Fifty days. That’s how long over half of Bitcoin’s circulating supply has been sitting at a loss. Ledger lines don’t lie. Every UTXO created when price was higher than today is a mark of unrealized pain. The metric is simple: supply in loss as a percentage of total supply. For the first time since the 2018 bear market, it has crossed 50% and stayed there for nearly two months. Market math supersedes market sentiment—but this particular math hides a nuance most traders miss.
Let’s define the data. “Supply in loss” counts every unspent transaction output (UTXO) whose acquisition price exceeds the current spot price. It’s calculated by walking the coin’s history, comparing each output’s creation block price to the current BTC/USD rate. The methodology is transparent—anyone with a Bitcoin node and a Python script can replicate it. Here’s the core logic:
# Simplified script to compute supply in loss
import requests
from datetime import datetime# Fetch UTXO set (requires full node RPC or public API) utxos = get_utxo_set() # placeholder current_price = get_current_price()
loss_supply = 0 total_supply = sum(utxo['amount'] for utxo in utxos)
for utxo in utxos: # Estimate acquisition price from block height at creation creation_price = get_block_price(utxo['block_height']) if creation_price > current_price: loss_supply += utxo['amount']
ratio = loss_supply / total_supply * 100 print(f"Supply in loss: {ratio:.2f}%") ```
Two implementations exist: one from Glassnode (using aggregated exchange data) and one from CoinMetrics (UTXO-based). The two can diverge by a few percent, but both currently read above 50%. This is the raw signal.
Historical context—every time supply in loss has surpassed 50% for an extended stretch, Bitcoin’s price has bottomed within four to eight weeks. In March 2020, it lasted 39 days before the Covid crash found a floor. In November 2018, it persisted 56 days, ending with the $3,100 low. The current 50-day streak is already longer than the 2020 episode and approaching 2018’s peak. But the environment is different.
Here’s where my own experience comes in. During the 2022 bear market, I spent months tracking the correlation between stablecoin de-pegging and Aave liquidations. I discovered that most cascading failures originated from positions above 80% LTV. That taught me a lesson about persistence: a metric alone is noise; its duration and velocity reveal the signal. For supply in loss, the key is not the 50% level but how fast it changes. Right now, the ratio has flatlined—hovering between 51% and 53% for weeks. This plateau suggests holders are not panic-selling, despite being underwater. They are waiting. The market is waiting.
The evidence chain goes deeper. If we overlay exchange net flow data with supply in loss, a pattern emerges: during the plateau, BTC has been moving from exchanges to cold storage at an accelerating rate. The last 30 days have seen over 100,000 BTC leave exchange wallets. This is not the behavior of fearful holders; it’s accumulation by entities that can tolerate drawdowns—likely institutional structures behind the ETFs. BlackRock’s IBIT and Fidelity’s FBTC continue to see inflows even as price stagnates. The supply in loss metric is being distorted by HODLers who refuse to sell, artificially keeping the loss count high while sell-side liquidity dries up.
This brings us to the contrarian angle. Correlation is not causation. The historical 50-day rule may be a relic of a retail-dominated market. Now, with ETF-driven capital, the loss coin may stay high even after a bottom forms. In fact, the metric could remain above 50% for months after the true low, because new buyers at the bottom create new UTXOs at low prices—so loss supply only drops when price rises significantly above those lows. The real bottom may have already occurred, and we just don’t see it in loss data yet. I call this the “loss lag trap.”
Another blind spot: the method for calculating acquisition price. Most analytics use a simplified approach—assigning the price at the first time the coin moved into a wallet. But coins that have been laundered through mixers or been through many hops have inaccurate cost bases. The loss figure for “old” coins (held for >1 year) might be overestimated if the coin was acquired during a previous cycle’s low. We need to segment by age. Coins held for more than 155 days rarely move, so their loss status is theoretical. Adjusting for coin age, the true “active” supply in loss is likely lower—maybe 35-40%. That changes the narrative.
Rules saved the portfolio. Again. In a sideways market, the only edge is patience and verification. The mistake would be to read the 50-day clock and assume fireworks are imminent. The data says: wait for supply in loss to break the plateau, either by a sharp drop (capitulation) or a slow decline (absorption). If it drops below 45% within the next two weeks, that’s a bullish divergence—price will have recovered or holders will have sold, clearing the loss overhead. If it stays above 50% for another 30 days, the historical pattern is broken, and the bottom is delayed.
In the bear market, survival is the only alpha. This market isn’t a bear; it’s a consolidation. But the principle holds: do not anticipate, respond. The on-chain evidence is clear: the supply in loss metric is a lagging indicator. Its current persistence signals pain, but not necessarily prudence. I advise readers to ignore the 50-day count and instead monitor two forward-looking metrics: the rate of change of loss supply (should be negative soon) and the realized price. Bitcoin’s realized price currently sits near $28,500. If price retests that level without loss supply expanding, the floor is confirmed. If price breaks below realized price and loss supply jumps to 60%, the bottom is not in.
What to watch next week: the weekly loss supply moving average. If it rolls over, that’s the buy signal. If it stays flat, keep your powder dry. Data detectives don’t chase headlines; they follow the ledger.
