Power law vs stock-to-flow
Stock-to-flow ties Bitcoin's value to its scarcity, so its model price jumps after every halving. Its widely cited claim was a $100,000 average for the 2020–2024 halving period. The actual average close over that period was $32,843. Today the S2F formula implies about $723K against a price of $81,462. The power law makes no halving claim at all and puts today's trend at $143,791; its weakness is that it is a curve fit to time, not a theory of value.
What stock-to-flow says
In March 2019 the pseudonymous analyst PlanB published “Modeling Bitcoin’s Value with Scarcity”. Stock-to-flow (SF) is the existing supply divided by the new supply mined in a year; gold has a high ratio, and every halving doubles Bitcoin's. PlanB fitted a regression of market value on SF and reported an R² of 95%:
ln(market value) = 3.3 × ln(SF) + 14.6
He wrote that the predicted market value after the May 2020 halving was $1 trillion, about $55,000 per bitcoin. In later versions the widely quoted figure became a $100,000 average for the 2020–2024 halving period, while a cross-asset variant (S2FX) called for $288,000 (Cointelegraph, June 2021). A related “floor model” of $98,000 for November 2021 and $135,000 for December 2021 also missed: the year closed below $100,000.
What actually happened
We recomputed PlanB's published regression from Coin Metrics supply data, with the flow taken as the coins issued over the previous 365 days. As in the original, the first million coins are left out as lost. This is a reconstruction: PlanB used 111 monthly data points (December 2009 to February 2019) and his price version had different parameters, so the exact numbers differ from his, but the shape is the same.
| Date | Stock-to-flow | S2F model price | Actual close | Actual ÷ model |
|---|---|---|---|---|
| 22 March 2019 | 24.8 | $4,956 | $3,984 | 0.80× |
| 1 January 2020 | 25.3 | $5,135 | $7,171 | 1.40× |
| 1 January 2021 | 39.1 | $21,096 | $29,381 | 1.39× |
| 1 January 2022 | 54.4 | $61,813 | $47,560 | 0.77× |
| 1 January 2023 | 54.9 | $62,707 | $16,607 | 0.26× |
| 1 January 2024 | 55.1 | $62,211 | $44,049 | 0.71× |
| 1 January 2025 | 86.9 | $277,078 | $94,456 | 0.34× |
| 1 January 2026 | 114.3 | $679,843 | $88,684 | 0.13× |
| 19 October 2026 | 116.7 | $723,173 | $81,918 | 0.11× |
Over the 2020–2024 halving period the average close was $32,843, and the highest close was $73,082. Neither the $100,000 average nor the reconstructed model average of $51,370 was reached. After the 2024 halving the gap widened: the model rose with the doubled ratio, the price did not.
| Period | Average actual | S2F model average | Power law trend average (fit made at the start) |
|---|---|---|---|
| 11 May 2020 – 19 Apr 2024 | $32,843 | $51,370 | $35,142 |
| Since 20 Apr 2024 | $83,936 | $503,470 | $106,186 |
For comparison, the last column is a power-law fit made at the start of each period with only the data known then. It was not perfect either, but its average sat closer to what the price did, because it never expected a step change.
Why the two models diverge
The power law describes price as a function of the network's age:
log10(price) = a + n × log10(days since 3 Jan 2009)
Time grows smoothly, so the trend grows smoothly, and its growth rate falls every year. Stock-to-flow is driven by the issuance schedule, which halves every four years, so the model price must roughly multiply by about ten each time the ratio doubles (2 to the power 3.3). After three halvings that step is the problem: each one demands an order of magnitude more money than the last. Critics pointed out that scarcity alone says nothing about demand. The power law has its own weaknesses: its exponent drifts, it has no causal mechanism built into the regression, and its bands were chosen with hindsight.
What each is good for
Stock-to-flow made a sharp, falsifiable claim, and that is to its credit; the claim did not hold. The power law makes a weaker claim, a slowly bending trend with a wide band around it, and that weaker claim has held up better so far. Neither is a price forecast for next month. Today the price is 0.11× the S2F model and 0.57× the power-law trend.
Frequently asked questions
What was the stock-to-flow prediction for 2020–2024?
Is the stock-to-flow model broken?
Which is better, stock-to-flow or the power law?
Is the power law also stock-to-flow in disguise?
More research
Sources
- PlanB, “Modeling Bitcoin’s Value with Scarcity”, Medium, 22 March 2019
- Cointelegraph, “$288K BTC price still in play, says PlanB”, 12 June 2021
- FXStreet, “Bitcoin stock-to-flow model invalidated as BTC closes 2021 below $100,000”, 1 January 2022
- Coin Metrics, community API, daily PriceUSD and SplyCur
Numbers updated 20 October 2026 · Coin Metrics daily closes through 19 October 2026 · price $81,462 · trend $143,791