Over the past five months, the headlines have been relentless: Bitcoin is losing the ETF war, institutions are fleeing, and gold is the true safe haven. But the raw numbers carve a different outline. Gold’s largest ETF, GLD, has bled $12 billion since March. Bitcoin’s entire ETF complex—all eleven spot products combined—has shed $8 billion over the same frame. The narrative is not false. It is incomplete.
Code is the oracle; data is the only scripture.
Let’s start with the context. GLD holds roughly $130 billion in assets under management. All spot Bitcoin ETFs together hover around $65 billion—exactly half the size. When an asset twice the size leaks more capital in absolute terms, the instinct is to cry ‘relative outperformance.’ But instincts are not analysis.
The numbers demand a forensic read. In March, GLD saw $4.2 billion in net outflows. April added $3.5 billion. May contributed $3.2 billion. June? Only $1.1 billion. July’s first two weeks recorded a whisper—less than $50 million. The bleeding is slowing, possibly stopping. Bitcoin ETFs, by contrast, accelerated in the opposite direction. Net outflows in May: $2.8 billion. June: $4.5 billion. The first two weeks of July show no sign of deceleration.
The Kobeissi Letter, which compiled these figures, notes that GLD’s total outflow since March is 50% larger than Bitcoin ETF outflows since January. But that comparison cheats on time windows. GLD’s outflow clock starts in March—after gold peaked near $5,600. Bitcoin’s clock starts in January, when BTC was already falling from $95,000. The asymmetry matters.
Liquidity flows like water; follow the evaporation.
Now the core—what the data actually reveals. First, the velocity of Bitcoin ETF outflows is higher. GLD lost 9.2% of its AUM over five months. Bitcoin ETFs lost 12.3% of their AUM over six months. On a per-month basis, Bitcoin ETFs are bleeding at 2.05% of AUM versus GLD’s 1.84%. The difference is small but directional. Second, the price impact is disproportionate. Gold fell from $5,600 to $4,000—a 29% drop—despite a 9% AUM outflow. Bitcoin fell from $95,000 to $57,700—a 39% drop—despite a 12% AUM outflow. The amplification factor is higher for Bitcoin. That suggests thinner order books, higher retail sensitivity, and less diversified demand.
During the 2022 Terra collapse, I monitored withdrawal rates in real-time. I noticed a 15% increase in large wallet withdrawals 48 hours before the public announcement. That experience taught me to look beyond absolute numbers to the structure of outflows. Here, the structure tells us that Bitcoin ETF outflows are concentrated in a few funds—likely those with higher retail exposure—while GLD outflows are more evenly distributed across institutional and retail holders. Concentration accelerates price impact.
The code does not lie, but it often omits.
The contrarian angle is uncomfortable but necessary. The ‘Bitcoin is winning compared to gold’ narrative is a trap. Comparing absolute outflows ignores the fact that gold has a parallel market—central bank reserves, physical bars, coins—that absorbs ETF selling without crashing spot price. Bitcoin has no parallel market. Every ETF redemption flows directly into Coinbase or Kraken order books. The same $1 billion outflow hits Bitcoin’s price three times harder than it hits gold’s.
Moreover, the timing of outflows suggests different drivers. GLD outflows peaked in March-April, coinciding with gold’s all-time high. Profit-taking by long-term holders. Bitcoin ETF outflows peaked in June, corresponding with the unwind of basis trades and the death of the ‘carry trade’ narrative. It is not the same capital rotating out of both. It is two separate pools of liquidity evaporating for different reasons.
So where does this leave us? The market is pricing fear. Bitcoin at $57,700 is discounting further ETF outflows. Gold at $4,000 is discounting a slowdown in central bank buying. The next signal is binary: if Bitcoin ETF outflows slow in July as dramatically as GLD outflows did in June, the bottom is likely in. If they continue at $4.5 billion per month, we are only halfway through the washout.
Based on my audit experience, I would watch the daily flow data from Farside with a specific filter: the ratio of Bitcoin ETF outflows to GLD outflows on a seven-day rolling basis. If that ratio falls below 1.0—meaning GLD is bleeding faster again—the narrative flips. If it stays above 2.0, the fear is justified.
The data does not lie, but it often omits the context that makes the difference between a signal and noise.
The next seven days will decide whether this is a buying opportunity or a trap. I am watching the hash, not the hype.