When the Prize Isn't Money: Green Cards, School Places, and the Draft Lottery That Wasn't Random
Most lotteries sell you a small chance at money you never had. A different kind hands out things money cannot reliably buy - a visa, a place at a school, a year of your life. Nobody enters these for entertainment, and that changes everything about the draw: when the stake is somebody's future rather than two euros, the randomness has to be beyond argument. On 1 December 1969 the United States ran the most consequential lottery in its history and got exactly that part wrong. It was not a conspiracy. It was a mixing problem - and anyone who plotted the results could see it.
When a Game Becomes an Allocation Method
In a money lottery the worst outcome is that two euros are gone. You opted in, you knew the odds, and nobody but you carries the consequences. All three of those things vanish the moment the prize stops being money.
An allocation lottery hands out something scarce that cannot be bought: a right of residence, a school place, a home - or, at the sharpest end, the order in which young men are sent to a war. The people in the draw often did not choose to be. And that shifts the standard entirely: randomness is no longer the product, it is the only thing legitimising the whole procedure.
Four Draws With No Cash Prize
The original allocation lottery
Athens filled most public offices not by election but by lot, using a stone slab called a kleroterion: citizens' identity tokens went into slots, and a tube of black and white balls decided which rows were selected. The reasoning was explicitly political - drawing lots cannot be canvassed, bribed or inherited. Randomness was chosen as a defence against influence, which is still the best argument for it.
The draft lottery that failed
To decide the order of conscription for Vietnam, 366 capsules - one per possible birthdate - were drawn on live television. The capsules had been loaded into the container month by month, January first and December last, tipped into a glass jar and never properly mixed. The draw preserved that loading order far too well: birthdates late in the year drew systematically low numbers, meaning young men born in the autumn were called up first.
The green card draw that was cancelled
The US Diversity Visa lottery offers a small number of permanent residencies - on the order of fifty thousand - against many millions of entries. In 2011 the State Department published results, then voided them: a programming error had drawn almost all selectees from applicants who entered in the first days of the window rather than across the whole period. Around 22,000 people were told they had won, and then told they had not.
School places, housing, and the quiet ballots
Where demand exceeds supply and no ranking feels legitimate, the ballot keeps reappearing: oversubscribed schools in many cities allocate places at random, affordable-housing schemes run public draws, and major event tickets are balloted rather than sold first-come. Researchers quietly love these draws, because a genuine lottery creates two comparable groups and turns an admissions policy into a natural experiment.
How to See Broken Randomness on Graph Paper
The remarkable thing about 1969 is how fast the flaw showed. You needed to know nothing about the procedure: it was enough to average the drawn number for each month. January sat high, December sat markedly lower, and in between the line slid down almost steadily. Under genuine randomness all twelve monthly averages should have wobbled around the same value.
The explanation was mundane, which is exactly why it teaches so well: capsules loaded in monthly blocks stay in monthly blocks unless something mixes them thoroughly. No intent was required, just too little physics. For the following year's draw the procedure was rebuilt, using two separate randomisation steps instead of one.
How modern draws are securedThree Lessons From Two Failures
Mix, then mix again
The 1969 failure was purely mechanical: things loaded in order stay roughly in order unless something works hard to break that up. The following year's draft lottery used two separate randomisations - one sequence for the dates, another for the numbers assigned to them - precisely so that no single mixing step had to carry the whole burden.
Publish enough to be checked
Nobody needed inside access to catch the draft lottery. The full results were public, so statisticians could plot the monthly averages and watch them slide downward across the year. An allocation lottery that releases its complete outcome invites exactly this kind of audit - and that is a feature, not a risk.
Software fails differently, not less
The green card case is the modern mirror of the capsule jar. Nothing was tumbling badly in a container; a piece of code simply weighted the pool towards early entries. Physical draws fail visibly and computational ones fail invisibly, which is why the serious operators lean on certified hardware randomness rather than a formula somebody wrote once.
Five Questions for Any Draw That Allocates
- 1Ask what the randomness is made of. Physical mixing, certified hardware noise and a plain software shuffle are not equivalent, and the first two are far harder to get quietly wrong.
- 2Look for a published, complete result. Partial announcements cannot be audited; a full dataset can be checked by anyone with a spreadsheet, which is how the 1969 flaw surfaced.
- 3Check whether entry timing or entry volume can matter. In a properly built draw, applying on the first day or the last should make no difference whatsoever.
- 4Treat a re-run as a serious cost, not a fix. Voiding a draw after publishing winners inflicts real harm - people resign jobs and make plans - so the integrity work belongs before the announcement.
- 5Remember who is entered without choosing. In a money lottery you opt in; in an allocation lottery you may be entered by birth, address or need - which is exactly why the standard has to be higher.
Why This Matters to Lottery Players
The detour through green cards and conscription leads straight back to the ordinary Saturday draw. The elaborate controls there - weighed balls, randomly chosen machines, auditors in the room, filmed test runs - look excessive for a game about money. They are the inheritance of cases like 1969, which showed how easily a draw can go wrong with nobody cheating at all.
Put differently: the care that goes into a lottery draw today was not invented for jackpots. It was worked out in places where a badly mixed container decided who had to go first.
Back to the Draws You Actually Enter
With the big lotteries you opt in yourself - and you can check the record. Run a fixed set of numbers through years of real draws, across every major lottery we cover, and see what would genuinely have come back.
Disclaimer: This article summarizes publicly available information about allocation lotteries, including the 1969 US draft lottery and the voided 2011 Diversity Visa draw. It is not legal or immigration advice; procedures and quotas change - always check the responsible authority's official information. Lottery remains a game of chance. Play responsibly. LottoROI is not affiliated with any lottery operator or government agency.