revealed that the certificate limits may be too low, so I will tweak them slightly.  I haven't revisited these in a long while.