Reading Bitcoin From First Principles

Most people reach for exponential growth to describe Bitcoin. The one thing that actually follows a clean exponential runs the opposite direction: issuance is exponential decay. A constant halving — the same mathematical curve as growth, mirrored. New supply is cut in half on a fixed schedule and marches toward zero.

• Provable, not predicted. ~450 new BTC issued per day now, dropping to ~225 after the April 2028 halving, halving again every four years until the last fraction around 2140.
• This is the anchor. Zero uncertainty. You can verify it yourself on any node. Everything else in this document is less certain than this — so we start here and reason outward.

Because issuance is front-loaded, the supply story is almost over. This is a fact with a date on it:

• The 20-millionth coin was mined March 9, 2026 (block 939,999). Over 95% of all Bitcoin that will ever exist is already in circulation.
• Fewer than 960,000 coins remain to mine — under 5% — trickling out over the next ~114 years.
• Effective supply is even tighter. After an estimated 3–4 million coins lost forever (Chainalysis / River / Ledger estimates), only ~16–17.5 million are genuinely available to buy, sell, or hold.

Adoption follows an S-curve — slow start, explosive middle, saturating top. This is the shape every monetary and technological diffusion has taken, from electricity to the internet. What we can’t know from inside the curve is exactly where we are on it. But we can triangulate better than most people admit, by anchoring on the countable tiers.

The demographic we CAN count: public companies

Public companies are the cleanest adoption signal, because every one is disclosed. The path is steep and recent:

One company to roughly two hundred in six years — and the rate of accumulation now outpaces new issuance. That is the unmistakable signature of the early steepening of an S-curve: still small in absolute terms, but the slope is turning upward.

The method: anchor on the countable, estimate the rest

You triangulate the curve’s position by building from what’s disclosed toward what isn’t — keeping each tier honest about its certainty:

• Tier 1 — Public companies (COUNTABLE). ~200, fully disclosed. Your hard anchor. Started from zero in 2020.
• Tier 2 — Banks & institutions (COUNTABLE, track separately). A distinct wave that began later than corporates and moves on regulation, not treasury policy — so it deserves its own line, not blending with companies.
• Tier 3 — Private companies (ESTIMATED from Tier 1). Not disclosed, but boundable: private firms vastly outnumber public ones and typically adopt new treasury practices later and slower. Use the public count and the known private-to-public ratio to bracket a range — explicitly a range, not a number.
• Tier 4 — Individuals & sovereigns. Individuals via on-chain + exchange proxies (wide error bars); sovereigns are few and mostly public.

Here is where honesty separates from hype. Price is not a fifth independent thing — it is the release valve for the other three. Supply can’t flex. So when adoption (demand) rises or falls, all of that pressure has exactly one place to go: price.

• This is why it’s so volatile, not despite the fixed supply but because of it. In any other market, a demand surge pulls forth new supply that cools the move. Here, nothing cools it. The full force lands on price, up and down.
• It’s the least modelable of the four, precisely because it carries all the adjustment. The firmer the supply, the more the price has to move to balance any change in demand.

A word on the popular price “laws”

Two get quoted most. Both describe the past; neither predicts the future — and each has a tell worth knowing:

• The power law (log-log straight line). Fits history remarkably well. But its errors are one-sided — the big deviations almost all run out the top (bubble overshoots), rarely the bottom. A model whose misses only go one direction isn’t a law; it’s a trend with an unmodeled force periodically overriding it. It describes; it cannot promise.
• “Deceleration” on a log chart is sleight-of-hand. When someone says the curve “decelerates,” they mean the percentage rate slows. But on a log chart, each equal-looking step — 1 to 10 to 100 to 1,000 — is 10× more actual dollars than the last. “Decelerating” in percent can still be staggering acceleration in dollars. Watch the word.

Read top to bottom, certainty drains away by design. That’s not a weakness in the analysis — it’s the analysis being honest about itself. Reason from the top, hold the bottom loosely, and you understand Bitcoin better than most of the people shouting about its price in either direction.