For every dollar of UNI that tracked whales pulled off exchanges between September 11 and October 10, 2026, about 73 cents of UNI went back in.
That ratio is the whole story of Uniswap's month, and it disappears the moment someone quotes a net figure. Net, $44.45M of UNI left exchanges, which sounds decisive. Gross, $162.32M left and $117.86M arrived. The net is the small difference between two very large flows.
Aave's month looked different. Smaller, at $68.80M of whale dollars against UNI's $280.32M, but more one-sided: $43.41M withdrawn and $25.29M deposited.
This comparison reads both tokens in both directions, then looks at what built each total: move sizes, the wallets behind them and the days they happened on.
The side-by-side numbers
| Metric | UNI | AAVE |
|---|---|---|
| Whale $ moved (30D) | $280.32M | $68.80M |
| Withdrawn from exchanges | $162.32M | $43.41M |
| Deposited to exchanges | $117.86M | $25.29M |
| Net leaving exchanges | $44.45M | $18.12M |
| Withdrawal share | 57.9% | 63.2% |
| Wallets | 150 | 96 |
| Whale moves | 1,094 | 482 |
| DEX share of whale $ | 0.0% | 0.1% |
| Whale $ as % of market cap | 5.78% | 2.62% |
UNI led on every measure of activity: about 4.1 times AAVE's dollars, 150 wallets against 96, and 1,094 moves against 482. UNI's market cap was only about 1.85 times AAVE's ($4.85B against $2.62B on October 11), so the gap holds after scaling. Whale moves equaled 5.78% of UNI's market cap over the month and 2.62% of AAVE's.
The DEX row sets the vocabulary. Swaps were 0.0% of UNI whale dollars and 0.1% of AAVE's. For reference, UNI's DEX portion was $21.8K of DEX swap buys and $114.8K of DEX swap sells; AAVE's was $51.7K and $48.2K. Everything else was coins moving onto and off exchanges, so "withdrawn" and "deposited" are the accurate words, not purchase or sale language.
The direction convention: coins leaving exchanges are the accumulation-side read, coins arriving are the distribution-side read. By that reading both tokens leaned toward accumulation. They just leaned with very different force.
How to read a withdrawal share like 57.9%
A withdrawal share is withdrawals divided by withdrawals plus deposits. It runs from 0% (all deposits) to 100% (all withdrawals), with 50% meaning the two sides matched.
UNI's 57.9% is the closer of the two to balanced. It says that a little more than half of UNI's exchange-crossing dollars were heading out. That is a lean, not a stampede, and on $280.32M of gross flow a lean can still produce a large net number.
Three habits keep this ratio honest:
- Always print the gross next to the net. "$44.45M net leaving, on $162.32M out and $117.86M in" tells a reader the flow was two-way. "$44.45M net leaving" alone does not.
- Compare the ratio, not just the net, across tokens. AAVE's net was less than half of UNI's, but its 63.2% withdrawal share was the more one-sided of the two.
- Check the ratio by week. A monthly 57.9% can hide a week that leaned the other way, which UNI had (more on that below).
Where the dollars came from: move sizes
| Move size | UNI moves | UNI share of $ | AAVE moves | AAVE share of $ |
|---|---|---|---|---|
| Under $10K | 221 | 0.4% | 160 | 0.9% |
| $10K–$100K | 482 | 5.7% | 220 | 15.5% |
| $100K–$1M | 324 | 41.3% | 86 | 44.8% |
| $1M and up | 67 | 52.6% | 16 | 38.8% |
UNI's month was built on large moves. Sixty-seven moves of $1M or more, averaging about $2.20M each, carried 52.6% of its whale dollars. Add the $100K–$1M bucket and about 94% of UNI's dollars came from moves of $100K or more.
AAVE leaned more on the middle. Its $100K–$1M bucket carried more dollars (44.8%) than its $1M-and-up bucket (38.8%), and moves under $100K made up 16.4% of its dollars against about 6% for UNI.
Read this way, AAVE whale flow was a wider spread of mid-sized transfers. UNI's was dominated by large ones, which is exactly the condition under which a few moves can swing a net figure.
The wallet that moved a third of UNI's month
This is the number that changes the comparison most.
| Wallet concentration | UNI | AAVE |
|---|---|---|
| Wallets active | 150 | 96 |
| Largest wallet's share of $ | 35.2% | 27.5% |
| Top 5 wallets' share | 60.0% | 54.6% |
| Top 10 wallets' share | 72.2% | 70.3% |
| Wallets net withdrawing | 112 | 75 |
| Wallets net depositing | 29 | 16 |
One wallet moved $98.63M of UNI in the window, 35.2% of every tracked UNI whale dollar. That is not visible from the largest single moves, which were modest: UNI's three biggest individual transfers were 7.8% of its month. The largest wallet got to a third of the total through many moves, not one.
This is a distinction worth keeping. Concentration by move asks whether one transfer made the month. Concentration by wallet asks whether one holder did. UNI scores low on the first and high on the second. A dashboard that checks only the largest transfers would call UNI's month broad; a wallet-level view shows one address doing a third of it.
AAVE's largest wallet moved $18.95M, 27.5% of its month, also well above the 11.1% its three largest single moves accounted for.
The headcount gives a third angle. In UNI, 112 wallets finished the window net withdrawing and 29 net depositing; in AAVE, 75 and 16. By headcount, roughly three-quarters of both tokens' whale wallets leaned toward the accumulation side, a firmer lean than either token's dollar withdrawal share. The gap suggests the deposit side in both tokens came from fewer wallets moving more per wallet.
The days that built UNI's month
UNI was active every day of the window, with a median day of about $9.19M. The busiest days were spread out rather than stacked:
- October 1: $29.96M moved, 10.7% of the month, from 103 moves by 31 wallets, net $2.36M leaving exchanges.
- September 15: $22.23M moved by just 14 wallets, net $11.04M leaving.
- September 21: $20.04M moved, net $7.72M arriving at exchanges.
Then, right after the busiest day, UNI went nearly silent: $350.4K on October 3 and $400.5K on October 4.
By week, one stretch stands out. From September 20 to 26, UNI's withdrawal share was 49.4% and $1.14M net arrived at exchanges, the only week of the month that leaned toward deposits. It contained the two largest single UNI moves of the month, both deposits ($7.93M on September 23 and $7.87M on September 21). From September 28 to October 6, UNI then finished nine straight days net leaving exchanges.
The largest single moves ran against the lean
| Token | Date (UTC) | Type | Size | Share of 30D whale $ |
|---|---|---|---|---|
| UNI | Sep 23 | Deposit to exchange | $7.93M | 2.8% |
| UNI | Sep 21 | Deposit to exchange | $7.87M | 2.8% |
| UNI | Sep 24 | Withdrawal from exchange | $6.13M | 2.2% |
| AAVE | Oct 7 | Deposit to exchange | $2.95M | 4.3% |
| AAVE | Sep 30 | Deposit to exchange | $2.37M | 3.4% |
| AAVE | Sep 30 | Withdrawal from exchange | $2.29M | 3.3% |
In both tokens, the two largest single moves were deposits to exchanges, the distribution-side read, while both months finished net leaving exchanges. UNI's two largest deposits together came to $15.80M, 13.4% of all UNI deposits for the month.
The shortened wallet behind the $7.87M UNI deposit, 0x1d4896…0270, also made the largest PEPE deposit in Deep Blue Alpha's memecoin data on the same day, September 21. One address, one day, a DeFi governance token and a memecoin. Wallets do not stay inside sector boundaries, and sector-level summaries are partly the same wallets counted twice.
Shared wallets
Over the window, 39 tracked wallets moved both UNI and AAVE: 40.6% of AAVE's 96 whale wallets and 26.0% of UNI's 150. Across both tokens there were 207 distinct wallets, so roughly one in five was active in both.
That is consistent with DeFi governance tokens often sitting in the same large wallets, but the data shows only the overlap. A shared wallet can withdraw one token and deposit the other in the same week, so shared activity is not shared direction.
How the numbers were measured
Figures count only wallets on Deep Blue Alpha's tracked whale list, about 21,880 Ethereum addresses at this update. "Whale $ moved" is exchange withdrawals plus exchange deposits plus DEX swaps, valued at the time of each move. Exchange-flow rows are retained for 30 days, so September 11 to October 10, 2026 (UTC) is the full available window. Wallet concentration ranks wallets by total dollars moved in each token, in both directions combined; "net withdrawing" and "net depositing" compare each wallet's own withdrawals with its own deposits. A small number of wallets in each token, nine in UNI and five in AAVE, had no net exchange lean either way, which happens when a wallet's activity is DEX-only or its withdrawals and deposits offset. Market data is from CoinGecko at October 11, 2026, 18:30 UTC.
How to check a move yourself on Etherscan
- Open UNI or AAVE on Deep Blue Alpha, set the window, and note a large move's date and shortened wallet.
- Search the full wallet address on Etherscan and open Token Transfers (ERC-20). Filter by the token's contract address.
- On the matching date, check the counterparty. A labelled exchange wallet sending to the address is a withdrawal; the address sending to a labelled exchange wallet is a deposit.
- To test wallet concentration, scroll the same tab across the whole window and add up the wallet's transfers. A wallet with dozens of mid-sized moves can outweigh a single headline transfer.
- If a transaction shows a router or pool and two tokens changing hands, it is a DEX swap, the only case where purchase or sale wording applies.
What this comparison cannot tell you
It cannot identify wallet owners, and none are attributed. It cannot say why the largest UNI wallet moved so much: market making, treasury management and custody changes all produce heavy two-way exchange flow. It sees tracked wallets on Ethereum only. And it does not explain price. Over the 30 days to October 11, CoinGecko showed AAVE up 34.2% and UNI up 26.5%, with UNI 82.7% and AAVE 74.3% below their all-time highs. Flow and price moved independently over this window.
Takeaway
UNI was the far bigger whale story in this window, but its $44.45M net hides a heavily two-way month built on large moves, with one wallet accounting for 35.2% of the dollars. AAVE was smaller, more one-sided and spread across more mid-sized transfers. For any token with flow this two-way, the useful figures are gross in, gross out, the largest wallet's share and the week-by-week ratio, in that order. The net comes last.
The full comparison, with the latest-week split and the largest moves, is at UNI vs AAVE: Whale Flow Compared. The framework behind withdrawal share is laid out in Exchange Inflows and Outflows Explained, and both tokens sit alongside the rest of DeFi on the live DeFi whale board.
Data: Deep Blue Alpha tracked-wallet data, whale-scoped, September 11 – October 10, 2026 (UTC), Ethereum only. Market data from CoinGecko, October 11, 2026, 18:30 UTC.
Deep Blue Alpha is an Ethereum whale intelligence platform tracking 20,000+ whale wallets in real time. This article is for informational purposes only and does not constitute financial advice. NFA/DYOR.
Track whale activity for free at deepbluealpha.io
Top comments (0)