Data is not guaranteed, and calculations will be off and sometimes wrong. Nothing here is investment advice. Use at your own risk. Full disclaimer

These are not price predictions

Every number on this page is arithmetic on assumptions you can change: a share of an assumed pool, a number of years, a discount rate. None of it is a forecast of Bitcoin's price, a target, or the odds of anything happening. The scenarios are illustrations, not a range of likely outcomes; the real outcome can be worse than all of them, including zero. Informational and educational only. Not investment advice.

The pool
The share. The speed. The return. Five possible futures for Bitcoin's share of monetary premium
$ trillion, fixed
$ a coin
% a year, nominal
Full supply21M
The numbers as a table, and the data
How we got here Observed market value
Loading Coin Metrics history…
What these scenarios would take Illustrative paths
The letter's model
Today's price implies
…
Loading today's price and difficulty…
The scenario What you are pricing
Pick one of the scenarios above. The model says what probability today's price assigns to it. Type the probability you believe. The model says what one coin is worth today. This is the half the letter does not print.
% a year, nominal
The same rate as the cards. A higher rate means more fear about the wait, and raises the implied probability.
$ trillion
What Bitcoin is worth if it stays a niche. Above zero lowers the implied probability.
%
Implied probability Success case down, discount rate across
outlined = your inputs
The ticket One hash's odds at today's difficulty
Every hash is a lottery ticket and the energy is its price. Enter one machine's hashrate to see how long it would wait for a block alone.
The network
One machine, running alone
TH/s
Here be dragons Skip this if you trust the page. It is here so you can check it.

The model treats Bitcoin's price as a probability-weighted blend of a success state and a failure state, discounted to today:

price = p × success ÷ (1 + r)years + (1 − p) × failure ÷ (1 + r)years

Run backward, solve for p. Run forward, solve for the price. Success and failure are values per coin over the full 21 million, the same basis as the cards, and the success state is one of the scenarios above: its share of the assumed premium pool, reached in its number of years, in the nominal dollars of that year, so r is a nominal rate. With no failure value, the backward answer is each card's break-even probability. The "one in N" on this page is the rounded reciprocal of p.

Three caveats from the letter. A higher discount rate (more fear) raises the implied probability, so the 10 percent column is on the conservative side of what the market thinks. A failure state above zero lowers it. And the pool is a scale reference, not a target: the full version of the model takes each asset class, estimates how much of its value is a store-of-value premium rather than use, and asks how much of that premium Bitcoin could take. Those inputs are judgment calls, which is why every one of them is editable here.

The letter itself priced 10% of the $891T visible pool in ten years, in today's dollars at a 10% real rate, and printed "1 in 20". This page prices the scenarios instead, on the assumed $300T premium, so its numbers are not the letter's; /odds/api/compute still runs the letter's version.

Live inputs: price from CoinGecko (mempool.space if it is unavailable); difficulty, hashrate and block height from mempool.space, refreshed every five minutes. Pool values are the letter's, from Savills, SIFMA and the World Gold Council. The scenarios, priced, are in the data link under the cards.

From The Odds, monthly commentary, October 2026: nxtmining.com/reports. Informational and educational only. Not investment advice.