A paper from the Federal Reserve Bank of Philadelphia found that Bitcoin wallets were more likely to trade in the same direction as a whale within 15 minutes after a large whale transaction became public. The finding describes an association in Bitcoin whale-trading patterns and does not establish that whales caused the subsequent trades.

In crypto terminology, a whale is a wallet that holds or moves a very large amount of cryptocurrency. The research examines how ordinary Bitcoin wallets behaved after one of these large wallets made a transaction visible through a public alert.

What the Philadelphia Fed Paper Found

The Federal Reserve Bank of Philadelphia published Working Paper 26-42, titled “How Do Large, Sophisticated Cryptocurrency Trades Impact Broader DeFi Market Dynamics?,” in September 2026. The authors are Keith Hazen, Julapa Jagtiani, and Loretta J. Mester. The manuscript is dated August 28, 2026.

The paper’s central result is that, after a large Bitcoin whale transaction became public, other wallets were more likely to trade in the same direction during the first 15 minutes. The term “more likely” is important: the study does not say that every wallet followed the whale, or that the behavior was automatic or guaranteed.

The researchers used public “whale alerts,” automated notifications that identify very large crypto transfers. They compared wallet activity during the first 0 to 15 minutes after an alert with activity during the 15 minutes before the alert.

The paper defines a whale wallet as one that made at least one transfer worth more than $50 million during the sample period. Wallets belonging to exchanges or smart contracts—automated programs that operate on a blockchain—were excluded. This is a transaction-based classification, rather than a determination of who owns a wallet or how much it holds at present.

Non-whale wallets were divided into small, medium, and large categories according to transaction sizes. Large wallets were at or above the 95th percentile, while small wallets were below the 50th percentile.

The largest changes appeared among medium-sized non-whale Bitcoin wallets. Their estimated buy participation increased by 23.72 percentage points during the first 15 minutes after a whale alert. Participation rose by 14.81 percentage points for small wallets and 3.50 percentage points for large wallets. All three results were statistically significant at the 1 percent level.

Bitcoin medium-wallet buy participation: +23.72 percentage points

The selling results showed a similar pattern. Medium wallets recorded an estimated 29.52-percentage-point increase in sell participation during the first 15 minutes. The corresponding increases were 12.95 percentage points for small wallets and 2.95 percentage points for large wallets.

Bitcoin medium-wallet sell participation: +29.52 percentage points

These figures measure changes in the share of wallets participating, not price movements or trading profits. Same-direction Bitcoin activity peaked during the first 15 minutes and moved back toward normal levels within 60 minutes. The result resembles coordinated flows that can occur when whale wallets absorb large amounts of Bitcoin, but it does not demonstrate that the wallets were copying the whale.

Details Behind the 15-Minute Window

The event window begins with a public whale alert, which is an automated flag of a very large transfer. The specific thresholds and methods remain important when evaluating individual estimates.

The alert data covered December 14, 2017, through December 31, 2025. Before filtering, the dataset contained 20,848 Bitcoin alerts and 12,104 Ethereum alerts. Ethereum is the second-largest cryptocurrency and the main platform for much of decentralized finance activity.

To isolate events, the authors retained only whale transactions with no other whale transaction within 120 minutes before or after the event. This left 6,645 Bitcoin whale transactions and 5,075 Ethereum whale transactions. Those figures are the isolated-event totals, not the final sample sizes used in the directional models. A footnote states that missing buy or sell labels removed 761 Bitcoin transactions and 232 Ethereum transactions from those models.

The wallet tiers were recalculated during fixed six-month intervals—January through June and July through December—using each wallet’s maximum transaction size. As a result, the classification of a wallet as small, medium, or large could change over the study period.

The study tracks blockchain addresses, not verified individuals or institutions. It therefore cannot establish who controls a particular wallet or whether every on-chain transfer represents a deliberate trade.

Bitcoin and Ethereum Results

The same-direction pattern was more pronounced for Bitcoin than for Ethereum. In the Ethereum results, the largest non-whale seller group showed only a 0.76-percentage-point change. Estimates from Ethereum’s buy models were not statistically significant for any wallet-size group.

The authors attributed the difference to persistent informational and structural asymmetries between large and small digital-asset investors. They also noted that the Bitcoin-Ethereum difference remained after Ethereum’s September 15, 2022, transition to proof of stake.

The paper does not claim market manipulation or explain any particular Bitcoin price movement. It also does not establish that following whale activity would produce profits. Although the abstract discusses volatility over a broader 24-hour period, the detailed results show volatility reversing within a day.

The document is preliminary research circulated for discussion. Its official cover says that the views expressed are those of the authors and do not necessarily reflect the Federal Reserve Bank of Philadelphia or the Federal Reserve System. It is therefore not a new regulation, an enforcement action, or an endorsement of copying whale transactions.

CryptoSlate also highlighted the limitation that the results show an association rather than proof of deliberate imitation in its September 11, 2026, summary: “Philadelphia Fed Finds Bitcoin Traders Follow Whale Signals Faster Than Ethereum Users”.

For Bitcoin holders, the study’s finding is observational rather than prescriptive. Public whale alerts appear to coincide with a burst of same-direction trading during the first 15 minutes, but the research alone cannot determine whether following that activity would help or hurt. Broader concerns about how large players affect markets have also contributed to institutional changes, including Bitcoin Suisse’s reported plan to cut 60 Swiss jobs: Bitcoin Suisse plans cut 60 Swiss jobs.

For related coverage, CoinLineup reported that Revolut disclosed identity and Bitcoin records after a fake agency request and that Blockstream rejected a ransom demand in a Liquid Bitcoin exploit.

Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital-asset markets carry significant risk. Readers should conduct their own research before making decisions.