Listed companies bought 166,984 BTC in the first half of 2025. Miners produced 81,153 BTC over the same period.
The math is brutal: net corporate demand exceeded new supply by 106%.
The ledger doesn't care about your hopium. It just records the imbalance.
Let me walk through the numbers, the structural shift, and why most retail traders are still reading this graph wrong.
Context: The BTCTreasuries data dump
The numbers come from BTCTreasuries, a public tracker that aggregates disclosed Bitcoin holdings of publicly traded companies. This includes MicroStrategy, Marathon Digital, Riot Platforms, and a handful of others.
Key distinction: This dataset only captures companies that voluntarily report their BTC holdings. It misses private entities, family offices, sovereign wealth funds, and ETF inflows. The actual institutional demand is almost certainly larger.
H1 2025 snapshot: - Listed companies net purchased: 166,984 BTC - Miner production: 81,153 BTC - Net corporate absorption rate: 205% of new supply
That means for every new Bitcoin mined, corporations bought 2.05. The remaining 1.05 BTC per coin came from existing market supply – coins from early adopters, speculative traders, or secondary sales.
This is not a blip. It's a structural regime change.
Core Analysis: The mechanics of a supply crisis
Let's break this down like an order book, because that's what it is.
Side A: The sellers Miners are forced sellers. They have operational costs – electricity, hardware leases, payroll. The 81,153 BTC represents their gross production, but not all of it hits the market immediately. Some is held as treasury, some is sold OTC. But the average miner sells 60-80% of block rewards within days.
Side B: The buyers Listed companies are not traders. They are accumulators. MicroStrategy alone added ~30,000 BTC in H1 via convertible note issuances. These are structural buys – not market-timed entries.
Net effect: The buyer pool absorbed not only all new supply, but also drained existing liquidity. Exchange balances across tracked platforms dropped by ~200,000 BTC in the same period.
Chain data confirms it: addresses associated with corporate treasuries grew by exactly 166,984 BTC.
Code does not lie, but liquidity does. The liquidity is disappearing from order books into cold storage. That creates a mathematical floor, not a ceiling.
Contrarian: Why this isn't a guaranteed rocket ship
Every narrative has a blind spot. Here are three.
1. Net vs. gross purchases "Net" means buys minus sells. If a company sold 50,000 BTC while another bought 216,984, the net is still 166,984. The dataset doesn't reveal gross flow. A single large sell order from a struggling miner or a company like Tesla (which has sold before) could distort the picture.
2. The Q4 flush risk Listed companies have quarterly reporting cycles. If BTC prices dip or if their stock underperforms, CFOs may be forced to liquidate to meet earnings expectations. The same institutions that bought in H1 could become sellers in Q4. The data is backward-looking.
3. The hidden seller: ETF arbitrage Spot ETFs are not included in this dataset, but they create a parallel demand channel. However, if ETF flows reverse, the selling pressure gets amplified because ETFs can redeem shares in creation units. The BTCTreasuries data doesn't capture that.
Survival is the first profit metric. Don't confuse a six-month trend with a permanent state.
Takeaway: What to do with this data
You don't trade on headlines. You trade on structural shifts.
This data confirms one thing: the buyer composition of Bitcoin has permanently changed. Retail no longer sets the price. Corporate treasuries and ETFs are the marginal price setters.
Actionable levels: - Monitor quarterly filings from MicroStrategy, MARA, RIOT. If net additions drop below 20,000 BTC per quarter, the absorption rate falls below 100%. - Track exchange reserve data. If reserves start rising while corporate holdings flatten, the supply squeeze is reversing. - Watch the miner-to-exchange flow. If it spikes above 60,000 BTC/month, miners are front-running corporate demand.
The moon is a myth. The ledger is the only truth.
I wrote my first on-chain audit script for the Parity multisig hack in 2017. That taught me to trust data over narratives. This dataset is clean. The signal is real. But execution still depends on timing.