Did Stock-to-Flow predict anything?
Bitcoin's price is determined by its scarcity, and trading below the model-implied value means it will revert upward.
Where we are now. Today the model implies $3.14m against a price of $83,094 — the gap is the reason this rule almost never fires. Shaded bands are the 2 occasions the rule fired, the same ones counted in the table below: 2011-09, 2014-09.
This rule has fired 2 times. A doubling followed within a year on 1 of the 2. On comparable days when it did not fire, a doubling followed 3.6% of the time.
| Times it fired | 2 | separate occasions, not days |
|---|---|---|
| What followed | 50.0% | a doubling within 365 days of that close |
| Could plausibly be | 12% to 88% | the range the true rate could take, on this many occasions, before you invoke anything but chance; a binomial interval that treats the episodes as independent trials |
| Happened anyway | 3.6% | comparable days when the rule did not fire |
| Difference | +46.4 pts | near zero means the signal added nothing |
Where the claim comes from
PlanB, 2019. The most influential Bitcoin valuation model ever published, and the source of the widely circulated $100,000 and $288,000 forecasts for the 2021 cycle.
How we tested it
Model price computed as 0.4 x (stock/flow)^3.3, using PlanB's exponent with a price-scale coefficient that is NOT one of his: his published regression is on market value, ln(mktcap) = 3.3 ln(SF) + 14.6. The 0.4 sits well above both his market-value and price forms, so the model line drawn here is high. Flow is the trailing 365-day change in supply. Trigger: price below the model. Scored as a bottom rule: doubling within 365 days.
Only days the rule could have fired are counted. Consecutive triggers within 90 days are one occasion. The baseline excludes each trigger and the 90 days after it, because the aftermath of a signal is not a fair comparison. No verdict is given below 20 occasions.
What the numbers say
Two occasions. 50.0% followed by a doubling within a year, against 3.6% on comparable days when the rule did not fire: a 46.4-point gap, positive, on 2 observations. That is the finding: on this specification the rule almost never fires, because after the 2020 halving the model-implied price sits far above the market. Two occasions cannot be scored.
What this does not mean
This is the weakest fit between a claim and our method on the site, and it is worth saying plainly. The coefficient is ours, not PlanB's, and the model line above is correspondingly high; only the exponent of 3.3 is his. Stock-to-Flow is not a trigger rule — it is a claim that a fitted relationship will continue to hold. A trigger test cannot assess that. The honest assessment of Stock-to-Flow is that its published forecasts for the 2021 cycle did not come to pass, which is a matter of record rather than of scoring, and that a model fitted to four halvings has four observations of the thing it claims to explain.
And it does not mean the threshold itself was chosen fairly. Every rule in this registry was specified by someone who had already seen bitcoin’s history, so the cut-off is fitted to the record it is being judged against. The interval above measures sampling variation between independent trials and nothing else: not the choice of threshold, not that cycle turns are few and related, and not that thirty-six rules were tested in total. Only the forward record, which fixes a rule before its answer is known, avoids this.
What would change the answer
Nothing we could compute here. Judging a fitted model needs out-of-sample forecasts made in advance and then checked, which is what the 2021 forecasts were.
If you cannot time it, the question changes
Most of the claims in this registry cannot be shown to beat a coin flip, and several are worse than one. That is not an argument against holding bitcoin. It is an argument that the decision worth spending effort on is how much to hold, not when to buy.
That is what How much bitcoin? is for: what fixed weights actually did in a conventional portfolio, and how much the answer depended on the year you started. For the valuation state today, and what has historically followed readings like it, see Where Things Stand.
Recomputed from the daily public snapshot by fetch/scorecard.py and published as scorecard.json. Method on the methods page; every claim in the scorecard.