Rig the Game or Play

crs48 with Claude · History

Everyone has a Monopoly story, and it is always the same story. Someone flipped the board. Someone went to bed. The game was abandoned at half past eleven with two players bankrupt, one player quietly reading a book, and a winner counting notes on a board where nothing further was going to happen. Nobody’s Monopoly story ends and then we finished it, and it was wonderful.

Hold onto that memory, because it is evidence, and it is evidence about something much larger than a board game. Then consider where the board came from.

The rule set that was supposed to lose

On 5 January 1904 the United States Patent Office granted patent number 748,626 to a stenographer, poet, actress and committed Georgist from Illinois named Elizabeth Magie. Read the patent today and the first thing you notice is how flatly it describes what you already know:

“The object of the game is to obtain as much Wealth or money as possible, the player having the greatest amount of wealth at the end of the game … being the winner.”

That is the whole object, as filed. No mention of landlords, or injustice, or the single tax. If the 1904 patent were the only surviving document, you would conclude that Magie set out to invent exactly the game we play.

But it isn’t the only document, and the other two are the interesting ones. Two years earlier, in the autumn 1902 issue of the Single Tax Review, she had already explained what the thing was for: “It is a practical demonstration of the present system of land-grabbing with all its usual outcomes and consequences.” Her hope was pedagogical and, read now, rather touching — that if children could “once see clearly the gross injustice of our present land system,” they would grow up and remedy it. And twenty years later, when she filed a second patent in 1924, she wrote the purpose into the specification itself: the game was meant “not only to afford amusement to the players, but to illustrate to them how under the present or prevailing system of land tenure, the landlord has an advantage over other enterprises and single tax would discourage land speculation.”

So the intent is documented, in her own words, on both sides of the patent that doesn’t mention it. And the game itself is the one you have played: buy land, build on it, charge rent to anyone unlucky enough to arrive, and keep charging until they have nothing left. The board art carries the slogan LABOR UPON MOTHER EARTH PRODUCES WAGES. Magie was a follower of Henry George, who argued that the value of land is created by the community around it and collected by whoever happens to hold the deed. She built a machine to make that feeling unmistakable — by doing it to you, over an evening, with your friends.

It worked. That is the whole problem. The demonstration was so effective that it detached from the lesson and travelled on its own — through Quaker households in Atlantic City, through college campuses, hand-copied and house-ruled for thirty years, until a version reached Charles Darrow, who sold it to Parker Brothers in 1935 as his own Depression-era invention. Parker Brothers bought Magie’s later patent separately, for $500, and had shelved her game within a year. It has since sold somewhere north of thirty million copies under the name of the thing it was built to indict.

You will read, in a lot of places, that Magie wrote two sets of rules so that players could try both and work out for themselves which was better. Be careful with that story — the patents contain no such thing, and we will come back to what is and isn’t documented at the foot of this essay. What does exist is a 1932 edition, published by the Adgame Company, that prints two ways to play side by side. One is called The Landlord’s Game, and its object is “to become the Big Landlord, which title he wins when he has accumulated $2,000.” The other is called Prosperity, and its finish is a genuinely strange thing to find on a board-game rules sheet:

“When the player with the least money has $7000 (double his original capital), Prosperity is achieved. The game may continue at the pleasure of the players.”

Read that twice. You do not win. Prosperity is achieved — a state of the whole table, measured at its poorest member — and then the game simply carries on for as long as everybody is enjoying it. Hold onto that last sentence; we will need it later.

What is not in doubt is the direction of travel. Of the two ways to play, the one that spread, and kept spreading, and is in a cupboard in your house right now, is the one where you win by bankrupting your friends.

A game that stops being a game

Set the politics aside for a moment and look at Monopoly the way a working game designer would. It is, by the standards of the craft, a catalogue of things you are taught not to do.

Its central engine is a positive feedback loop: money buys property, property collects rent, rent buys more property. The advantage of being ahead is that you get further ahead. Designers call the result the runaway leader problem, and modern games are built with elaborate machinery to prevent it — catch-up mechanics, rubber-banding, escalating costs, turn order that favours the trailing player. There is a long-running list on BoardGameGeek of nothing but solutions to this one problem. Monopoly has none of them. It is, in a precise sense, a machine for converting a small early lead into a certainty.

It has player elimination, which means the punishment for losing is being made to watch. And it has what is politely called a long endgame and honestly called a dead zone: the stretch — often an hour, sometimes two — between the moment the outcome is actually decided and the moment the rules admit it. During that stretch the dice are still rolling and nothing is being decided. The players who are out have no moves. The players who are nearly out have no moves that matter. The player who is winning is performing an arithmetic chore.

This is what a rigged game feels like from the inside, and the feeling arrives long before the injustice does. You do not resent the winner. You are just no longer playing anything. The decisions have drained out of the evening and left the procedure behind.

flowchart LR
  P["Play"] --> V["Variety:<br/>surprising moves"]
  V --> P
  P --> R["Rig"]
  R --> C["Advantage<br/>compounds"]
  C --> E["Players leave<br/>or are removed"]
  E --> S["Search stops.<br/>No new moves."]
  S -.->|"a board nobody<br/>is playing on"| R

And then we make it worse on purpose

Here is the part that changed how I think about all of this.

Almost nobody plays Monopoly by its actual rules, and the two most common house rules both push in the same direction.

The first: the official rules contain no jackpot on Free Parking. Hasbro’s rulebook could hardly be blunter — a player landing there “does not receive any money, property or reward of any kind. This is just a ‘free’ resting place.” The near-universal table custom of pooling fines and taxes in the middle and paying them out to whoever lands on the corner is an invention, and what it invents is a random cash injection — which lengthens the game and feeds the snowball, because a windfall is worth far more to a player who already owns things to build on.

The second, and the important one: the official rules require an auction. “If you do not wish to buy the property, the Banker sells it at auction to the highest bidder,” and — this is the good part — “any player, including the one who declined the option to buy it at the printed price, may bid. Bidding may start at any price.” The property does not stay in the box. It goes to whoever wants it most, immediately, at a price the table sets. Most people skip this entirely. Many have never heard of it.

Think about what those two edits do together. The auction is the game’s only real competitive mechanism: the one place where judgement beats dice, where you must price an asset against your rivals under cash-flow pressure, and where a player who is behind can strike. The Free Parking jackpot is pure unearned windfall. So the two things people reliably do to Monopoly are remove the competition and add the extraction — and the result is the four-hour slog everyone complains about, which is then blamed on the box.

That is not a fact about board games. That is a fact about people, and it is the whole essay in miniature. Nobody sits down on a Tuesday and decides to rig a market. They make a series of individually reasonable edits — this bidding process is a hassle, this default is more convenient, this integration should just work for our own products — and each edit quietly removes a place where someone could have competed, or quietly adds a place where value arrives without being earned. Rigging is rarely a decision. It is an accumulation of house rules.

Boredom is a measurement

The instinct to treat “this isn’t fun any more” as a soft complaint is wrong. In a system with many participants, fun is a fairly precise proxy for something quite hard-edged: whether the system is still generating moves nobody had thought of yet.

In 1968 Friedrich Hayek gave a lecture with the deliberately unromantic title Competition as a Discovery Procedure, and the line that matters is this one: competition, he said, is “a procedure for discovering facts which, if the procedure did not exist, would remain unknown or at least would not be used.” His deeper point is sharper than the quotation. He thought the textbook idea of “perfect competition” — every actor knowing every price, every cost, every preference — actually describes a world in which competition has become pointless, because there is nothing left to find out. Competition earns its keep precisely because everyone is ignorant. It is a search.

A monopoly is what a search looks like when it stops. Not a search that concluded — a search that was switched off, with most of the space unvisited. And the monopolist is not merely indifferent to continuing it; they are structurally disinclined. Kenneth Arrow made the argument in 1962 and it has been awkward for incumbents ever since: a firm with a monopoly has less incentive to innovate than a firm facing competition, because its innovation mostly cannibalises its own existing profits, while a challenger’s innovation displaces somebody else’s. The economists call it the replacement effect. Anyone who has watched a dominant product go quiet for five years has seen it.

So when a market stops producing surprising things — when every app looks like every other app, when the interesting move is always the one nobody is allowed to make — that is not a mood or a nostalgia. It is a reading off an instrument. The boredom is the search having stopped. (There is a version of this argument that runs through soil rather than strategy — a field of identical corn is a search that stopped too. We made it in The Forest and the Field.)

The honest shape of the curve

Now the part that costs us something, because an essay that only collects supporting evidence is an advertisement.

“Competition good, monopoly bad” is not what the economics says. The best-known empirical treatment — Aghion, Bloom, Blundell, Griffith and Howitt, in 2005 — found the relationship between competition and innovation is an inverted U. Too little competition suppresses innovation, as Arrow predicted. But so does too much: in a market atomised to the point where margins are gone, nobody can fund the attempt, and firms far enough behind stop trying at all because they cannot see a path to catching up. The most inventive place on the curve is the middle, where rivals are close enough to each other that pulling ahead looks achievable — what the authors call escaping competition.

Schumpeter deserves his hearing too. His later argument was that scale and market power are what make serious innovation affordable: someone has to absorb the risk and fund the laboratory. But note what his defence actually defends. It is a defence of temporary monopoly, won by being first and held only until someone does better — monopoly as the prize in a game still being played. It is not a defence of a position that cannot be taken from you. Those are opposite things wearing the same word.

So the claim worth making is narrower than the one that would be more fun to write. It is not that competition is always good, or that scale is bad, or that every large company is a parasite. It is this: a monopoly that cannot be lost is the worst point on the curve, and it is the point that every dominant strategy walks toward on its own. The trouble with rent is not only that it is unfair. It is that it is an absorbing state. Once you are in it, the cheapest way to stay there is to keep the game from being played.

The third rule set

Which brings us back to the board, and to the possibility Magie’s admirers wanted so badly to be true.

In 1986 the philosopher James Carse opened a small strange book with a distinction that has been quietly rearranging people’s thinking ever since: “A finite game is played for the purpose of winning, an infinite game for the purpose of continuing the play.” Finite games have fixed rules, fixed players and an ending. Infinite games have rules too — but the rules exist to stop the game from ending, and they get changed whenever they threaten to.

Carse was writing in 1986. But look again at that 1932 rules sheet, which beat him to it by half a century and without any of the vocabulary: the game may continue at the pleasure of the players. That is an infinite game, written down on a piece of cardboard, as a finish condition. The Landlord’s Game ends when one person has $2,000 and everyone else is finished. Prosperity doesn’t end at all; it reaches a state, and then defers to whether people are still having a good time.

That is the real difference between the two ways of playing Magie’s board, and it is not fundamentally about money. The monopolist rule set is finite-game logic taken to its conclusion: you win by ending the game, and the game ends because everyone else has been removed from it. The other is infinite-game logic: you are trying to keep enough players at the table that the thing keeps generating moves — including the moves that beat you.

But cooperation is not a free win, and this is where the nice version of the story has to be given up. Cooperative games have their own way of collapsing, and it has a name: quarterbacking. In a game like Pandemic, where everyone can see everything, one strong player can simply compute the whole table’s optimal move, and the others become an audience holding cards. Everybody wins and nobody played. The best cooperative designs — Spirit Island is the usual example — solve it structurally, by giving each player enough private complexity that no single person can solve the table.

Which tells you the failure is not really about competition or cooperation at all. Monopoly’s dead zone and co-op quarterbacking are the same disease with opposite politics: the game collapsed into one player and an audience. Every rule set has a way of doing that. The design question is never “are we competing or cooperating” — it is “which collapse are we guarding against, and what in the rules does the guarding?”

Communities that get this right tend to arrive at the answer the hard way, by writing rules that keep the table populated: who may join, who watches, what happens on a first offence, who gets to change the rules. We wrote about one such place — a members-only music library that rebuilt most of Elinor Ostrom’s commons principles from scratch without having read her — in The World’s Greatest Record Store.

How anyone knows any of this

There is one more thing about Lizzie Magie, and it is the part I find hardest to shake.

For roughly forty years, the public story of Monopoly was the one Parker Brothers printed: an unemployed salesman in Depression-era Pennsylvania invented a game at his kitchen table and saved his family. Magie gave interviews in January 1936 — she brought out her original boards, showed them to the Washington Post and the Washington Evening Star, and said plainly that this was hers. It changed nothing. The company that owned the story had no reason to tell a different one, and no obligation to.

What eventually changed it was a lawsuit, and not even one about her. In 1974 Parker Brothers sent a cease-and-desist to Ralph Anspach, an economics professor who had made a game called Anti-Monopoly. He fought it, and the litigation ground on for nine years, through a reversal, a remand and a second appeal, until the Supreme Court declined to hear it in 1983. Building his defence — that the trademark should not stand because the game predated the 1935 purchase and had circulated for thirty years as folk property — Anspach went looking for the game’s real lineage, and found Magie’s patents, and the Quaker households, and the hand-copied prehistory. The histories written since, Mary Pilon’s especially, rest heavily on what he dug up. Essentially everything the public now knows about the origin of the best-selling board game in history surfaced because a large company sued a small one and the small one would not settle.

Sit with the general form of that. The rules of a rigged game are usually exposed by someone who got hurt and would not go away — not by the operator’s transparency report. Nobody with the documents volunteers them. The information asymmetry is not a bug in the arrangement; it is a load-bearing part of it. Which is an uncomfortable thing to notice if you are, as we are, a company writing an essay about how honest it intends to be.

A court can name the rig without ending it

A modern example, in case the board game feels like a long way from anything.

On 5 August 2024, in a 277-page opinion, Judge Amit Mehta wrote: “Google is a monopolist, and it has acted as one to maintain its monopoly.” The case turned substantially on default placement — the payments that made Google the pre-set search engine on other people’s devices and browsers. Not a better product forced on anyone: a position purchased and held. The rig, found as fact, in a federal court.

And then the remedies were largely procedural. No divestiture, no structural break-up. Which is the detail worth carrying away, and the reason this section is not a triumphant one: a court can determine that the game was rigged and still not restore the thing that was lost. The lost thing was never the market share. It was fifteen years of moves nobody got to make — the search engines not started, the ranking ideas not tried, because the position at the top was not actually available to compete for. You cannot order those into existence afterwards. The dead zone does not get replayed.

This is why treating monopoly as purely a fairness problem under-describes it. Unfairness can, in principle, be compensated. A search that did not happen cannot be.

So: our rules, published in advance

We build a company, and one of the awkward facts about writing this essay is that every incentive described in it applies to us. The temptation to rig arrives dressed as reasonableness — as convenience, as fairness to the people who do the work, as a default that would just be better for everyone. It never arrives labelled.

The only defence we have found is the one that works on house rules: write the rules down while nobody is losing, and make changing them expensive. Ours are in the Charter, and every new way we might make money has to pass four tests before it ships.

Three of them ask whether a lane is fair to you. The improvement test: does the margin pay for labour, capital or operations we actually provide — or for access to something you would own anyway? The BATNA test: after this ships, is self-hosting still a real, undegraded alternative? The vanish test: if this company disappeared tomorrow, does what you paid for survive us?

The fourth one is different, and it is the one this essay is really about. The sleep test asks: if a well-funded competitor released our entire feature set as open source tomorrow, which of our revenue lines would still be there? A business whose honest answer is none is not a business, it is a position — and a company defending a position will eventually break every promise in its charter rather than fall off it. We would rather find that out now, on paper, than discover it in a bad quarter.

The concrete version is unglamorous, and most of it is just the right to leave made mechanical. Your data leaves in a signed, verifiable bundle, free, and a fork is a complete replica rather than a reference. Your keys are yours and work on any hub. The wire format and the hub are MIT-licensed, so a hub you run yourself never phones home to us. There is no take rate on direct sales of your work. If we ever build an index, it has to be reproducible by somebody else running the same crawler, because an index you cannot reproduce is a chokepoint no matter how open the format is.

None of that is charity, and we should be honest that it is not primarily generosity either. It is a bet that the slope is a better place to stand than the cliff: that being genuinely better is a position you can lose gradually and win back, whereas being unavoidable is a position you hold entirely until you don’t. Rent fails all at once. Improvement fails a bit at a time. You only lose sleep on cliffs.

And underneath the arithmetic there is a plainer reason, which is the one I actually believe. The reason to keep a game honest is not that rigging it is wicked. It is that a rigged game is not worth playing — not even for the person winning. What the rig destroys first is the thing everybody came for: the creative plays, the moves nobody anticipated, the evolution that only happens when the outcome is genuinely unsettled. Magie built a board to show people something about land. What it demonstrates just as clearly, to anyone who has sat through the last hour of a game, is that the winning strategy and the good evening are two different things, and the first one eats the second.

You can rig the game, or you can play. It turns out you cannot do both, because a rigged game is not a game any more — it is just an arrangement, and the dice on the table are decoration.

Sources

• Magie’s own words: her first known commentary on the game in the Single Tax Review, Autumn 1902 (“a practical demonstration of the present system of land-grabbing”, and the passage about children and the land system), and the two patents — US 748,626 (granted 5 January 1904; its stated object is “to obtain as much Wealth or money as possible”) and US 1,509,312 (granted 23 September 1924), which is where the statement about landlords, land tenure and the single tax actually appears.
• The two rule sets: The Landlord’s Game and Prosperity, Adgame Co., 1932 — the primary document for both the “Big Landlord … $2,000” finish and the Prosperity finish quoted in full above. On what is and isn’t verifiable about it, see Snopes, 27 December 2024, and Wikipedia — The Landlord’s Game.
• The litigation that recovered the history: Anti-Monopoly, Inc. v. General Mills Fun Group, 515 F. Supp. 448 (N.D. Cal. 1981), docket No. C-74-0529 SW, through cert denied in 1983; Ralph Anspach’s papers are catalogued in the Anspach Collection. The synthesised account is Mary Pilon’s The Monopolists (2015) — NPR’s review summarises it. Note that the accessible appellate opinion does not itself discuss Magie; the reconstruction is reported by the histories built on Anspach’s research.
• The rules everyone breaks: Hasbro’s official Monopoly rulebook (instruction sheet 00009) for the mandatory auction — “any player, including the one who declined the option to buy it at the printed price, may bid” — and for Free Parking being “just a ‘free’ resting place”. On runaway leaders as a design problem, University XP’s survey of the runaway-leader problem; BoardGameGeek maintains a long geeklist (204332) of designers’ solutions to the same “rich get richer” loop.
• The economics: F. A. Hayek, Competition as a Discovery Procedure (lecture, 1968), quoted from Marcellus Snow’s translation, Quarterly Journal of Austrian Economics 5(3), 2002, p. 9; Kenneth Arrow, Economic Welfare and the Allocation of Resources for Invention (1962), for the replacement effect; and Aghion, Bloom, Blundell, Griffith & Howitt, Competition and Innovation: An Inverted-U Relationship, Quarterly Journal of Economics 120(2), 2005, pp. 701–728 (linked as the NBER working paper, w9269).
• The games: James P. Carse, Finite and Infinite Games (Free Press, 1986), ch. 1; on quarterbacking, Gideon’s Gaming on the co-op alpha-player problem and Spirit Island as the structural counter-design. On commons that keep players at the table, Elinor Ostrom, Governing the Commons (Cambridge University Press, 1990) — discussed at greater length in our essay on Oink’s Pink Palace.
• The modern rig: United States v. Google LLC, Judge Amit P. Mehta, D.D.C., 5 August 2024 (“Google is a monopolist…”), with the procedural-remedies outcome.
• Our own rules, in the open: the four tests in the Charter §6, the reasoning and the honest gaps in ECONOMICS.md, and the working notes behind this essay in exploration 0363 — including a list of the corrections the fact-check forced.

Several widely-quoted game-design aphorisms about Monopoly could not be traced to a primary source while writing this, so none of them appear here. Where this essay makes a claim about design rather than reporting one, it is ours to defend.