On-chain investigationAirdropsBNB Chain, Ethereum, BaseJan. to Sept. 2026

Where airdrop tokens really go: some of 2026's biggest 'claims' were the project

On day two of Bitway's airdrop, a batch of new wallets appeared to land the biggest payouts of the launch. They were not claimers. We followed them, and the pots of 2026's other big airdrops, to see where airdrop tokens really go.

At a glance
An airdrop pot is the stack of tokens a project sets aside for its users. Across 35 of 2026's big drops paid from a simple claim contract, the median pot sent 87% to claimers. The rest sat unclaimed, went back to the project, or left in a few big transfers. Espresso took back 85% of its pot in June. Some of the biggest wallets in an airdrop's history were not claimers at all. At Bitway, 44 new wallets got 196 million BTW in one transfer, more than every claimer combined. Within a week all of it sat at two exchange deposit addresses that Bitway's main token wallet also used.
87%
Median share of a pot that reached claimers (35 drops)
85%
Of Espresso's pot taken back in June
196M BTW
Sent to 44 wallets in one transfer, then to two exchanges
4 pushes
Of one KITE wallet list, twice as a 'rehearsal' token
Key findingsEach line has its own share link
  1. 87%

    The median share of an airdrop pot that reached claimers, across 35 drops paid from a simple claim contract. 14 paid out almost all of it.

  2. 9 of 35

    Pots with a tenth or more still unclaimed at the end of September, led by TermMax and XYZVERSE. Three of them pay in stages.

  3. 85%

    Of Espresso's pot went back to the wallet that funded it, on June 23. Claimers had taken the other 15%.

  4. 196M BTW

    Sent to 44 new wallets in one transfer the day after Bitway's claim opened. All of it reached a Bitget or HTX deposit address within a week.

  5. 4 pushes

    Of the same 97,756-wallet KITE list in two weeks, twice as a token named Distribution Rehearsal. We did not count it as an airdrop.

01 · One transferMore than every claimer combined

Bitway, a project building Bitcoin-based yield products, launched its BTW token on BNB Chain on March 2. Part of the launch was an airdrop run with Binance Wallet, and 65,850 claimers took part. The typical one got 1,000 BTW, worth about $8 at the launch price.

The next morning, one transaction sent BTW to 44 wallets. Each got a block of three to six million tokens, 196 million in all. That is about one and a half times what every claimer received together.

One transfer, more than every claimBitway (BTW)
44 wallets, one transfer65,850 claimersMarch 3, 10:42every claim, from March 2196.3M132.8MBTW received. The 44 wallets each got one block of three to six million tokens in a single transaction.
The 44 wallets got about one and a half times what all claimers received.

Within six days, every one of those tokens was at an exchange. Thirty-nine of them sent their whole block to the same Bitget deposit address, an address Bitget gives to one customer account. The other five used one HTX deposit address. And the 196 million had come, eight minutes before the batch, from the wallet that received most of Bitway's supply.

To anyone scanning the chain, those 44 look like the airdrop's biggest winners cashing out. Our own study first counted them as airdrop payouts. They were tokens from the project's main wallet, on their way to exchanges.

That raised a wider question. When a project sets tokens aside for an airdrop, where do they actually go? We checked this year's big EVM airdrops to find out.

02 · The potFour ways out of a claim contract

Most airdrops work the same way. The project loads its pot, the tokens set aside for users, into a claim contract. Eligible users then call the contract and collect their share. Tokens can leave that contract in four ways: to claimers, in big transfers the project sends itself, back to the project, or not at all.

We looked at the 35 drops in our 2026 airdrop scorecard that paid from a simple claim contract, one funded by the project and used only for the airdrop. For most of them the answer is simple. The pot went to claimers.

Where 35 airdrop pots wentShare of each pot
0%25%50%75%100%Beat (BNB)Paimon pPOLY (BNB)BigShort (BNB)OFC (Base)ZK Coin (ETH)Arcium (BNB)Perle (BNB)OpenGradient (BNB)Cap (BNB)Unitas (BNB)OFC (Base)LABUBU (BNB)Limitless (Base)ODIC (BNB)o1.exchange (Base)CREATOR (BNB)Re Protocol (ETH)Chip (Arb)Sport.fun (Base)CAT (BNB)Avantis (Base)ENA (ETH)edgeX (ETH)MarsCat (BNB)ROBO (ETH)Staked ENA (ETH)EarthMeta (ETH)Elsa (Base)DAPPOS (ETH)Brevis (ETH)XYZVERSE (BNB)TermMax (BNB)Aligned (Base)Espresso (ETH)Zenchain (BNB)ClaimersBig payoutsBack to projectOther payoutsStill in contractShare of each pot, 35 drops paid from a simple claim contract, most paid out first. Other payouts: to wallets that had also paid in.
Most pots went almost entirely to claimers; the exceptions stand out.

The median pot sent 87% of its tokens to claimers, and 14 paid out almost everything. The exceptions are where the story gets interesting. Some pots were still close to half full in September. A few were taken back. And two, Aligned on Base and EarthMeta, sent a big share out in a handful of large blocks. Most of those tokens still sit in the wallets that received them.

03 · LeftoversTokens nobody has claimed

In nine of the 35 drops, a tenth or more of the pot was still sitting in the contract at the end of September. TermMax, XYZVERSE, Brevis and DAPPOS each still held more than two fifths.

Pots still waitingUnclaimed at the end of September
TermMax (BNB)XYZVERSE (BNB)Brevis (ETH)DAPPOS (ETH)Staked ENA (ETH)ENA (ETH)MarsCat (BNB)Avantis (Base)CAT (BNB)56%47% (pays in stages)45%43% (pays in stages)25%23%22%20% (pays in stages)19%Unclaimed so farPays in stages: part not due yetShare of the pot still in the claim contract on September 29, drops with a tenth or more left.
Some leftovers are stages not yet due; the rest is tokens nobody has collected.

Not all of that is forgotten. XYZVERSE, DAPPOS and Avantis pay in stages, so part of what remains is simply not due yet. The rest is tokens that nobody has collected. Some never will be. What happens to them is up to the project, and at six drops part of the answer is already on the chain.

04 · Taking it backSix pots sent tokens home

Espresso opened its claim on Ethereum in mid-February. Claims stopped about two weeks later, by which time 52,621 claimers had collected 15% of the pot. On June 23, the other 271 million ESP went back to the wallet that had funded the contract.

Four pots that sent a big share backClaimed against taken back
EspressoElsaROBOedgeXback on June 23back on Feb. 26back on Aug. 5back on April 1915% claimed66% pushed74% claimed76% claimed85% back34% back26% back24% backShare of each pot. Espresso and Elsa returned tokens to the wallet that funded the contract; ROBO to another address; edgeX to its exchange vault.
At Espresso, most of the pot went back to the project.

Three other drops took back a large share. Elsa pushed its tokens straight into about a quarter of a million wallets. The day after the last push, the third of the pot it had not sent went back to its funding wallet. In August, the ROBO claim contract sent the last quarter of its pot to an address that has not moved it since. That drop is the subject of our ROBO airdrop investigation. In April, edgeX moved the last quarter of its EDGE pot into the edgeX exchange's own vault. Limitless and OFC sent back a few percent each.

Zenchain is a separate case. It moved two thirds of its pot to its own treasury on the day claims opened, which is why most of its bar in the pots chart is pink.

A take-back is not a scandal in itself. Claim windows close, and projects usually announce the deadline. It does mean the size of an airdrop, as announced, can be far larger than what users end up holding.

05 · Look-alikesBig payouts that were not claims

The Bitway batch was not the only large transfer that looks like an airdrop payout on the chain. We traced the biggest outsized payouts in the 51 drops, the ones more than a thousand times the typical claim. Several turned out to be something else.

AirdropLooked likeWhat it was
Bitway196M BTW to 44 wallets on BNB ChainThe project's main wallet, via a middle wallet; all of it to one Bitget and one HTX deposit address
Aligned1.86B ALIGN on Ethereum in six payouts1.66B to one wallet in four transfers within 10 minutes, barely moved since; 150M to Protocol Guild, which funds Ethereum core developers
Spacecoin1.55B SPACE on BNB Chain in eleven payoutsNew tokens created straight into two wallets; about a third of what the larger one sent on went into the Wormhole bridge between chains
Zenchain574M ZTC on BNB Chain in seven payouts420M to two wallets that never moved it; 105M into PancakeSwap
ETHGasEleven payouts on Ethereum, ten of exactly 19,999,900 GWEIEight passed on in halves and quarters, some to exchange deposit addresses

None of this is hidden. It shows how easily a raw count of "airdrop recipients" can mislead. A single wallet that took a billion tokens drags every average with it, whatever it turns out to be. We keep these payouts out of every claimer figure in our airdrop work. Bitway's trail was the clearest, so we followed it to the end.

06 · BitwayFollowing the 196 million

The trail starts in December, when all ten billion BTW were created in one wallet. On January 21 that wallet passed 95% of the supply to a second one, which we call Bitway's main token wallet, 0x7d45…9256. In late February the main wallet funded the airdrop's claim contract. It also put tokens into PancakeSwap, the main DEX on BNB Chain, days before trading opened.

How 196 million BTW reached two exchangesBitway, January to April
Main token walletMiddle wallet44 new walletsBitget depositHTX depositClaim contract9.5B BTW, Jan. 2195% of supplyMarch 3196M in one txone addressone addressand PancakeSwap50M claims, 111M200M, 10:34PancakeSwap, Feb. 26BTW flows, 2026 (times UTC). The main wallet used both deposit addresses before the 44 wallets did.120M, March 1-310M, March 210:425 wallets, 23.6M39 wallets172.8M, March 8-9Bitget sends 255M back to the main wallet, March 9 to April 1
The same two exchange accounts sit at both ends of the trail.

The main wallet used the two exchange accounts first. In the first days of March it sent 120 million BTW to the Bitget deposit address 0x9ee9…6eed and 10 million to the HTX one. On the morning of March 3 it sent 200 million BTW to a middle wallet. Eight minutes later the middle wallet pushed the 44 blocks out in one batch, through a bulk-transfer contract that has moved more than 200 different tokens.

The 44 wallets were new. Forty-two of them had received exactly one transfer before, each a withdrawal from an exchange (Bybit, Bitget, Gate, HTX or OKX) in the five weeks before the launch. The block sizes look typed by hand: 3,456,789, 4,567,890 and 5,678,901 BTW, with most of the others built on runs of digits like these.

For almost five days their BTW sat still. Then, on March 8 and 9, they emptied into the two deposit addresses. At that week's prices their BTW was worth about $4.6 million. Bitget then sent 255 million BTW back to the main wallet between March 9 and April 1. From April to June, the main wallet sent BTW to, or got BTW from, 15 of the 44 wallets.

What happened inside the exchange accounts is not on the chain. Bitget ran a BTW listing promotion that week with 33 million BTW in rewards, a small part of these deposits. The tokens could have been sold, lent to a market maker, or held. The chain shows that the biggest "airdrop" wallets were filled from the project's main wallet, and that every token went to two exchanges. Anyone can follow this trail with the Bitquery MCP.

07 · The rehearsalA push that was not an airdrop

One more drop nearly made our 2026 list. On August 24, a KITE contract on Ethereum pushed tokens to 97,756 wallets. KITE itself has traded since late 2025. At claim-day prices it would have been the year's third-largest airdrop.

We left it out. Almost all of them had received the same amounts three times in the two weeks before, from three other token contracts. Two of those tokens were named "Distribution Rehearsal".

One list, four pushesKITE on Ethereum, August
0k50k100k100,466 wallets100,595 wallets97,757 wallets97,756 walletsAug. 11Aug. 12Aug. 23Aug. 24KITERehearsalRehearsalKITEToken named KITEToken named Distribution Rehearsal
Nearly the same wallets got the same amounts each time.

A list replayed from a practice run looks like a copy of an existing holder list, and it rewards nobody new. That also explains the shape. Just 82 wallets took 92% of the tokens, while the typical one got about 15 KITE. Four in five recipients were wallets in active use, yet only about one in 18 has moved the token since.

08 · Reading an airdropWhat to check before you trust the numbers

For a claimer, the pot is the headline and the claim window is the fine print. If you are eligible, claim early. Closing dates are real, and six of 35 pots in our sample sent tokens back to the project.

For anyone reading airdrop data, a big wallet in the transfer history is where the questions start. Check who sent the tokens and where they went next. A wave of "airdrop whales" heading to exchanges can be the project moving its own supply. At Bitway it was.

09 · The answerHow we checked, and what it means

We used Bitquery's archive of token transfers from January 1 to September 29, 2026, with exchange and DEX addresses from our address labels and prices from our price data. A pot is what the project loaded into the claim contract. For each contract we added up what went to claimers, to outsized payouts, back to the project, and what is still there. Payouts made through smart-wallet services count as claims.

So where do airdrop tokens go? Mostly to the people who claim them. The median pot in our sample did what it promised. But the biggest numbers in an airdrop's history can belong to someone else. At Espresso most of the pot went home. At Bitway, the largest "payouts" of the launch were the project's own tokens, sent through 44 new wallets to two exchanges.

FAQ

Where do unclaimed airdrop tokens go?

They stay in the claim contract until the project moves them. In our sample, nine of 35 pots still held a tenth or more at the end of September. Six sent tokens back to project wallets, led by Espresso, Elsa, ROBO and edgeX.

What is an airdrop clawback?

A clawback is when a project takes back tokens it had set aside for an airdrop, usually after the claim window closes. Espresso took back 85% of its pot on June 23, 2026.

Were Bitway's 44 big wallets airdrop winners?

No. They got their BTW in one batch transfer from a wallet funded by Bitway's main token wallet. None of it came from the claim contract. Within a week all 196 million tokens went to one Bitget and one HTX deposit address that the main wallet also used.

Was KITE's August 2026 push an airdrop?

We do not think so. The same list of wallets and amounts was pushed four times in two weeks, twice as a token named Distribution Rehearsal. That looks like a copy of an existing holder list.

Check it yourself

Follow any airdrop pot

Every claim contract, batch transfer and exchange deposit here is public. You can pull a token's transfers, see who funded a wallet and follow tokens onward.

Token transfersAddress labelsWallet fundingToken prices
Open the Bitquery MCP →Transfers up to September 29, 2026
Scope, limits and attribution

This article is provided for informational and educational purposes only and reflects analysis of publicly available on-chain data as of the times stated. It does not constitute legal, financial, compliance, or investment advice.

Wallets are identified by address only. 'Bitway's main token wallet' means the wallet that received 95% of BTW's supply on January 21, 2026; nothing here states which person or company controls it. What happened to tokens inside exchange accounts is not visible on-chain, and nothing here says they were sold.

References to Bitway, Espresso, Elsa, ROBO, Aligned, Spacecoin, Zenchain, ETHGas, KITE, Bitget, HTX and any other project or exchange describe on-chain activity or public statements. No wrongdoing is alleged.

Nothing herein should be relied upon as a definitive determination of fact. Readers should conduct their own independent verification before taking any action. The authors and publisher accept no liability for any loss or damage arising from the use of this article.

Reported by Gaurav Agarwal for Bitquery Research, with AI tools; every figure was checked against the raw data.

See where any token really went

Bitquery shows every transfer, holder and trade for a token, so you can check an airdrop before you trust its numbers.