Is it a shakeout or a trend reversal signal? BTC retraces all of May's gains, ETF six-week inflow ends

Following the prior strong rebound, Bitcoin (BTC) saw a sharp pullback in mid-May. In just 48 hours, the price gave back nearly all of the gains from earlier in the month, returning to consolidation above $76,000 USD. Under the dual blow of macro data coming in far above expectations and a rapid reversal in ETF fund flows, market sentiment abruptly shifted from “greedy or neutral” to the “fear” zone.

Behind the Sudden Surge and Plunge: How Macro Data Changes Market Expectations

In this round of declines, the crypto market displayed clear macro-driven characteristics. The trigger for this deep pullback was a comprehensive upside surprise in U.S. April inflation data. The U.S. Bureau of Labor Statistics shows that April PPI year-over-year surged to 6%, far above the 4.9% expected by the market, reaching a historical high since December 2022. The month-over-month growth rate came in at 1.4%, the largest single-month increase in more than four years. Before the PPI data was released, CPI year-over-year had already reached 3.8%, exceeding the market expectation of 3.7%. This broad inflation upside surprise directly drove the market to reprice the Federal Reserve’s monetary policy path. CME FedWatch data indicates that, as of May 16, the market’s probability of rate hikes before December 2026 had risen to 39.1%. The shift in interest rate expectations delivered a fundamental shock to valuation benchmarks for risk assets.

End of Six-Week ETF Inflow Record: Why Did Institutional Funds Suddenly Withdraw?

In the second week of May, the U.S. spot Bitcoin ETF market experienced a notable turn. According to SoSoValue data, from May 11 to 15, Bitcoin spot ETFs recorded total net outflows of approximately $1.039 billion, officially ending the six consecutive weeks of strong inflows. This reversal was not gradual. At the start of the week there was a modest inflow of $27.29 million; on Tuesday, outflows surged to $233 million; on Wednesday, outflows sharply widened to $635 million. On Thursday, there was a brief return of $131 million, but on Friday, outflows picked up again by $290 million. The fragmented “hit-and-run” pattern of capital reflects that institutions’ reactions to macro uncertainty were becoming increasingly consistent. From product structure, ARKB and IBIT led with weekly outflow sizes of about $324 million and $317 million, respectively. The historical peak of cumulative net ETF inflows occurred in October 2025, at about $61.19 billion. After sustained outflows of roughly $6.38 billion from November 2025 to February 2026, the recent $3.29 billion rebound repaired nearly half of the gap; however, the more than $1 billion outflow in mid-May reveals the fragility of demand.

Leverage Markets Take a Severe Hit: How Liquidation Data Reflects Risk Concentration

The downward move driven by macro shocks and the reversal of ETF capital flows directly triggered a chain of liquidations in the derivatives market. Coinglass data shows that within the 24 hours as of May 18, the total liquidation amount across the whole network reached $657.9 million. Long positions accounted for nearly 89% of the total, with long liquidations totaling approximately $584 million. Ethereum, with liquidations of $257 million, was the hardest-hit liquidation asset in the same period. The largest single liquidation order was an ETH/USDT contract worth approximately $28.49 million, and about 135,604 traders were liquidated within the same 24 hours. The fact that nearly 90% of liquidations were longs highlights how extremely skewed leverage positioning was before this selloff: large accumulations of 20 to 50x leveraged long positions around $80,000. When the price failed to hold the $78,000 USD support and moved lower again, dense long exposure directly triggered a spiral of forced position closures.

Strategic Holders vs. Hedge Funds: What Does Behavioral Divergence Between Institutions Mean?

As the market panicked and retail investors exited via stop-losses, institutions were experiencing a significant internal divergence. Between May 11 and 17, Strategy spent roughly $2.01 billion, adding 24,869 BTC at an average price of $80,985 USD, bringing its total holdings to 843,738 BTC—accounting for more than 4% of Bitcoin’s total supply. Strategy’s cumulative average cost basis is approximately $75,700 USD. As of May 17, the position still showed unrealized gains. However, not all institutions were “betting” in the same direction. Goldman Sachs’ 13F filings show the firm fully liquidated all XRP and Solana ETFs and significantly reduced its holdings in Ethereum ETFs by about 70%. Goldman Sachs’ holdings in XRP ETF products once approached $154 million, and the complete exit in Q1 2026 indicates a strategic shift in focus.

On-Chain Panic and Additional Buying Coexist: The Sentiment Time Gap Between Retail Traders and Whales

Behind the apparent panic is a clear divergence in on-chain flows. On May 19, the crypto fear and greed index fell to 28, entering the “fear” zone. Only a week earlier, the index was at a neutral level of 48; within a week it dropped by nearly 42%. Meanwhile, however, the number of whale addresses holding at least 100 BTC rose to 20,229, up about 11.2% from 18,191 a year earlier. Addresses holding between 10 and 10,000 BTC accumulated a net increase of roughly 56,227 BTC since December 2025, showing a distinct bullish divergence. Retail investors reduce positions and take profits amid fear, while large holders systematically increase allocation—this kind of divergence has been repeatedly validated across multiple cycles. The fear index reading of 28 more reflects a contraction in short-term risk appetite than a comprehensive deterioration of the crypto market’s internal structure.

Dropping Below the $80,000 Threshold: Analyzing Market Structure After Support Fails

Throughout the entire process of the market falling from the $80,000 high, the market’s “chip” structure itself already contained adjustment pressure. In the $79,800 to $80,500 range, the order book shows sell orders that are more than 3 times thicker than buy orders. On Deribit, the notional value of $80,000 call options expiring from May to June exceeds $1.5 billion. Market makers’ long gamma hedging mechanism leads to a passive increase in sell volume as prices approach $80,000, creating a top-pressure effect that reinforces itself. On-chain data shows that the short-term holders’ cost basis is in the $80,000 to $81.8k range. When the price hits $80,000, profits are distributed at a rate of about $4 million per hour, while mid- to long-term positions held for 2–3 years are also being realized at roughly $210 million per hour. Glassnode describes this as “when the price approaches the cost basis of short-term holders, the motivation to exit exceeds the demand to enter.” The market action matches a typical bear-market response—there is a lack of systemic conviction for new trend-setting buying.

The Transmission Chain of Geopolitical Shocks: Energy Prices, Inflation Expectations, and Risk Pricing

Crude oil prices are the core variable linking geopolitical conflicts, inflation expectations, and valuation of the crypto market. In mid-May 2026, escalating geopolitical tensions in the Middle East pushed Brent crude oil prices rapidly up into the $111 to $112 range, while Bitcoin prices fell in parallel to below $77,000 USD. With commercial shipping through the Strait of Hormuz restricted, international oil prices continued to stay above $100 per barrel. Energy prices therefore kept rising, driving an upside rebound in inflation expectations beyond forecasts. The transmission mechanism is clear and direct: geopolitical tensions → energy prices rise → inflation expectations strengthen → expectations for loose monetary policy shrink → valuation pressure on risk assets. In this transmission chain, crypto assets are often hit first due to their high volatility and strong risk-sensitivity.

Three Scenario Projections the Market May Face After This Selloff

Based on the current macro environment and capital structure, three possible follow-on evolution scenarios can be projected. Scenario 1—short-term repair after panic is released: if subsequent inflation data shows marginal easing, and ETF outflows enter a slowdown period after reaching the $1 billion scale, the market may build a new game center around the $75,000 to $78,000 USD range. Scenario 2—rate-hike expectations further intensify: if future inflation data remains elevated, the probability of additional hikes will keep climbing. The crypto market would then face continued macro-driven valuation compression, and the liquidity premium would further narrow. Scenario 3—rebalancing institutional strategies in a more differentiated way: Strategy’s continued accumulation and Goldman Sachs’ strategic pullback occur at the same time. This in itself suggests a lack of consistent market expectations among institutions. This structure typically corresponds to the continuation of range-bound/choppy trading rather than the start of a sustained uptrend or downtrend. The common premise behind all three scenarios is: as long as oil prices remain high and geopolitical risks have not cooled off, macro pressures on risk assets will keep persisting.

Summary

Bitcoin’s 48-hour retracement of all May gains is the result of multiple factors converging: PPI year-over-year at 6%—a three-year high; the probability of Fed rate hikes rising to 39%; spot ETF inflows ending after six weeks with net outflows exceeding $1 billion; together with nearly $657.9 million in liquidations across the derivatives market and the transmission of geopolitical shocks. However, institutions are not all exiting in a uniform way. Strategy added $2.01 billion against the panic, while Goldman Sachs significantly adjusted its ETF position structure—both representing markedly different strategic orientations. In the short term, macro pressure and capital structure tend to favor range-bound oscillation; in the medium to long term, the direction depends on whether inflation trends can show a substantive inflection point and whether institutional capital allocation logic undergoes a fundamental change.

FAQ

Q: What is the core trigger for this BTC decline?

The primary trigger is the macro-level inflation data coming in fully above expectations—April PPI at 6% year-over-year, setting a three-year high—combined with higher CPI. This pushed market expectations for Fed rate hikes to above 39%, which in turn triggered systemic repricing of risk assets.

Q: How did the capital flows of Bitcoin ETFs change?

For the week ending May 15, Bitcoin spot ETFs officially ended the six-week streak of net inflows, recording approximately $1.039 billion in net outflows. Among them, ARKB and IBIT were the two products with the largest outflow sizes.

Q: How large were the long liquidations?

In the past 24 hours, the total liquidations across the whole network were approximately $657.9 million. Long positions accounted for nearly 89%, and long liquidation amounts were about $584 million, highlighting the extreme concentration of leverage in the prior market.

Q: Are institutional investors uniformly bearish?

No, not at all. Strategy increased holdings by purchasing about 24,869 BTC with $2.01 billion during the panic period. Goldman Sachs also made major adjustments to its crypto asset allocation, liquidating its XRP and Solana ETFs. Together, these reflect different investment strategies.

Q: What is the current market sentiment level?

The fear and greed index has dropped to 28 and is in the “fear” zone, down nearly 42% within a week. However, on-chain data show that the number of whale addresses is still increasing. This sentiment divergence suggests that any adjustment is more driven by short-term external shocks rather than a systemic deterioration of the internal structure of the crypto market.