Looking for input on Stock Market Mechanic

I'm a CS major myself and have spent some time trying to design exactly what you've done, using a pc to compliment a board game. The issue with this game is it seems you have time sensitive transactions, and no way to record or process transactions instantly. Basically players would be scrambling for the sake of scrambling. Why not just simulate the market in a more "supply and demand" style? Something really simple, but more interactive than a random market simulator program.

Say at the start of each market round each player has a secret notepad that they can write transactions on. Then players reveal and then the price of the goods each player purchases scales with the number of players buying/selling that same resource. If a player can't afford an order they are take only what they can afford but can't replace orders (because the market has closed).

So say you have ore, wood, oil, and gold for example. Each player places bids like so

Player A:
Buy 2 Ore, 2 Wood
Sell 1 Oil

Player B:
Buy 3 Oil
Sell 5 Wood

Player C:
Buy 1 Ore, 1 Wood, 3 Gold
Sell 1 Oil

Player D:
Buy 2 Gold

And then for the price of each good just keep a sliding scale (1-10) starting a 5 for each resource and keep track through the game. For each order to buy a good the price goes up 1, for each order to sell the good the price goes down 1. So say this is the first round, first you move the price indicators based on each order leaving the prices as such:

Ore: +3 (3 buys, no sells) -> 8
Wood: -2 (3 buys, 5 sells) -> 3
Oil: +1 (3 buys, 2 sells) -> 6
Gold: +5 (5 buys, no sells) -> 10

Then each player transfers resources and pays the new market price. This is a simple system (relatively), but allows a lot of great interactions. Players can attempt to drive the price up on goods, or flood the market with them to lower their value. And assuming you do a good job of making each resource equally useful (or at least universally useful at all points in the game), then the market should self balance since players will instinctively sell high priced goods and buy low priced goods.

You could even put a mechanic in the game where if any 3 resources reach 1 price in the same turn, the market crashes all players lose half their wealth or some other global negative effect (halt on production as the workers strike, for example).

This system benefits from not needing a pc to play a boardgame, as well as being more more cerebral and strategic, while still being simple and elegant enough to need only one section of the board to be the market price ticker and 4 pegs to track the prices.

A stock or commodity marketplace can be described as a version of the minority game where the payoff is inversely correlated to the number of other people who make the same choice. If everyone is selling, you want to buy. If everyone is buying, you want to sell. All things being equal.

Now, in the real marketplace, there are exogenous reasons for both buying and selling. It won't do you much good to be buy up all the stock in the Titanic as it is sinking just because everyone else is selling. So... as long as your market mechanic is tied to some other mechanic that puts those commodities to use (producing and consuming them), it no longer becomes a strict minority game.

Likewise, speculative behavior in the marketplace cannot be described as a strict minority game. In speculation, you aren't buying or selling something because you or someone else requires its use value, but because of your belief in its future exchange value. So, in this case, you're choosing to buy something precisely because you think everyone else will. However, this can actually still be seen as a form of the minority game, once you introduce time. No one wants it now... but they will.

So, when both external use value and temporal exchange value matter, you end up with a very complicated valuation system that sufficiently complicates the minority game to make it interesting to players. Perhaps that is why both trading and stockmarket exchanges are some of the most popular game dynamics outside of auctions. They give you a wide variety of strategies to pursue whose advantage depend greatly upon those of the other players.

You have already noticed a few issue with simultaneous trades (false or naked bids), but I think there are a few others that might turn out to be a nuisance:

What if the value of an asset falls below 0? This is a problem not only for asset valuation but also for liquidity.
What if the value of an asset rises above 10? The math required to process these trades may become quite taxing.
What if I buy something now, and sell it the next turn without anyone else trading that commodity? Under your pricing system, I would lose $1. The same problem also occurs if you sell then buy. Transactions with the market inherently lose you money unless someone else is involved. Is this fair? And who gets this money?

A marketplace really represents an aggregation of buyers and sellers and the brokers who maintain the liquidity. Marketplaces without a sufficient amount of liquidity often breakdown and operate in strange, counter-productive, ways because there aren't enough buyers to cover all the sellers (or vice versa). Under these circumstances, it actually becomes much more efficient to bypass the marketplace and buy/sell/trade directly between the asset owners. This occurs whenever the owners are selling/buying in such large volumes that it represents a significant proportion of the total liquidity or when there aren't enough counterparties willing to participate (such as in an economic crash).

Without a computer and a thousand other players, you are likely to find yourself in a situation where liquidity matters. The proportion of goods each player holds compared to the total available will be very high. You will end up with situations where asset prices will swing wildly and perhaps even dip below 0. Although speculators like volatility, I'm not sure that this kind of volatility will be fun for a player. At best it will be a degenerate marketplace. But, more likely, it will be an empty one as player's natural risk aversion will make them cautious to trade in an environment where the value of a good is prospectively indeterminant.

As a result, you will probably want to introduce an actual pool/market with structure (turn based decisions) or allow players to trade goods/assets directly with each other. If there is a turn based structure, then the marketplace will provide a certain degree of liquidity and prices will rise fall incrementally as players buy/sell to it. Speculation, hedging and commodity hording can still take place as long as there are enough turns and enough commodities that players want at different times and you will not have a problem with verification ( fake or naked bid) . But, you won't be able to pump and dump or do other things that rely upon other people's ignorance and there is a chance that there may be permanent commodity shortages if the use value of a good is higher than its market value.

On the other hand, trading goods directly will allow players to always determine the true value between themselves. Liquidity is dictated by the mechanism that generates goods to trade and the rate that producers decide to consume them. And, verification would be self-enforcing in each bilateral interaction. However, direct trades will enable greater collusion and deliberate favoritism... which may or may not be enjoyable. As long as trade is voluntary, a trading game can frequently curtail the trading opportunities for the winning player. Since it is often strategically better to trade with person in last place, there is a built in catch up mechanism which may or may not seem fair.

Interesting.

Would the price reset at the beginning of each round, or would it persist between rounds and reshufflings? If there are an equal number of positive as well as negative numbers, I believe that the random walk will create a normal distribution with a bias towards the initial price fluctuation. So... if the first card is negative, the mean value of the commodity will tend to be slightly negative unless a significant number of trials are performed (probably too many for your game).

However, if each card has multiple values, there will also be a correlative relationship between all the commodity values. Depending upon how many commodities you have, you may be able to balance this correlative effect by including all possible combinations of values to prevent any systemic bias in the game. But, there will still be a dynamic correlation (based upon which cards are drawn) in each game. The end result, I suspect, will be a somewhat complex statistical environment, but not something that is responsive to play decisions.

Since random card draws will determine commodity values, the buying/selling mechanic will be completely independent of the rest of the game. Players won't necessarily be responding to S&D as much as exploiting statistical imbalances. As a result, it is not clear to me how this system will be interactive in any meaningful way. Like blackjack with a 1000 deck shoe, it really doesn't matter what anyone else does... your decision will always be based upon what the house has and what you have. When a price is lower than average, you buy. When it is higher than average, you sell. All things being equal.

There may be exogenous reasons for why you would want to buy high or sell low, but they won't be based upon what other players are buying. So... you're not really talking about a stockmarket at all as much as a store. In a stockmarket, the price that is discovered through each voluntary buy/sell order is really the outcome of a trade where each party has a different opinion of the direction of the price... And that direction is as much determined by what other people think as it is the use value of the good/asset.

But, it sounds to me that you need to model a commodities market more than a stockmarket. In a merchantile exchange, people buy/sell contracts for certain goods that are yet to be produced, but which they anticipate that they will need. Producers sell these contracts in order to raise funds to employ people to produce the good based upon their projected costs, estimated yield and desired profitability. The farther out the time frame that a contract is, the more likely it will be either undervalued or overvalued because there are a number of events which could happen in the meantime that affect supply and demand. So... merchantile traders buy these contracts and buy and sell them to each other and end consumers and charge prices based upon their most current estimates of price when the contract becomes due... exploiting informational arbitrage in most cases. The closer to the close of the contract, the more 'true' the value of the price of the commodity will be.

Since speculative behavior in an exchange is based equally upon exogenous S&D as from their own internal minority gamesmanship, it is possible for merchantile exchanges to discover prices which are wildly out of balance with sustainable S&D. In fact, asset bubbles and commodity spikes frequently occur when there is a positive feedback between the process of merchantile evaluation and the supply of the goods being evaluated. But, under normal circumstances, commodity prices bridge the rate of consumption with the rate of production.

In games like Power Grid, the production of commodities is steady, but the consumption is determined by the actions of the players. In general, you want to buy a type of power plant that no one else is using so that you don't have to compete for fuel, but there are, of course, a lot of other reasons to buy them, least of which is the question of timing in the game. To make sure you run your plants at peak efficiency, you can stockpile fuel, but usually the opportunity cost of doing so makes stockpiling limited. If there were a mechanism to sell back the fuel for the current market rate, you might be able to engage in speculative behavior. But, ultimately, the game is structured tightly enough that it would be difficult to make this a significant advantage. With players constantly upgrading their plants, a built in mechanism that favors self-sufficient energy sources and so few rounds in the game, it is not likely that buying up goods would overcome the opportunity cost of doing so. So, Power Grid often comes down to having the right amount of money at the right time. With a side show of auction based price discovery that tests your familiarity with the rules, expected values of resources and the cost of lines on the map.

But, the lesson of Power Grid for any player is to avoid competition: try to be the only one building in this area, the only one buying this fuel, the only one with cash to bid, etc... And, the end result of the game is a somewhat disappointing realization that the game never escapes a king-maker scenario. From the very beginning, players are forced to choose who they will compete with. Once two people have chosen a separate direction. The third person is forced to compete with one or the other and in so doing will lose the game for both players (all things being equal). But, this degenerate multiplayer Hawk-Dove dynamic is dressed up in enough complexity that it is still worth a few games, and they keep making new versions... so apparently it is something fun.

However, a lot of Power Grid's dynamic emerges from mechanics exogenous to the commodity exchange. The actual purchasing mechanism is quite robust and as I said, it might allow players to engage in some degree of speculator behavior if there were enough rounds to do so. So, you might consider using something similar to Power Grid if you want to build gamesmanship into how players acquire commodities. An even more robust system may also allow players to also produce commodities to add to the market. But it really all depends upon the level of complexity and number of steps you wish to introduce.

Catan gets away with a simple trading system because players are both consumers and producers of goods. The barter system works because you risk losing what you have accumulated beyond 7 cards and the price structure is not very fine grained. On the other hand, all you can do in Power Grid is bet on the level of demand because there are a significant number of issues that effect it and a variety of other factors to consider. So, perhaps, if the market mechanic is only a minor part of the game you have in mind, the Texas Hold'em idea will work. But, I believe that it still needs to have some way for the players to effect price based upon either the supply, the demand or both, or otherwise it becomes a game of statistics and you might as well keep the price constant.

The problem is non-linear. If I buy then sell two of the same good, I lose 4. Three, I lose 9. Although this will prohibit players from 'gaming' the other part of the system, it will have a significant chilling effect on speculative behavior. Your loss due to speculation will be exponentially inversely ( 1 / x^2) related to the degree of uncertainty about the actions of your counterparties. If they don't do what you expect them to do, then you risk losing a lot. But, what is the corresponding payoff. Let's say you guess right. You correctly expect the value of a good to be driven up by demand, so you sell into it, and to drop again, so you buy it back. What's the upside:

If it starts at 5 and you have unlimited stocks, and everyone else wants to collectively buy 5, then you will make 4 if you sell one, 6 if you sell two, 6 if you sell three, 4 if you sell 4 and 0 if you sell five. The optimum is to sell two to gain 6. The same can be said for selling.

Now, let's see how that number changes based upon true demand. If it is even higher, you earn windfall profits, and possibly lose out to having sold 3 or 4 into higher demand, but always less than half the number of the actual demand. On the other hand, if there is only 2 demand, you earn nothing, and everything less than 2 is a loss to you.

So... the calculation shows that you need to not only be pretty sure of direction but also of degree if you are going to make any money speculating. And, when the yield is compared to the principal, it is very likely that only correct guesses with big demand will overcome the opportunity cost of putting those same resources to work for production/consumption.

I don't know exactly how many rounds or the number of commodities traded each round there will be... but it becomes clear that unless there is both a large number of rounds and a large number of commodities traded (two conditions unlikely in a tabletop game), this mechanism will not support speculative behavior.

And, the problem with not supporting speculative behavior is the problem of the degenerate coordination game (aka.... multiplayer Hawk-Dove). If it doesn't make sense for you to both buy and sell, then you will either aways be selling (if you are a producer) or you will always be buying (if you are a consumer). If you are the consumer, you basically want to pick the commodity that no one else is buying to keep prices low. The same can be said as a producer if you are selling. Unfortunately, let's say that you are absolutely certain that player X will buy Y copies of commodity Z. If you also buy it, you both lose. If you buy something else, you do nothing to stop him and if it turns out that he wants what you want, then you will lose no matter what you do.

Being able to sell into an aggressive buyer is what allows people to take advantage of another player's lack of hidden agenda or desperation. But, if the actual mechanism for doing so has a negative expectation value, then everyone will be naturally disinclined to exercise this option in the absence of perfect knowledge. If accuracy is distributed evenly at random across the entirety of the game, then you will not only lose through a speculative strategy, you will end up with less than you started with.

So, I still think you need a mechanism that has no transaction cost. Something where price movements are symmetric. Where if you buy at a price and sell back immediately, the cost is 0. Such a system requires incremental price increases (and hence probably turn based structure) or direct selling between buyers and sellers.