Research · Relative Value
Bitcoin Relative Value
What Bitcoin buys relative to other assets. Bitcoin is compared with gold, silver, the S&P 500, Nasdaq, ether and Solana using historical price ratios. The page also tests simple rotation rules at three lookbacks, so one favourable parameter cannot become the headline, and shows how Bitcoin's relationship with other assets changes over time.
Current readings
Computed from the daily public snapshot for 14 October 2026, published 2026-10-15 11:22 UTC. The charts and tables below are drawn in the browser from the same file; these figures are the ones they show.
| Price | $83,094 | daily close |
|---|---|---|
| Price vs 200-day average | 15.2% | 200-day $72,129 |
| 90-day range | $62,760 – $86,608 |
WHAT THE MEASUREMENTS SAY
One bitcoin buys 18.6 ounces of gold today, against 0.002 grams in 2010-08 — and 50% below its peak in 2024-12. Rotation rules between bitcoin and its rivals do not survive an honest test: 1 of 9 momentum lookbacks beat holding, the others lose, and mean reversion wins only 2 of its 18 tests, all against ether. Right now bitcoin is trading like money.
None of this is a recommendation. A ratio says what one asset has done relative to another; it does not say what either will do, and a rotation rule that wins at one lookback and loses at the next is evidence about the fragility of rotation rules, not about which side to hold.
What one bitcoin has bought, since inception
Who won each calendar year
| Year | vs gold | vs S&P 500 | vs ether |
|---|---|---|---|
| 2026 so far | other -10% | other -19% | bitcoin +11% |
| 2025 | other -44% | other -21% | bitcoin +9% |
| 2024 | bitcoin +78% | bitcoin +81% | bitcoin +44% |
| 2023 | bitcoin +122% | bitcoin +109% | bitcoin +37% |
| 2022 | other -66% | other -59% | bitcoin +12% |
| 2021 | bitcoin +138% | bitcoin +81% | other -65% |
| 2020 | bitcoin +138% | bitcoin +157% | other -34% |
| 2019 | bitcoin +65% | bitcoin +58% | bitcoin +53% |
| 2018 | other -75% | other -74% | bitcoin +49% |
| 2017 | bitcoin +1563% | bitcoin +1432% | other -78% |
| 2016 | bitcoin +80% | bitcoin +77% | – |
| 2015 | bitcoin +37% | bitcoin +23% | – |
| 2014 | other -56% | other -62% | – |
| 2013 | bitcoin +8126% | bitcoin +4594% | – |
| 2012 | bitcoin +269% | bitcoin +231% | – |
| 2011 | bitcoin +1099% | bitcoin +1311% | – |
- The long recordbitcoin gained on gold in 11 of 16 calendar years and on the S&P 500 in 11 of 16
- The recent recordgold has now won two consecutive years, which had not happened before in bitcoin's history
- What a year meansa calendar year is one observation, and there are 16 of them; the table is a record, not a forecast
Rotation rules, tested at three lookbacks
Bitcoin against gold · 2010-08 to 2026-09 · 17 switching events at the 12-month rule
| from 2017-01 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 88.2× -75% dd | 88.2× -75% dd | 88.2× -75% dd |
| Hold the denominator | 3.6× -19% dd | 3.6× -19% dd | 3.6× -19% dd |
| 50/50, rebalanced monthly | 28.7× -48% dd | 28.7× -48% dd | 28.7× -48% dd |
| Momentum | 71.8× -63% dd · 20 sw | 88.7× -59% dd · 13 sw | 30.0× -77% dd · 9 sw |
| Mean reversion | 4.1× -60% dd · 20 sw | 3.4× -63% dd · 13 sw | 10.3× -36% dd · 9 sw |
| from 2020-01 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 9.7× -72% dd | 9.7× -72% dd | 9.7× -72% dd |
| Hold the denominator | 2.8× -19% dd | 2.8× -19% dd | 2.8× -19% dd |
| 50/50, rebalanced monthly | 6.6× -46% dd | 6.6× -46% dd | 6.6× -46% dd |
| Momentum | 6.9× -63% dd · 17 sw | 11.7× -40% dd · 10 sw | 7.4× -63% dd · 6 sw |
| Mean reversion | 3.6× -60% dd · 17 sw | 2.2× -63% dd · 10 sw | 3.5× -36% dd · 6 sw |
Bitcoin against S&P 500 · 2010-08 to 2026-09 · 15 switching events at the 12-month rule
| from 2017-01 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 88.2× -75% dd | 88.2× -75% dd | 88.2× -75% dd |
| Hold the denominator | 3.4× -20% dd | 3.4× -20% dd | 3.4× -20% dd |
| 50/50, rebalanced monthly | 25.9× -50% dd | 25.9× -50% dd | 25.9× -50% dd |
| Momentum | 66.5× -64% dd · 15 sw | 74.0× -65% dd · 13 sw | 29.9× -82% dd · 7 sw |
| Mean reversion | 4.2× -60% dd · 15 sw | 3.8× -60% dd · 13 sw | 9.7× -43% dd · 7 sw |
| from 2020-01 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 9.7× -72% dd | 9.7× -72% dd | 9.7× -72% dd |
| Hold the denominator | 2.3× -20% dd | 2.3× -20% dd | 2.3× -20% dd |
| 50/50, rebalanced monthly | 5.8× -50% dd | 5.8× -50% dd | 5.8× -50% dd |
| Momentum | 5.8× -64% dd · 12 sw | 10.4× -45% dd · 10 sw | 8.2× -61% dd · 4 sw |
| Mean reversion | 3.7× -60% dd · 12 sw | 2.1× -60% dd · 10 sw | 2.7× -43% dd · 4 sw |
Bitcoin against ether · 2016-05 to 2026-10 · 12 switching events at the 12-month rule
| from 2018-05 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 9.9× -72% dd | 9.9× -72% dd | 9.9× -72% dd |
| Hold the denominator | 3.8× -84% dd | 3.8× -84% dd | 3.8× -84% dd |
| 50/50, rebalanced monthly | 7.4× -73% dd | 7.4× -73% dd | 7.4× -73% dd |
| Momentum | 4.9× -82% dd · 20 sw | 9.1× -79% dd · 11 sw | 6.3× -83% dd · 6 sw |
| Mean reversion | 7.1× -81% dd · 20 sw | 4.0× -75% dd · 11 sw | 5.8× -72% dd · 6 sw |
| from 2020-01 | 6 months | 12 months | 24 months |
|---|---|---|---|
| Hold bitcoin | 10.1× -72% dd | 10.1× -72% dd | 10.1× -72% dd |
| Hold the denominator | 16.4× -72% dd | 16.4× -72% dd | 16.4× -72% dd |
| 50/50, rebalanced monthly | 14.7× -72% dd | 14.7× -72% dd | 14.7× -72% dd |
| Momentum | 6.8× -82% dd · 17 sw | 14.8× -79% dd · 9 sw | 17.0× -72% dd · 5 sw |
| Mean reversion | 22.8× -59% dd · 17 sw | 10.8× -63% dd · 9 sw | 9.6× -72% dd · 5 sw |
- The parameter decides the resultacross the momentum tests from 2017, 1 of 9 lookback choices beat simply holding bitcoin; the same rule with a different window loses. That is what specification search looks like from the inside, and it is why all three are shown
- Mean reversion mostly losesbuying the lagging side trails holding in 16 of 18 tests; the two it wins are both against ether, where the result again depends on the lookback and the start date
- What a rotation actually buyswhere momentum beats holding, it also cut the worst drawdown; a reader who cares about the path and not the endpoint may still prefer it, and should know it cost about 14 taxable switches per decade
What Bitcoin is trading like
- The two scales disagree, and that is the readingover the last 90 days of daily returns bitcoin reads as money; over the last 24 months of monthly returns it reads as mixed (0.43 with equities, -0.07 with gold). A short window is more current and more easily moved by a single episode; the long window is what an allocator would experience
- The long recordon the monthly basis the equity correlation peaked at 0.59 in July 2022 and the gold correlation at 0.64 in March 2024; over the full history bitcoin has co-moved with equities far more often than with gold
- The daily record, where it existsthe equity correlation peaked at 0.62 (Feb 11, 2026) and bottomed at -0.34 (Aug 9, 2019); it averaged 0.52 across 2022. Equity history begins Jan 12, 2017 because the free FRED series carries ten years
- How it is computedthe chart uses monthly average prices for all three assets (bitcoin from the daily close series, gold from the World Bank, the S&P 500 from the Shiller series), so both lines run from 2012 on identical footing. The 90-day reading uses daily closes on both sides; its gold leg exists only since tokenised gold listed on Coinbase in May 2025, which is why no long daily gold line is drawn
- How much of the wiggle is realeach point on the monthly chart rests on 24 observations, so its standard error is about 0.20; a move of a quarter or so is inside the noise, and only sustained shifts of half or more — equities from near zero to 0.5 over 2013 to 2020, gold from −0.4 to +0.6 over 2016 to 2024 — should be read as changes in regime
- What it cannot tell youcorrelation is a description of a window, not a property of the asset, and the two scales can point different ways at the same moment, as they may today
Inside crypto
Method
Long-history denominators are monthly averages: gold and silver from the World Bank Commodity Markets “Pink Sheet” (public domain; gold falls back to the datahub mirror of the same series if the World Bank file is unavailable), and the S&P 500 from Robert Shiller's monthly series (public domain, mirrored on GitHub), used alone for the long ratio so that a partial current month from the daily feed cannot sit beside full-month averages; FRED's daily closes serve the correlation panel. Bitcoin is the monthly mean of the daily close over the same months, so both sides of every long ratio are like-for-like averages. Daily denominators for the recent period are Coinbase Exchange closes for ether, Solana and tokenised gold (PAXG) and FRED daily closes for the S&P 500 and Nasdaq; the correlation panel uses these. Every source is keyless and named per series in the snapshot's provenance block.
Rotation tests rebalance monthly on the prior month's ratio against its trailing mean over 6, 12 or 24 months, charge 0.2% per switch, ignore tax, and report the ending multiple and maximum drawdown from two start dates. The number of switching events is the count of times the ratio crossed its trailing mean, and it is stated beside each table because that, not the number of months, is the sample. The events are not independent draws: consecutive crossings share the same trend episode.
What this page tells you, and what it does not
Claimed: these are the historical records of each ratio and each rule, reproducible from the snapshot. Not claimed: that any rotation rule is a strategy. The clearest result on this page is that the answer depends on the lookback, which is what specification search looks like from the inside; the page shows all three so the reader can see it rather than be shown the one that worked. On the data to 14 October 2026, mean-reversion rotation — buying whichever side has lagged, the logic behind every “alt season” narrative — trails holding under every lookback tested.
Sources: the Crypto Exponentials daily snapshot (Blockchain.com price; World Bank Commodity Markets; FRED; Coinbase Exchange). Nothing here is investment advice.