Airdrop farming: what actually qualifies you
The clearest evidence that nobody knows the criteria in advance is when they get published. In almost every distribution I have followed, the eligibility rules and the claim page go live in the same announcement, on the same day. The team writes the rules, then opens the door immediately.
There is no window during which the criteria exist and the behaviour they measure has not already happened.
That ordering is the whole subject. A checklist sold to you in July, for a protocol that will decide its formula in November, is a forecast wearing the clothes of a specification.
Rules get written against a dataset
When a team sits down to distribute, the entire history is in front of them. Not a sample. Every address that ever touched the contracts, with timestamps, amounts, funding paths and exit behaviour attached.
Their job is to draw a line through that table so that most of the users they want to keep land on one side and most of the wallet farms land on the other. That is an engineering problem shaped by the specific mess in front of them.
Which is why the outputs vary so much. A protocol overrun by one operator running fifteen thousand addresses ends up with a harsher formula than a quiet protocol with four hundred genuine users.
Neither team is applying a shared standard, because there is no shared standard to apply.
It also explains appeals windows. Rules written in a fortnight against a messy table will catch honest wallets, so teams open a form and reverse a handful of decisions. That process is routine, and it is the best evidence available that the criteria were never sitting in a drawer waiting.
What published formulas have actually weighted
Across the distributions that eventually showed their working, a few factors keep reappearing. This is tendency, not a rule you can bank on, and none of it predicts whether a given protocol will distribute anything or what a token would be worth if it did.
Duration comes up most. Wallets spread across many months tend to score better than wallets with the same action count compressed into one week. A week of activity is something a person can buy in an afternoon. Having been present in March is not purchasable in November, at any price, which is exactly why it carries weight.
Value at risk comes up nearly as often. Formulas frequently weight position size and how long positions stayed open, and several have set a floor below which small transfers contribute nothing at all. A hundred dust transactions can score lower than one position that sat open through a volatile month.
Depth inside a single product shows up too. Bridging in and stopping there produces a thin profile. Supplying collateral, borrowing against it, unwinding, then voting on a parameter change produces a wallet that touched five surfaces of the protocol.
Anyone reading that table can tell the difference by eye. No statistics required.
One more property is worth knowing about, because it breaks checklists entirely. Several formulas have been relative rather than absolute: tiers set by percentile, or a fixed pool split proportionally across everyone who cleared the bar. Under that design your score depends on what every other wallet did. A threshold that looked comfortable in September can sit below the cut by the time the snapshot lands, without you changing a thing.
Nobody can hand you a number in advance for a formula like that. The number is a function of a crowd that has not finished forming.
Persistence after the fact is the one people forget. Some formulas carry a term for what a wallet did with a previous round from the same team, and dumping everything inside seven days is a row that stays in the record forever. What happened after one distribution feeds the design of the next.
Why honest wallets get filtered
This side of the topic usually gets written up as an evasion guide, which is a waste of the material. The useful framing for someone running two or three wallets for real reasons is defensive: know which signals are cheap to compute, so you recognise when your own setup produces one by accident.
Timing is the cheapest signal there is. A set of addresses that all transact inside a narrow, repeating window describes a scheduler, and a scheduler describes a single operator. People are irregular. They act at strange hours, skip weeks, forget entirely. The absence of that irregularity is trivial to spot.
Funding topology is next. Addresses funded from one exchange withdrawal, or from a single hot wallet that paid out to all of them in one session, form a star in the transaction graph with an obvious hub. Finding that shape is a graph query somebody writes in an afternoon.
Sequence identity is third. Identical contracts, in identical order, with round amounts, repeated across addresses. One wallet doing that is a habit. Forty wallets doing it is a script with forty runs.
The asymmetry in the incentives is worth sitting with. A team that filters too loosely hands supply to a farm and gets criticised publicly for it. A team that filters too tightly inconveniences a few hundred genuine users, most of whom never file an appeal. Given that trade, the pressure runs toward filtering hard and sorting out the mistakes afterwards, which is roughly what has happened.
The uncomfortable implication for anyone running several wallets legitimately is that convincing difference costs money. Separate funding paths, genuinely varied timing, genuinely varied behaviour, sustained over months. That is more fees and far more attention than most setups are willing to spend, so they buy the cheap imitation instead, and the cheap imitation is precisely what the filters were built to catch.
The three costs nobody puts in the sheet
Gas is the obvious one and still gets undercounted. An approval plus a swap on a congested layer one can run a few dollars. On a rollup it might be fractions of a cent.
Neither number feels like anything on its own. That is the trap.
Multiply by months, by protocols, by wallets, and it becomes a bill you have already settled regardless of what happens next. The fee is certain and paid today. The reward is contingent and arrives, if ever, an unknown number of months later.
Locked capital is the invisible one. Two thousand dollars sitting in a lending position for five months is two thousand dollars unavailable for five months. That carries a cost even when the position closes at the exact price it opened, and I have almost never seen anyone account for it.
Time is the one people refuse to price. An hour a week for eight months is thirty-two hours. Value your own hour at whatever you consider fair; leaving it at zero is how a losing activity keeps feeling like a winning one.
Add all three honestly and most farming does not clear its own costs. The distributions people quote at each other are memorable because they were outliers. Nobody writes a thread about the eleven protocols that never launched anything.
What I got wrong
I ran the volume theory for about a year. The reasoning was that qualification is a numbers game, so more addresses and more actions meant more chances at the same prize. I spent somewhere around six hundred dollars in fees disproving it.
Two allocations came out of that stretch. One was smaller than what I had already spent on that single protocol.
The other I sold badly, which is on me and not the protocol.
The wallet that actually paid was one I had been using for months with my own money, for reasons that had nothing to do with farming. No schedule, no fleet, no strategy attached to it at all. It qualified for several times what the entire operation produced.
I also mis-sized a position once because I had read a leaked formula from a competing protocol in the same category and assumed the thresholds would carry across. They did not. Categories rhyme. They do not copy.
Fewer protocols, held longer
The practical version of all this is boring, which is why nobody sells it. Use fewer protocols. Stay in them longer. Only expose capital you were already willing to expose. Stop paying anyone for criteria that have not been written yet.
That approach guarantees nothing and I am not going to imply otherwise. It is not financial advice, it says nothing about what any token will be worth, and it cannot make any protocol safe. What it does is stop you spending a year optimising against a rulebook that does not exist.
If you want the tested side of it, the tooling, the wallet hygiene, and the trackers worth checking daily, start at the home page.
Get new guides and videos first — join the Telegram channel.