When demand is two or three times supply, a first-come drop is a race, and bots are built for races. A raffle is a lottery, and bots are built for volume instead. Neither system is fair in the abstract. The question is which kind of unfair your customers will accept, and which kind your queue can actually enforce.
Why first-come breaks when demand runs hot
First-come, first-served works when demand is close to supply. Everyone who wants the product gets it, and the order barely matters. At two to three times supply, the system stops being a sale and becomes a race, and races are won by whoever is fastest. Bots are built for exactly that contest: they watch the clock, hit the endpoint the millisecond it opens, and check out before a human has read the page.
The result is a drop where the winners are mostly automated and the losers are mostly human, and the seller cannot tell the difference from the sales numbers alone. The inventory sells out either way. The difference is who bought it, and whether they will ever come back.
How a raffle changes the bot math
A raffle changes what the bot optimizes for. Speed stops mattering, because the entry window stays open for hours or days. Volume takes over: a raffle bot submits as many entries as it can, from as many identities as it can create. The defense changes with it. Entry limits, identity checks, and one entry per person become the whole game, and a raffle without those controls is a first-come drop with extra steps.
That is why the entry window matters. A long window gives real customers time to enter, and gives the seller time to dedupe entries and flag the farms before the draw.
The SNKRS model, in practice
Nike's SNKRS app runs the best-known version of this system. The LEO queue holds entrants for two minutes when demand spikes, and the DAN draw runs a ten-minute entry window before selecting winners at random. The two pieces do different jobs: the queue absorbs the flood, and the draw removes timing from the outcome.
The model works because the rules are public and the window is fixed. Customers know when the draw opens, when it closes, and that their chance is the same whether they entered in the first minute or the last. That predictability is what keeps people entering the next draw after losing the last one.
What to tell the people who lose
Most entrants lose. In a raffle for a scarce sneaker, losing is the normal outcome, and the seller's job is to make losing feel fair. Publish the odds, the entry count, and the selection method. Say how many entries were removed as duplicates. Queue-it found that 76 percent of customers were satisfied with a queue when it was explained to them, and the same logic applies here: people accept losing to a lottery they understand, and they resent losing to one they suspect is rigged.
Keep reading
Related guides on this site: how digital dashes hide rolled-back odometers a first-timer's guide to salvage auctions how salvage titles get washed across state lines More from Bot Protection For Auto Marketplaces.
Frequently asked questions
Is a raffle actually fairer than first-come?
Neither is fair in the abstract. First-come rewards speed, and bots are faster than people. A raffle rewards nobody in particular, and bots win by volume instead. The fairer system is the one whose failure mode your customers accept.
Can bots still win a raffle?
Yes, by volume. A raffle bot submits many entries, so the defense is entry limits, identity checks, and one entry per person. A raffle without those controls is a first-come drop with extra steps.
Why do some drops use a queue and some use a raffle?
Because they solve different problems. A queue handles a flood of simultaneous demand for a live sale. A raffle handles a fixed supply of a scarce item where timing should not matter. The product and the crowd decide, not the preference of the seller.
What should a drop page tell people who did not get a pair?
The odds, the entry count, and the selection method. Queue-it found 76 percent of customers were satisfied with a queue when it was explained to them, and the same logic applies to a raffle: people accept losing to a lottery they understand.