A working paper from the Federal Reserve Bank of Philadelphia, published this month, shows that small Bitcoin holders adjust their positions within 15 minutes of a major transfer alert, while Ethereum displays almost no reaction to the same signal.
🔑 Key takeaways
- The study covered 6,645 BTC whale transactions and 5,075 ETH whale transactions through end-2025
- After a whale buy, medium-sized wallet participation jumped 23.72 percentage points in under 15 minutes
- The same-direction effect on Ethereum barely reached the 10% statistical significance threshold
- Bitcoin realized volatility spiked briefly, then reversed within 24 hours of the alert
1. Methodology and the whale wallet definition
A USD 50 million threshold
To isolate meaningful flows, the authors defined a whale wallet as any address that had executed at least one transfer larger than USD 50 million, excluding addresses tied to centralized exchanges and smart contracts. This convention strips out technical flows and exchange market-making rebalancing.
Dataset and clean events
The study matched the exact timestamps of public alerts issued by Whale Alert against on-chain (blockchain-level) activity on Bitcoin, Ethereum and Wrapped Bitcoin (WBTC). To avoid statistical confusion, the authors kept only events with no other whale transaction in the two-hour window before or after, leaving 6,645 BTC transactions and 5,075 ETH transactions across the full sample through end-2025.
2. Bitcoin: a 15-minute flash reaction
The peak response on Bitcoin clusters in the first 15 minutes after the alert. Beyond that window, directional activity fades back toward pre-alert levels within an hour, consistent with a short-lived herding effect.
Buys synchronized with whale buys
After a whale buy alert, small-wallet buying participation rose by 14.81 percentage points, medium wallets by 23.72 points, and large non-whale wallets by 3.50 points. Mid-sized holders are therefore the most reactive cohort.
Sells aligned with whale sells
The symmetric sell-side reaction is just as clear, and arguably stronger. After a whale sell alert, small-wallet selling participation rose by 12.95 points, medium wallets by 29.52 points, and large wallets by 2.95 points. Medium sellers outpace their buyer counterparts, suggesting heightened sensitivity to bearish signals.
| Wallet cohort | Buy after BTC alert | Sell after BTC alert |
|---|---|---|
| Small wallets | +14.81 pp | +12.95 pp |
| Medium wallets | +23.72 pp | +29.52 pp |
| Large non-whales | +3.50 pp | +2.95 pp |
3. Ethereum: a largely ignored signal
The clearest ETH result is still statistically weak
On Ethereum, same-direction participation around the alerts stayed broadly flat. The strongest reading came from the largest non-whale cohort after whale sells, while medium ETH sellers only barely crossed the 10% significance threshold, the lowest cut-off used in the paper.
A structural gap, not a consensus effect
The authors link the divergence to market structure. On Ethereum, a large share of activity flows through exchanges, smart contracts (self-executing programs on the blockchain) and Layer-2 solutions (secondary networks that process transactions off the main chain to cut costs and congestion), where many user transactions are aggregated into larger balance transfers. That aggregation dilutes the individual response visible on Bitcoin.
« The contrast between the two networks persists after Ethereum’s move to proof-of-stake in September 2022, which argues against an explanation based solely on the consensus mechanism. »
Authors of the working paper, Federal Reserve Bank of Philadelphia
4. Volatility and study limitations
A short-lived spike, then a reversal
Whale alerts were associated with a temporary spike in Bitcoin realized volatility (volatility computed from observed past returns) at short horizons. After 24 hours, the effect on BTC volatility had reversed for both BTC and ETH alerts. The response to WBTC alerts was not statistically distinguishable from zero. Ethereum realized volatility, by contrast, tended to drift lower after the alerts without a meaningful initial jump.
An observational study
The authors stress that the evidence is observational. Wallet-size buckets are approximations based on transactions, and a single owner may control several addresses. Exchange activity has been filtered out. The event-study design establishes correlations between public alerts and on-chain activity, but does not prove a strict causal link between each alert and each observed reaction.
In the same window, Bitcoin gained 0.46% over the prior 24 hours and remains ranked number one by market capitalization, although that price move cannot be directly attributed to the alerts analyzed in the paper.
Conclusion
The Philadelphia Fed working paper highlights a clear behavioral gap between Bitcoin and Ethereum holders when public whale alerts hit the wire. Bitcoin’s spot market shows a flash herding response in the first quarter-hour, with directional shifts of 12.95 to 29.52 percentage points, while Ethereum absorbs the same events with a delay and dilution that reflects the complexity of its on-chain ecosystem and aggregation layers.
Even though the methodology is limited to observational correlations, the implications for market-making (continuous liquidity provision by specialized desks) and market surveillance are concrete: whale alerts act as a much stronger short-term signal on Bitcoin than on Ethereum. For investors and trading desks, this is a reminder that visible liquidity and network structure materially change how information travels and turns into actual on-chain transactions.
Sources
- CryptoSlate — Philadelphia Fed finds Bitcoin traders follow whale signals faster than Ethereum users
- Mitrade — Bitcoin and Ethereum whale alert analysis (May 2026)
- CoinDesk — Bitcoin wallet activity coverage (August 2026)
- Yahoo Finance — Crypto markets coverage
This article is published for informational and educational purposes only. It does not constitute investment advice. Do your own research (DYOR) before making any decision.
