Monday, 22 February 2010

On Asset Stripping (Again)

As I've stated before, I want Britain Under Steam to have companies that change in value throughout the game, so that players are continually faced with investment decisions. This means that the game must have many companies in operation. In turn, this could provide many opportunities for asset stripping.

Asset stripping is an important factor in many 18xx games. One company can buy a train from another at a grossly inflated price, or a measly undervalued one, thus transferring wealth between them.

This is a potential problem in my game. A player could start a new company, strip it of its assets, and sell it back to the bank. This could leave several companies that are such poor value that there is no point in buying them, which would defeat the object of the design. Many 18xx games restrict the sale of shares so that it is harder to get rid of asset-stripped companies, but I want players to be able to swap shares freely. I just want to ensure that companies that are of little worth at one point may become more valuable later on.

I will need to see what happens in playtests. It may be that the asset-stripped companies take loans to buy new trains and thus end up becoming more valuable. Nonetheless, here are some thoughts on ways to limit excessive asset stripping.
  • Require each newly-floated company to buy its first train from the bank. So it can't just give all it cash to an existing company. (It could, of course, sell the new train cheaply)
  • Forbid a company from buying the last train of another company, unless the purchasing company has no trains. (This rule is used in 1860. It should combine well with the previous point).
  • If a company sells a train for less than half price, drop its share price. Conversely, drop its share price if it buys a train for more than full price. (This gives more volatility to the stock market, as well as limiting asset stripping per se).
  • Additionally, an insolvent company may not sell a train for less than half price or buy a train for more than full price.
  • When a player sells the director's certificate of company, give him only the value of a single share instead of the two represented by the certificate.
  • Additionally, if a company is insolvent, its shares sell for half value (so directors are discouraged from maladministration).
  • If the above are too feeble, do not allow companies to sell trains until phase three (or some other phase as fits the design).

Sunday, 21 February 2010

When Trains Rust

My aim for Britain Under Steam is to have companies rising and falling in value, so that players are selling some shares and buying others throughout the game, trying to spot which shares will return most value. The key 18xx mechanism to trigger changes in value is the obsolescence of trains as newer ones replace them. So the question that arises is, what should happen when a company loses all its trains and doesn't have enough funds to buy a new one? I'll borrow a term from 1860 and call such companies "insolvent".

In 1830 and its derivatives, the director of an insolvent company must provide funds from his own hand. If he can't, he is bankrupted. That's not the kind of game I'm looking to design. In any case, one feature of the British railway boom was the innovation of limited liability companies, in which the directors did not have to personally bankroll their corporations.

Some other 18xx games allow insolvent companies to take loans to fund the purchase of new trains. They may then have to pay interest on these loans, be required to repay loans before making further purchases, and/or suffer a reduced share price. This is more in line with my goal, provided that the penalty for loans does noticeably affect the value of the company. 1861, for example, is probably too lenient for my aims; in that game a company's share price is unaffected by loans taken.

The original 18xx game, 1829, and descendants such as 1825 and 1860, allow insolvent companies without directors to borrow a notional train from the bank. A company can collect income from running this borrowed train until it has enough to buy a train of its own. As it is not paying dividends, its share price drops each turn that it remains insolvent. This "train borrowing" mechanism seems a little clunky but it does work.

Perhaps an alternative is to combine the most appropriate aspects of the two approaches, as follows. An insolvent company must take loans to buy a train. A company with loans may not pay dividends, but must withhold income. Hence an insolvent company will drop in share value until it has paid for its trains, just as if we were using the "borrowing" approach.

Into this mix there is also the option for a 5-share company to convert to a 10-share company, gaining the capital to buy a train in the process. I hope that this will be a worthwhile option, so that I don't have to bring in a rule that forces an insolvent company to convert. So directors will have the choice of converting, taking loans, or selling the company altogether.

Tuesday, 16 February 2010

Terrain changes

I'm making several changes to the map. There are several reasons for this, such as improving the way hills and estuaries are represented, adapting to the new N+M trains, and reviewing the historical population of the towns and cities. I'm a little concerned in case some of this is "churn", rather than progress, but some of the changes are definitely improvements.

To start with, I've fixed the problem I noted in a previous post. Companies now have to pay to build through difficult terrain (and can still only do so from phase 5 onwards). I've changed the map in Northern England so that the only north-south route along the Pennines requires two such payments. To make this work, I've also moved Carlisle one hex south-east, so that only one hex lies between Carlisle and Newcastle.

The introduction of a cost for building through difficult terrain (brown hexsides) has a knock-on effect in Scotland. I had brown hexsides for both the Forth and Tay estuaries, which meant that companies would have to pay twice to build the route from Edinburgh to Dundee This is a route I actually want to encourage. In one attempt to fix this, I've collapsed Fife to a single hexside, moving Perth and Dundee one hex south. This seems to work, although it removes some track building options in the area. I was pleased when a change on another part of the board suggested another solution, so I'll explain that next.

The area around the Bristol Channel has always caused some hassle on this map. I was reasonably happy with the current solution, which has a sea hex between Bristol and Cardiff. It meant the Severn Tunnel got built to the north-west of Bristol, and Newport was rather to the north of its actual location, but I could live with that. Then, when I added a tight curve on the Bristol side of the sea hex, I noticed another option. That hex has always included a shallow curve on the Cardiff side; the addition of the new curve produced a pattern that could be upgraded to a brown tile, to add a straight track between Bristol and Cardiff. So instead of a sea hex, I've made this a green hex, upgradable from phase 5 onwards (at a cost), and the Severn Tunnel can now run in the right place.

This suggested an alternative fix for the Forth and Tay estuaries. Instead of collapsing them to a single hexside, I think I can change the intervening hex to a green hex. As with the Severn Tunnel, this can be upgraded from phase 5 onwards at a cost. With only a single tight curve on the green hex, this upgrade has more options than the Severn equivalent.

I'm pretty satisfied with these changes. Each of them improve the relevant area of the map and fix particular problems.

Tuesday, 9 February 2010

On Trains and Towns

For a while, I struggled to find a good way of dealing with the towns that 18xx games represent as small stations, those in which companies cannot place markers. At the start of the game, I wanted 2 trains to run through them (either skipping them as in 1861 or scoring their income as in 18EU). Soon after, I wanted them to block 2 and 3 trains from connecting stations that I deem to be "too far" apart for small trains - such as London to Manchester. Later on, I wanted 5 or 6 trains to be able to run through them. This was, of course, contradictory.

I came up with two solutions, which led to a short discussion on the 18xx list. The first was to limit the number of hexes that trains could run in addition to the number of towns & cities they can call at. So 2 trains would be limited to 5 hexes, 3 trains to 7 hexes, 5 trains to 11 hexes, and so on. You could call these 2/5H, 3/7H, ... trains. (I don't want to just use hex trains because other parts of the map have plenty of cities in close proximity and hex trains would make fortunes).

The second solution was to use + trains, e.g. 2+1 trains that can call at 2 cities and 1 town, 3+1 trains that can call at 3 cities and 1 town (or 2 cities and 2 towns), and so on.

The clear preference from the 18xx list was for the + trains, which agreed with my own intuition. Another suggestion was to use "N of M" trains, which can call at M stations but only count income from N of them (allowing them to skip low value stations). I think the + trains fit my game better, because I don't want trains to skip low value large stations.

Like many solutions, this one looks obvious in retrospect. It took several weeks of pondering and I only stumbled upon the answer when setting up a game of 1812, which uses them.

Monday, 1 February 2010

The Blind Spot

There's always the risk, when designing a game, that you miss something obvious. I've just noticed such an example. I essayed a simple rule for difficult terrain: certain hexsides may not have track laid across them until a certain point of the game. The idea is that by that time, Victorian engineering had advanced and could build routes such as the Severn Tunnel, the Forth Bridge or the Settle-Carlisle line. It's a nice, simple rule.

What I hadn't noticed was that it allowed a company to build track straight along the Pennines and into the Scottish Lowlands, by-passing both Newcastle and Carlisle. This would be rather unhistorical and would affect game balance too.

So I have to do better. One obvious fix would be to make companies pay every time they build across such a hexside. Then someone could build the Pennine Way, if they had enough money and enough incentive. And this would be entirely consistent with other 18xx games.

Another fix would be to add some impassable hexsides, so that the route would not even be possible.

I need to experiment to work out which is best.