kommunitas logo
BlogInvesting

Why Bitcoin Fell 50% From Its Peak: The 5 Forces Behind the Crash, Explained

Why Bitcoin Fell 50% From Its Peak: The 5 Forces Behind the Crash, Explained

Why Bitcoin Fell 50% From Its Peak: The 5 Forces Behind the Crash, Explained

Investing • November 20, 2026

By Priyo Harjiyono

TL;DR: Bitcoin traded near $62,000 on the morning of November 20, 2026 — about 51% below its all-time high of $126,080, set on October 6, 2025. The fall was not one event but a pile-up: seven straight sessions of US spot Bitcoin ETF outflows ($1.91 billion), a $6.3 billion hack at Binance still being cleaned up, an AI-stock selloff that triggered margin calls, the largest IPO in history draining cash, and liquidations that turned a slide into a rout. A viral rumor that the US government dumped $2 billion of Silk Road bitcoin is false — on-chain data shows government wallets dormant. Nobody can say which force mattered most.

Key Takeaways:

  • The arithmetic: $126,080 (CoinGecko, October 6, 2025) to about $62,000 (November 19–20, 2026) is about 51% down; the June low below $59,000 was deeper, over 53% off the peak.
  • The measured flow: US spot Bitcoin ETFs lost $1.91 billion over seven straight sessions (November 5–13), and November's total reached $2.35 billion after nine trading days.
  • Tom Lee's five possible causes — the Nvidia selloff, Anthropic's IPO, Binance hack refunds, AI security fears, and chip export rules — are explicitly unconfirmed, and Binance founder CZ said the same.
  • The Silk Road rumor is false: government wallets did not move, and since March 2025 forfeited bitcoin has legally been a reserve asset, not inventory to sell.
  • The disagreement ahead: Bitwise's Matt Hougan sees a 2018-style grind, while Strive's Matt Cole sees a post-IPO recovery toward $90,000 by late February 2027. Both are analogies, not forecasts.

Last updated: November 2026

Bitcoin did not crash on a single day. It bled for six straight weeks, then fell off a ledge in mid-November, and by the morning of November 20, 2026 it was trading near $62,000 — roughly half of what it was worth thirteen months earlier. The Crypto Fear & Greed Index read 10 out of 100, extreme fear, for the seventh straight day. This article starts with the numbers and the arithmetic, then takes the five forces one at a time, with what supports each and what limits it.

How far Bitcoin has actually fallen

The peak is not in dispute. CoinGecko's all-time-high data puts bitcoin's record at $126,080, set on October 6, 2025. Other feeds differ by a few hundred dollars but agree on the day and the level: about $126,000.

DateEvent
Oct 6, 2025All-time high: $126,080 (CoinGecko)
Jun 30, 2026Falls below $59,000 — more than 53% below the record, the year's low
Oct 2026Recovery to about $86,000
Oct 29, 2026Binance hacked for about $6.3 billion; bitcoin below $78,000 by Nov 1
Nov 5–13, 2026Seven straight sessions of spot Bitcoin ETF outflows: $1.91 billion
Nov 19, 2026Anthropic starts trading on Nasdaq; bitcoin low of $61,448.83
Nov 20, 2026Bitcoin near $61,995; Fear & Greed at 10, extreme fear

Two details in that timeline matter most. First, the slide lasted six straight weeks — the selling was already entrenched before November's worst days. Second, this is not the year's worst drawdown: the June 30 low below $59,000 was more than 53% off the peak, deeper than where bitcoin trades now.

The math: is "down 50%" right?

Yes, with rounding. Take the CoinGecko peak of $126,080 and the November 20 price of about $61,995: the decline is ($126,080 − $61,995) / $126,080, which is 50.8%. Against the November 19 intraday low of $61,448.83, it is 51.3%. The title's "50%" is a fair round number; "about 51%" is the more precise one.

Context: past bear markets fell 77% to 85% peak-to-trough (CoinDesk), so a 51% decline is severe but not historically extreme.

Force 1: The ETF bid reversed

The one force with a clean paper trail is the money leaving US spot Bitcoin ETFs. Farside Investors' daily flow data shows seven consecutive sessions of net outflows from November 5 to 13, totaling $1,907.4 million. BlackRock's IBIT accounted for $1,416.4 million of it — about 74% — and has now had eleven straight outflow days since October 30. The week of November 9–13 alone saw $1.07 billion leave, the second straight week above $1 billion, and November's total reached $2.35 billion after just nine trading days. For 2026 as a whole, the funds are now about $1.98 billion in net outflows.

Keep it in proportion: the $1.91 billion is under 4% of the roughly $54.7 billion in net inflows these funds have gathered since launching in January 2024. The patient money did not all leave; it rotated out at the margin, and at the margin is where prices are set.

With the average ETF holder's cost basis near $82,300, per Glassnode, the typical holder was about 21% underwater at $64,800. Underwater holders redeem, and redemptions force the funds to sell bitcoin. That is the mechanical loop at the center of this force, and it needs no narrative to work.

Force 2: The Anthropic IPO vacuum

The second force got the most attention and has the thinnest causal chain. Fundstrat's Tom Lee put it second on his list on November 14: investors were raising cash for Anthropic's IPO, selling holdings — including crypto — to fund subscriptions. Binance founder Changpeng Zhao, known as CZ, put it first on his own list the next day: money rotating into the offering.

The offering itself is genuinely enormous. Anthropic launched its IPO roadshow on November 10, initially marketing 825 million shares at $70 to $78, which would have raised up to $64.35 billion. On November 18 the company priced at $83 a share — above the range — and upsized to 980 million shares, raising $81.34 billion in the largest IPO on record, plus a separate $26 billion private placement. The shares began trading on Nasdaq under the ticker ANT on November 19, opened at $111.50, and closed the first day at $97.17, up 17.1%, for a fully diluted valuation near $1.5 trillion.

The theory is that a deal this size vacuums up cash: investors sell what they hold to buy the new thing, and up to 20% of the offering was reserved for retail investors, with Robinhood named as a likely channel. But no flow data shows crypto being sold to fund Anthropic subscriptions; subscriptions are paid in dollars, and the dollars could have come from anywhere. The timing fits: the theory predicts selling before the listing, and bitcoin's worst days, November 14–15, were in that window. But the stock then rallied 17% on day one, which is not what a market drained of cash usually does. The IPO is a real event and a plausible pressure, not a proven cause.

Force 3: The Binance hack aftershock

On October 29, attackers drained about $6.3 billion from Binance in what was described as the largest hack on record. On October 31, the FBI attributed the theft to North Korean actors associated with the TraderTraitor group. Binance committed more than $6.34 billion of company funds to cover affected customer balances and said it had replenished about 98% of the stolen assets by November 3.

Lee's third force is what happened next: customers whose balances had been restored might be converting their holdings into cash or stablecoins, creating selling pressure. A large group of people suddenly regaining access to funds they thought were gone, in a falling market, are more likely to sell than to hold. But no data has been published on how much of the refunded money was actually sold.

The denial is documented too. As bitcoin fell toward $62,000, speculation spread that Binance itself was selling crypto to cover the breach. CZ denied it flatly on November 15 — "No, Binance did not sell crypto" — and said he had personally bought crypto in recent days.

The hack's wider footprint is measurable in one place: stablecoins. By Glassnode's count, major stablecoin supply fell $14.2 billion over the 30 days through November 11 — the sharpest contraction since the 2022 bear market — with USDT and USDC accounting for nearly $11 billion of it. Stablecoins are the cash of the crypto market; when their supply shrinks by that much, there is less money available to buy anything. Whether that shrinkage caused the fall or merely accompanied it is not something the data settles.

Force 4: AI stocks fell first

Lee's first force has the clearest mechanism. Nvidia and other AI and semiconductor stocks sold off hard, and that selloff triggered margin calls among leveraged retail investors in Asia, forcing them to sell assets — including crypto — to meet obligations. A trader who holds both Nvidia shares and bitcoin on margin does not get to choose which one the broker sells. When the stock leg falls, the crypto leg goes with it.

The week's calendar cooperated. Nvidia reported earnings on November 17–18, and bitcoin briefly climbed to $64,600 on the result, then fell straight back toward $64,000. The macro backdrop was no help: US producer prices rose 0.3% in October, lifting the 12-month rate to 6.5%, and the 10-year Treasury yield closed at 5.15% on November 13. Hot inflation raises the odds of another Federal Reserve rate hike, and higher yields weigh on every risk asset at once.

This is also the force that connects to the bigger shift in what moves bitcoin at all. K33 Research's argument — that institutional flows, ETFs, and macro conditions now dominate price, while the halving's supply shock has shrunk to a rounding error — is exactly what a week like this demonstrates. Bitcoin fell with tech stocks, on tech-stock news, through tech-stock mechanisms. In November 2026 it traded like a leveraged tech stock with no earnings.

Force 5: Chip fears, security fears, and politics

The last two items on Lee's list are the softest, and he presented them that way. Fourth: fears that AI could weaken existing cybersecurity defenses — Anthropic itself had discussed AI's potential to compromise current security systems. Fifth: tighter US scrutiny of chip exports to Asia.

CZ's version of this bucket was broader: hot inflation, and political uncertainty after the US midterm elections, in which Democrats took control of both chambers of Congress. Bitwise's Matt Hougan named the same three pressures — the midterms, the Binance episode, and tight monetary policy. One concrete data point sits underneath the politics: the CLARITY Act, the crypto market-structure bill, failed a Senate procedural vote 49 to 50 on September 15, and it is still not law.

ForceWhat can be measuredWhat cannot be proven
1. ETF outflows$1.91B over 7 sessions; IBIT 74% of itThe flows' share of the move
2. Anthropic IPO$81.34B raised; priced $83; trading Nov 19Any crypto sold to fund subscriptions
3. Binance hack~$6.3B stolen; 98% replenished by Nov 3; stablecoins −$14.2B in 30 daysHow much refunded money was sold
4. AI stock selloffNvidia earnings Nov 17–18; 10Y yield 5.15%; PPI +0.3%How many margin calls hit crypto
5. Chips, security, politicsCLARITY Act failed 49–50 on Sep 15Whether it moved any trade

The rumor that wasn't: no $2 billion Silk Road sale

As prices fell, a rumor spread across X and Reddit that the US Department of Justice had moved $2 billion of seized Silk Road bitcoin to Coinbase for immediate liquidation. On-chain analysis confirms the claim is entirely false: government-controlled wallets remain dormant.

The rumor worked because the underlying fact is real and large. The US government is the largest sovereign holder of bitcoin, with roughly 307,000 to 325,000 BTC — about $20 billion to $28 billion — spread across hundreds of wallets, most of it from the Silk Road cases, the 2016 Bitfinex hack, and the October 2025 Prince Group forfeiture. But moving and selling are different things: in March 2025, an executive order created the Strategic Bitcoin Reserve and directed that forfeited bitcoin be held as a reserve asset rather than sold.

That does not mean government coins never move. Roughly $1 billion in bitcoin recently moved to Coinbase Prime custody addresses, and an earlier batch of about 3,940 BTC moved the same way in July — but those are custody moves, and no sale has been confirmed from them. The contrast that proves the point is Germany: in the summer of 2024, Berlin moved roughly 50,000 bitcoin to exchanges and sold it over a few weeks for about $2.9 billion. A government sale looks like that.

Altcoins fell harder

Bitcoin's 51% is the headline, but the damage was worse further out the risk curve. XRP fell below $1 on November 14, then dropped to $0.89 in the early hours of November 15 — a two-year low — and ended the weekend down about 27% since late October and 21% in November. US spot XRP ETFs recorded $19.13 million in net withdrawals in the week ending November 13, their largest weekly outflow since March. Ether traded near $1,711 to $1,800 over the same weekend. Over seven days, Solana fell about 11%, Hyperliquid about 16%, and Zcash about 15%.

The mechanism is the same one that hits every crash: leverage and liquidity. CoinGlass data put total crypto liquidations at about $1.96 billion over a single 24-hour stretch, with bitcoin accounting for roughly $750 million and long positions making up about 80% of the forced selling. When leveraged longs are liquidated, the exchange sells their collateral into a falling market, which pushes the price lower and liquidates the next set of longs. Altcoins, with thinner order books and more leverage per dollar of market value, fall further on the same flow.

Scheduled supply made it worse. November is a heavy month for token unlocks, and what usually happens to price around a cliff unlock is well documented: pressure tends to start about 30 days before the date, not on it. Monad's unlock on November 24 hangs over the same week.

What the analysts disagree on

The sharpest disagreement of the week is not about what happened. It is about what it resembles.

Bitwise chief investment officer Matt Hougan, in an interview published November 16, said the current environment reminds him of mid-2018, when prolonged losses left traders discouraged and some participants abandoned the industry. His illustration: in 2018, bitcoin fell from $20,000 toward $5,000, traders concluded the worst was over, and then it dropped to about $3,000. He cited the midterm results, the Binance hack, and tight monetary policy as pressures that could take time to work through. His one condition for optimism is concrete: if bitcoin holds above the June low near $58,000 for a few more weeks, 2027 could shape up as a strong year. He still calls bitcoin the best risk-adjusted trade in finance for the next ten years.

Strive CEO Matt Cole offered the opposite comparison. He noted that bitcoin fell 18% in the 50 days before Anthropic's listing, similar to a 19% decline before SpaceX went public in June; after SpaceX listed, bitcoin bottomed 18 days later and recovered to stand 3.7% above its starting level by day 100. On that pattern, Cole suggested a similar recovery could take bitcoin toward $90,000 by late February 2027 — while writing that markets rarely follow a comparison like this exactly.

Both men are reasoning by analogy, built on one prior example each — 2018 for Hougan, SpaceX for Cole. Neither is a forecast. The useful part is what they agree on without saying it: the question is no longer whether the halving calendar says bitcoin should be rising. The driver has changed to flows and macro, and whether Bitcoin's four-year cycle is even still alive is now the same question as whether demand returns.

What to watch next

If the fall was driven by flows, the recovery — if one comes — will show up in flows first. The daily ETF numbers from Farside are the fastest read: a flip back to net inflows would be the first evidence the selling has stopped. The second thing to watch is the June low near $58,000, the level Hougan named; a break below it opens the conversation about the 2018 parallel in earnest, while holding it keeps the October recovery in play as the template.

The Federal Reserve meets on December 9, with the November CPI report landing on December 10. And November 30 is the MSCI deadline for Strategy: Strategy's $10B paper loss and the November 30 MSCI deadline is a separate story from this one — it is about a company's shares, not bitcoin's price.

None of this is financial advice. What the week proved, again, is that bitcoin can lose half its value without a single protocol failing, a single coin being stolen from the network itself, or a single government selling. The causes were outside the protocol: funds redeeming, stocks falling, an IPO absorbing cash, a hack being cleaned up, and leverage unwinding all at once.

Frequently Asked Questions

How far has Bitcoin fallen from its peak?

Bitcoin's all-time high was $126,080 on October 6, 2025, per CoinGecko. At about $62,000 on November 19–20, 2026, that is a decline of about 51%.

What caused the crash?

No single cause has been proven. The measured flow is $1.91 billion leaving US spot Bitcoin ETFs over seven sessions in November, plus about $1.96 billion in forced liquidations in a day. The most-quoted explanations — Tom Lee's five forces and CZ's five pressures — were presented as possibilities, not findings.

Did the US government sell $2 billion of Silk Road Bitcoin?

No. On-chain analysis shows government-controlled wallets did not move, and since the March 2025 executive order creating the Strategic Bitcoin Reserve, forfeited bitcoin is legally held as a reserve asset rather than sold. Custody moves to Coinbase Prime have happened, but no sale has been confirmed from them.

Why did altcoins fall more than Bitcoin?

Leverage and liquidity. About $1.96 billion in leveraged positions was liquidated in 24 hours, roughly 80% of it long positions, and forced selling hits thin altcoin order books harder.

Is this a bear market like 2018?

Bitwise's Matt Hougan says the mood resembles mid-2018, when bitcoin fell from $20,000 toward $5,000 and then to about $3,000. The differences matter too: this market has spot ETFs and institutional holders. There is no consensus yet, and Hougan himself conditions his optimism on the June low near $58,000 holding.

Could Bitcoin recover to $90,000?

Strive CEO Matt Cole has sketched a scenario, based on the SpaceX IPO pattern, in which bitcoin recovers toward $90,000 by late February 2027. It is an analogy, and Cole himself wrote that markets rarely follow a comparison like this exactly.

Sources

This article is for educational purposes only and is not financial advice. Prices and flows are snapshots dated November 2026 and move fast; verify current data before acting on any of it.

Share This Article

This link will open in a new window