Author: Yuanshan Insight
Data sources: Farside Investors, SoSoValue, Federal Reserve H.4.1 report, CryptoQuant
On the first trading day of 2026, BTC ETFs saw a net inflow of $471 million.
What does this number mean?
Over the two months of November and December, spot BTC ETFs had a combined net outflow of approximately $4.57 billion; with a net outflow of about $1.09 billion in December alone.
Many people frantically sold at a loss above 93K, and institutions bought back about one-tenth of that in just one day on January 2nd.
Simultaneously occurring were:
- The Federal Reserve's balance sheet increased by approximately $59.4 billion weekly (WALCL: 6.6406 trillion on 12/31, an increase of about $59.4 billion compared to 12/24)
- A new whale's holdings surpassed 100,000 BTC ($12 billion)
- BTC rebounded from 87.5K to 93K (+6.8%)
The simultaneous appearance of these three data points indicates a shift in capital flows.
The 2025 rally was driven by "narratives" (halving, ETF launch), while the 2026 rally is driven by "real money" (Fed liquidity injection, institutional subscriptions, whale accumulation).
This is the second stage of the market: shifting from emotion-driven to capital-driven.
01| What Happened: Three Simultaneous Signals
Signal 1: ETFs Reverse Selling Pressure
Over the two months of November and December, spot BTC ETFs had a combined net outflow of approximately $4.57 billion; with a net outflow of about $1.09 billion in December alone. Retail investors frantically sold at a loss in the 90–93K range, spreading panic.
But on January 2nd, BTC ETFs saw a net inflow of $471 million, the highest single-day inflow since November 11th, 2025.
What does this mean? Institutions are buying at the levels where retail is selling.
The data is more直观:
BlackRock's IBIT is currently the largest single BTC spot ETF; in terms of trading activity, IBIT is often统计 as accounting for nearly 70% of the trading volume share.
The total net asset size of spot BTC ETFs is at the hundred-billion-dollar level.
Cumulative trading volume of US crypto ETFs has exceeded $2 trillion.
Signal 2: The Fed Shifts to Expanding Its Balance Sheet
In March '22, the Fed initiated QT (Quantitative Tightening), which lasted nearly 3 years. The essence of QT is to drain liquidity from the market, which was the root cause of the暴跌 in all risk assets during '22–'23.
But according to authoritative sources (Reuters, Fed reports, etc.), QT stopped/ended balance sheet reduction on December 1st, 2025.
Starting in January, the Fed is not only not draining liquidity but is also adding water back in.
The Federal Reserve's balance sheet increased by approximately $59.4 billion weekly (WALCL: 6.6406 trillion on 12/31, an increase of about $59.4 billion compared to 12/24)
Starting in December, the Fed automatically began technical purchases of short-term Treasury bills from the market for the purpose of replenishing reserves (RMP), about $40 billion in the first week; the market expects后续 to maintain a节奏 of "slowly expanding the balance sheet to replenish reserves," but on a more controlled scale.
This means the key turning point is from "draining" to "adding" liquidity.
Signal 3: New Whales Accelerate Accumulation
On-chain data shows that new whales are accumulating BTC at a record pace:
New addresses hold over 100,000 BTC, worth approximately $12 billion.
Tether purchased 8,888 BTC ($780 million) on New Year's Eve 2025, total holdings exceed 96,000 BTC
Long-term Holders have switched to a "net accumulation" state over the past 30 days
But there is an important controversy: CryptoQuant's head of research points out that some "whale accumulation" data might be misled by exchange internal wallet consolidation. After filtering out exchange factors, true whale addresses (100–1000 BTC) are actually减持 slightly.
Real buying mainly comes from: new whales (small,分散 addresses) + ETF institutional subscriptions.
The common point of these three signals: Money is entering the market, and it's "smart money."
02| Why Institutions Enter When Retail Sells
First Layer: Fed Liquidity Injection Creates a Base
The Fed started QT in March '22, reducing its balance sheet from $9 trillion to $6.6 trillion, cumulatively draining $2.4 trillion in liquidity.
What happened during QT?
2022: Nasdaq fell 33%, BTC fell 65%
2023: Interest rates raised to 5.5%, FTX bankruptcy, Luna归零
All risk assets were under pressure.
But in December 2025, QT officially stopped. Starting in January, the Fed shifted to "reserve management purchases." This is not QE (Quantitative Easing), but at least liquidity is no longer flowing out and has begun有限 flowing in.
What does this mean for BTC?
Refer to history: In March 2020, the Fed launched unlimited QE, BTC rose from $3,800 to $69,000 (+1,715%). The scale this time is much smaller than 2020, but the direction has changed.
More dollars entering the market will seek high-yield assets. BTC, as "digital gold," is a natural receiver.
Second Layer: ETFs Become the "Highway" for Institutional Allocation
In January '24, BTC spot ETFs launched, significantly lowering the barrier for institutions to allocate to BTC.
- No need to learn private keys, cold wallets, on-chain transfers
- Compliant channel, can be included in pension funds, hedge funds, family office asset allocation
- Good liquidity, can be bought and sold随时, no withdrawal limits
Why outflows in December? Retail FOMO chased highs, buying near 93K.
Why inflows in January? Institutions配置 rationally, buying the dip at 87–90K.
Key data:
- BlackRock's IBIT holds 770.8K BTC, is the largest single BTC ETF
- ETF cumulative trading volume exceeded $2 trillion
Before ETFs, institutions wanting to allocate BTC needed to build cold wallets, train teams, handle regulatory risks. After ETFs, it's just a few clicks in a brokerage account.
Third Layer: "Generational Changeover" of New Whales
Traditional whales (entered 2013–2017) may take profits at highs. Their cost basis is extremely low (hundreds, thousands of dollars), 90K is astronomical profits.
But new whales (entered 2023–2026) are taking over. Their cost basis is around $50–70K, 90K is just the starting point.
Tether's logic is most typical: Since May 2023, use 15% of profits each quarter to buy BTC. Regardless of whether BTC is at 60K or 40K, they坚持购买. Executed for 10 consecutive quarters, never interrupted.
Average cost $51,117, current price 93K, floating profit over $3.5 billion
This isn't luck, it's discipline.
This is a "generational changeover of筹码," from "early believers" to "institutional allocators." Old whales take profits, new whales take over. Market structure is healthier, holders are more分散.
03| Three Risks That Cannot Be Ignored
Risk 1: Controversy Over "New Whale" Data
CryptoQuant's head of research points out that recent "whale accumulation" data might be misleading: Exchange internal wallet consolidation can be misjudged as "whale buying"
After filtering out exchange factors, true whale addresses (100–1000 BTC) are actually减持 slightly
Real buying mainly comes from: new whales (small,分散 addresses) + ETF institutional subscriptions
What does this mean?
Data needs discernment, cannot be blindly believed. Real buying still exists, but not as exaggerated as surface data suggests. Market rises more靠 "continuous small buying" rather than "large purchases."
This is actually good. It means the market is more分散, not reliant on a few large holders.
Risk 2: The "Limited Nature" of Fed Balance Sheet Expansion
The expansion is "technical buying to replenish reserves," different from QE, and more controlled in scale. If the market overinterprets it as expectations for 2020-style QE, it will be disappointed.
Current RMP is technical buying, not the same as actively injecting liquidity into the market, scale is much smaller than 2020 QE (which expanded over $1,000 billion monthly on average then).
This means liquidity improvement is有限. BTC won't rise "mindlessly" like '20–'21 (from 4K to 69K). Need to wait for clearer monetary policy shifts (like rate cuts, restarting QE).
2026 might be a "slow bull."
Risk 3: The "Time Lag Trap" Between Retail and Institutions
Institutions buy at 87–90K, retail chases at 93K. If BTC corrects to 88K:
Institutions are still profitable, continue holding; retail gets trapped, panics and sells
Result: Institutions buy again at low levels.
This is the eternal cycle:
- Institutions look at 4-year cycles, retail looks at 4-week fluctuations
- Institutions have discipline, retail relies on feeling
- Institutions buy against the trend, retail chases rallies and sells dips
The November-December data is the best proof: Retail sold at a loss at 93K (combined net outflow ~$4.57B over two months), institutions bought at 87K (net inflow $471M on Jan 2). Institutions profit from the money retail panics and sells.
04| How This Rally Differs from 2025
2025 Rally: Narrative-Driven
Core logic: Halving + ETF launch + post-halving supply shock
Funding source: Retail FOMO, institutional试探性配置
Price performance: Rose from 25K to 73K (+192%)
Risk: After narratives were realized, capital receded (combined net outflow ~$4.57B over Nov-Dec)
2026 Rally: Capital-Driven
Core logic: Fed liquidity injection + continuous ETF inflows + new whale accumulation
Funding source: Institutional long-term allocation, sovereign funds, family offices
Price performance: Rebounded from 87K to 93K (+6.8%, just started)
Advantage: Capital-driven is more sustainable than narrative-driven
Key Differences:
Driving force changed: '25靠 "expectation," '26靠 "real money." Narratives can change overnight (like SEC attitude shift, regulatory policy adjustments), but capital inflow is real buying pressure.
Sustainability changed: Narratives fade (halving effect diminishes, ETF launch novelty wears off), but capital stays (institutional allocation is long-term, not frequent trading).
Volatility differs: Capital-driven phase has lower volatility. Institutions won't chase rallies and sell dips like retail; they have clear allocation plans and discipline.
This means '26 might not see "暴涨暴跌" like '21,更像是 "slow bull": climbing up step by step, with smaller corrections.
Retail needs to adapt to the new节奏, don't expect "double overnight," have patience.
Refer to gold 2019–2024, gold rose from $1,300 to $2,700 (+107%), took 5 years. No暴涨, but also no暴跌. This is the characteristic of an institution-dominated market.
05| Three Revelations for Us
First, Learn to Read the Moves of "Smart Money".
Don't follow the K-line, follow the money:
- ETF inflow = Institutions are buying
- Fed balance sheet expansion = Liquidity improving
- New whale accumulation = Long-term signal
These three indicators are more important than any technical analysis. K-lines can lie (can画门, wash trades, fake breakouts), but capital flow doesn't lie.
Second, Understand the "Time Lag" Trap.
Institutions buy when retail panics, sell when retail FOMOs. If you always chase rallies and sell dips, you are the one being harvested.
Learn to:
- Buy when institutions buy (even if scared at the time)
- Sell when institutions sell (even if excited at the time)
- Use ETF flow data to judge, not feelings
Third, 2026 Might Be a "Slow Bull," Have Patience.
The 2021 "暴涨" won't repeat. This bull market is more like:
- Rising 5–10% monthly
- Lasting 12–18 months
- Ultimately reaching new highs, but the path is more曲折
If you expect "get rich overnight," you'll be disappointed. But if you have patience, you might find this bull market more "comfortable" than the last. Smaller corrections, no need for daily心惊胆战.
Previously, BTC fell from 69K to 15K, a 78% drop. Many sold at various levels like 60K, 50K, 40K,最终绝望 at 15K.
If 2026 is a slow bull, corrections might only be 15–20%. A drop from 90K to 75K, not from 90K to 20K. In this market environment, holding is easier, mentality is steadier.
Final word: Understanding institutional capital dynamics is more important than predicting prices. When you understand capital flows, you won't panic when you should buy, nor be greedy when you should sell.
Retail sold at a loss at 93K in December, institutions added at 87K in January. This is the gap.
