By Marcus ReeveWhat Ten Years of Ether Priced in Bitcoin Actually Tell You
Quick answer: The ETH/BTC ratio closed at 0.0297 on October 15, 2026 (Bitfinex and OKX 0.02971; CoinGecko 0.02971). That is the 30th percentile of all daily closes since March 2016 and 80.2% below the record 0.1503 set on June 12, 2017.
Updated October 2026: latest ratio, year-to-date returns and current-swing levels refreshed.
On 15 October 2026, one Ether was worth 0.02971 Bitcoin. That is the entire ETH/BTC ratio: not a dollar price, not a market cap, just how much Bitcoin a single Ether buys. It sounds like a niche number. In practice it is one of the most-watched relative-value figures in crypto, because it decides whether holding Ethereum instead of Bitcoin was the right call.
Today's reading sits in the 30th percentile of every session since March 2016 — cheaper than 70% of the last decade, and 80.2% below the all-time high set on 12 June 2017. Ether has been cheap against Bitcoin for a long time now, and the obvious question is whether that means anything.
This is a data piece, not a forecast. Every number below is computed from daily closes and the method is spelled out at the end so you can reproduce it.
What the ETH/BTC ratio actually measures
The ratio is simply ETH's price divided by BTC's price. When it rises, Ether is outperforming Bitcoin; when it falls, Bitcoin is winning. Because both legs are crypto, the ratio strips out the thing that dominates every dollar chart — the overall crypto cycle — and leaves only the relative question.
That makes it the cleanest available read on a few things traders actually care about:
- Risk appetite inside crypto. Capital rotating out of Bitcoin and into Ether (and from there into smaller assets) historically shows up here first. Our reference on Bitcoin dominance cycles and when altseason starts tracks the same rotation from the dominance side.
- Whether "Ethereum is dead" is a price statement or a network statement. The ratio is a price statement. It says nothing about usage, fees, or the Ethereum roadmap.
- Relative-value entries. If you are going to hold crypto anyway, the ratio tells you which side of the pair the market is discounting.
What it does not measure is profit. That distinction turns out to matter enormously, and we come back to it below.
Every major ETH/BTC cycle since 2016
Below is the complete cycle table. Pivots are identified with a 45% reversal ZigZag on daily closes — a swing high is confirmed only once the ratio falls 45% from it, and a swing low only once it rises 45%. That threshold is deliberately coarse: it ignores ordinary chop and isolates the moves that actually redefined the range. Sixteen legs qualify.
| # | Direction | From | To | Days | Ratio | Ratio move | ETH in USD | BTC in USD |
|---|---|---|---|---|---|---|---|---|
| 1 | Down | 13 Mar 2016 | 26 Apr 2016 | 44 | 0.0352 → 0.0158 | −55.1% | −48.7% | +13.7% |
| 2 | Up | 26 Apr 2016 | 19 May 2016 | 23 | 0.0158 → 0.0337 | +113.5% | +98.1% | −7.2% |
| 3 | Down | 19 May 2016 | 18 Jun 2016 | 30 | 0.0337 → 0.0150 | −55.5% | −23.1% | +74.3% |
| 4 | Up | 18 Jun 2016 | 20 Sep 2016 | 94 | 0.0150 → 0.0237 | +58.3% | +26.2% | −20.8% |
| 5 | Down | 20 Sep 2016 | 28 Dec 2016 | 99 | 0.0237 → 0.0077 | −67.5% | −46.6% | +63.6% |
| 6 | Up | 28 Dec 2016 | 12 Jun 2017 | 166 | 0.0077 → 0.1503 | +1,851.1% | +4,980.2% | +161.8% |
| 7 | Down | 12 Jun 2017 | 07 Dec 2017 | 178 | 0.1503 → 0.0250 | −83.4% | +7.3% | +546.4% |
| 8 | Up | 07 Dec 2017 | 01 Feb 2018 | 56 | 0.0250 → 0.1129 | +351.6% | +149.4% | −44.7% |
| 9 | Down | 01 Feb 2018 | 29 Mar 2018 | 56 | 0.1129 → 0.0541 | −52.1% | −62.9% | −22.7% |
| 10 | Up | 29 Mar 2018 | 16 May 2018 | 48 | 0.0541 → 0.0847 | +56.6% | +84.0% | +17.5% |
| 11 | Down | 16 May 2018 | 12 Dec 2018 | 210 | 0.0847 → 0.0259 | −69.4% | −87.0% | −57.6% |
| 12 | Up | 12 Dec 2018 | 05 Jan 2019 | 24 | 0.0259 → 0.0405 | +56.2% | +72.3% | +10.6% |
| 13 | Down | 05 Jan 2019 | 06 Sep 2019 | 244 | 0.0405 → 0.0164 | −59.5% | +7.0% | +163.4% |
| 14 | Up | 06 Sep 2019 | 08 Dec 2021 | 824 | 0.0164 → 0.0879 | +435.8% | +2,522.3% | +390.1% |
| 15 | Down | 08 Dec 2021 | 21 Apr 2025 | 1,230 | 0.0879 → 0.0181 | −79.5% | −64.4% | +73.2% |
| 16 | Up | 21 Apr 2025 | 24 Aug 2025 | 125 | 0.0181 → 0.0421 | +133.2% | +203.2% | +30.0% |
Four things fall straight out of that table.
Down legs last roughly twice as long as up legs. The median up leg ran 75 days for a +123% gain; the median down leg ran 139 days for a −64% loss. Ether's outperformance arrives in short, sharp bursts; its underperformance is slow.
The 2021–2025 decline was a different animal. Leg 15 lasted 1,230 days — longer than legs 1 through 12 combined — and took the ratio down 79.5%. Nothing else in the record comes close for duration. If your mental model of ETH/BTC was formed after December 2021, you have only ever seen one regime.
The record high and record low are six months apart. The ratio bottomed at 0.0077 on 28 December 2016 and peaked at 0.1503 on 12 June 2017 — a 19.5x move in 166 days. Any claim about where the ratio "should" trade has to survive the fact that it has traded across a 19-fold range inside half a year.
Leg 7 is the one to memorise. Between June and December 2017 the ratio fell 83.4% — the deepest collapse on record — while Ether rose 7.3% in dollars. Everyone who held ETH through that period made money and underperformed, simultaneously. Which brings us to the most useful finding in this dataset.
A falling ratio is not the same as losing money
This is where ETH/BTC commentary often goes wrong. A declining ratio is treated as evidence that Ether is a bad hold, when it only ever says Ether is a worse hold than Bitcoin.
We measured every rolling 365-day window since 2016. In 1,912 of them the ratio fell. In 891 of those — 47% — Ether was still up in dollar terms over the same window. Nearly half the time, "ETH is bleeding against BTC" and "ETH holders made money" were both true.
The calendar-year view makes it concrete:
| Year | ETH/BTC ratio | ETH in USD | BTC in USD |
|---|---|---|---|
| 2016 | −71.7% | −28.5% | +134.0% |
| 2017 | +541.8% | +9,107.3% | +1,324.5% |
| 2018 | −33.6% | −81.5% | −72.2% |
| 2019 | −49.5% | −4.8% | +88.2% |
| 2020 | +41.6% | +468.5% | +301.5% |
| 2021 | +213.0% | +399.8% | +59.7% |
| 2022 | −9.1% | −67.5% | −64.2% |
| 2023 | −25.4% | +90.9% | +155.5% |
| 2024 | −33.9% | +46.5% | +121.5% |
| 2025 | −4.9% | −11.1% | −6.4% |
| 2026 YTD* | −12.4% | −17.9% | −6.3% |
Through 15 October 2026.
In 2023 and 2024 the ratio fell by a quarter and a third respectively — and Ether returned +90.9% and +46.5% in dollars. Two of the better years an ETH holder has had were, on the ratio chart, part of an unbroken slide.
The other half of the picture is less comfortable: the ratio has fallen in 8 of the last 11 calendar years. It has risen in three. This is not a mean-reverting oscillator around a fixed centre; over the full sample it is a downtrending series with occasional large countertrend rallies.
The percentile rule: what the ratio's level has predicted
If the ratio is a relative-value measure, the honest test is whether its level has predicted relative performance. So we ran it directly: take every trading day since 2016, rank that day's ratio against the entire history, sort into five percentile buckets, then measure the following 365 days.
| Ratio percentile at entry | Windows | ETH beat BTC over next year | Median ETH +1y | Median BTC +1y |
|---|---|---|---|---|
| Cheapest 0–20% | 758 | 100% | +873.2% | +284.6% |
| 20–40% | 543 | 60% | +62.8% | +68.4% |
| 40–60% | 608 | 30% | +25.5% | +55.6% |
| 60–80% | 758 | 23% | +21.6% | +65.3% |
| Richest 80–100% | 757 | 10% | −31.4% | +23.0% |
The relationship is monotonic and it is steep. Buying Ether when the ratio sat in its cheapest fifth beat Bitcoin over the next year in every single one of 758 windows. Buying it in the richest fifth beat Bitcoin 10% of the time and lost money outright 61% of the time.
The obvious objection is that the cheapest bucket is stuffed with 2016 dates that happen to precede the 2017 mania. It is a fair objection, so we re-ran the test twice more with the early years removed:
| Sample | Cheapest 20% beat BTC | Richest 20% beat BTC |
|---|---|---|
| All sessions (2016–2025 entries) | 100% | 10% |
| Excluding 2016–17 (entries from 2018) | 100% | 11% |
| Modern era only (entries from 2020) | 100% | 15% |
| — median ETH +1y, ex-2016/17 | +117.3% | −39.8% |
The magnitudes shrink a lot once 2017 is removed — the ex-2016/17 median drops from +873% to +117% — but the direction survives every cut. Cheap ratio, ETH outperforms; rich ratio, ETH underperforms. That is the finding.
Where the ratio sits right now
The 15 October 2026 close of 0.02971 lands in the 30th percentile — the 20–40% bucket. Historically that bucket has been close to a coin flip: Ether beat Bitcoin over the following year 60% of the time across the full sample, 57% excluding 2016–17. Better than even, nothing like the cheapest bucket's clean sweep.
The recent path matters too. The ratio peaked at 0.0421 on 24 August 2025, bottomed at 0.02578 on 6 June 2026 (−38.8%), rose to 0.03262 on 12 September 2026 (+26.5% off the low), and on 15 October stood +15.3% above the June low and 29.5% below the August 2025 peak. Neither move has reached the 45% threshold, so the current leg is an open, unconfirmed swing, not a resolved cycle. Over the last 30 days the ratio is −6.3%, with Ether at +1.8% against Bitcoin's +8.6%.
For dollar context: Ether closed at $2,443 and Bitcoin at $82,233 on 15 October 2026. Both are down year to date (ETH −17.9%, BTC −6.3%) after recovering from summer lows, in a market shaped by a Fed that raised rates in September 2026 and by an oil-driven risk premium, forces we traced in Bitcoin's push back above $66K and in the regulatory reset from the CLARITY Act.
Three levels worth marking
| Level | What it is | Why it matters |
|---|---|---|
| 0.02578 | 6 June 2026 low | The floor of the current leg; a daily close below reopens the 2025 lows at 0.0181 |
| 0.0297 | 15 Oct 2026 close | 30th percentile — inside the "modestly cheap" 20–40% bucket |
| 0.0421 | 24 Aug 2025 peak | A close above confirms leg 16 resumed and ends the 2025–26 downswing |
Above 0.0328 the ratio leaves the 20–40% bucket for the middle third of its range, where the historical edge flips against Ether; above 0.0421 leg 16 resumes outright. Below 0.0251 it drops back into the cheapest bucket — the only one that has never failed to beat Bitcoin over a year, on a sample of 758 windows — and below 0.0181 it sets a new decade low.
How to track this without watching a chart all day
The ratio moves slowly and then all at once, which is exactly the sort of thing worth automating rather than staring at.
- SimianX AI runs multi-agent analysis on both legs of the pair — technical, fundamental, news and a decision agent — so you get a read on ETH and BTC from the same framework rather than two disconnected chart windows.
- SimianX Crypto Leaderboard — dozens of AI models from several providers trading real positions, so you can see which models are actually positioned in ETH versus BTC rather than guessing at sentiment.
- Live Command Room — start a session on ETH-USDT or BTC-USDT and watch the agents work through the same levels listed above.
- Autopilots re-run the analysis on your schedule and notify you when something like "ratio closes above 0.0421" actually happens.
- Market Pulse flags 52-week and 30-day extremes across crypto and equities as they print.
If you want to know which models have historically been right on calls like this, we published the full scoreboard in Which AI Model Is the Best Trader?, and a study of how they behave in drawdowns in Do AI Models Panic-Sell in a Crash?.
Frequently asked questions
What is the ETH/BTC ratio today?
0.0297 at the 15 October 2026 daily close, with Ether at $2,443 and Bitcoin at $82,233. The ratio is down 12.4% in 2026 and sits in the 30th percentile of all closes since March 2016.
What is a good ETH/BTC ratio to buy at?
Historically, the cheapest fifth of the range (below about 0.0250 on the tested sample) is the only bucket that beat Bitcoin in 100% of one-year windows. The current 0.0297 is in the 20–40% bucket, which beat Bitcoin about 57–60% of the time. That is a base rate, not a guarantee.
What was the all-time high for ETH/BTC?
0.15029 on 12 June 2017, on a daily-close basis. The 15 October 2026 close is 80.2% below that level, 3,412 days later. The record low is 0.0077 on 28 December 2016.
Does a falling ETH/BTC ratio mean Ethereum is failing?
No. It means Ether is underperforming Bitcoin. In 47% of the rolling one-year windows where the ratio fell, Ether still rose in dollars; in 2023 and 2024 it returned +91% and +46% while the ratio dropped 25% and 34%.
Has the ETH/BTC ratio ever recovered from a decline this long?
The 2021–2025 decline ran 1,230 days and ended with a 133% rally in 125 days. Before that, the 2019 decline ran 244 days and was followed by an 824-day, +436% advance. Long declines have ended, but not on a schedule visible in the chart.
Is the ETH/BTC ratio mean-reverting?
Not around a fixed level. It has fallen in 8 of the last 11 calendar years, counting 2026 to date, so no single number works as "fair value". The percentile framework measures the ratio against its realised range instead.
How does this compare to Bitcoin's own cycles?
Bitcoin's drawdown-and-recovery record is in Every Bitcoin Crash Over 50%, its seasonal pattern in Bitcoin Quarterly Returns, and its supply-cycle framework in Bitcoin Halving Cycles.
Methodology and limitations
Prices are daily closes from the Bitfinex public API for ETH/BTC, ETH/USD and BTC/USD, covering 9 March 2016 through 30 July 2026 — 3,789 sessions with a complete ratio series. Bitfinex was chosen because it carries the longest continuous ETH/BTC history of the major venues; the July reading was cross-checked against OKX (0.02963) and CoinGecko (0.029631). The October 2026 update extends the series to 15 October 2026 (3,866 sessions); that day's close of 0.02971 matches OKX (0.02971) and CoinGecko's 00:00 UTC 16 October snapshot (0.029706) to four decimal places. The cycle legs and percentile-bucket tests above still use entries through mid-2025, the last dates with a complete forward year at the time of the original analysis.
Cycle pivots use a 45% reversal ZigZag on closes. Percentile buckets rank each session's ratio against the full sample and measure the subsequent 365 calendar days; windows without a complete forward year are excluded, which is why the entry sample ends in mid-2025.
Three limitations worth stating plainly. Overlapping windows — consecutive daily entries share most of their forward year, so the 758 windows in a bucket are nowhere near 758 independent observations; treat the percentages as descriptive, not as a significance test. Regime dependence — the cheapest bucket draws heavily on 2016, 2019, 2020 and 2025, four very different markets. Survivorship of the framing — the ratio's level is one input among many; the Merge, the ETF approvals, and the regulatory shifts of 2026 all changed the underlying assets in ways a price ratio cannot capture.
None of this is investment advice. It is a reference table with its workings shown.
Related Reading
- Bitcoin Dominance Cycles: When Altseason Starts 2017-2026
- Every Bitcoin Crash Over 50%: Recovery Times Since 2011
- BTC vs ETH vs SOL: Best Buy Zones in the 2026 Crypto Dip
- Bitcoin Quarterly Returns: Full History Table 2013–2026
- Bitcoin Halving Cycles: Complete Returns Reference 2012-2028
- Bitcoin After Every Fed Rate Cut: 2019-2026 Reference
- Every Major Stablecoin Depeg: A Complete Reference Table
- Bitcoin One Year After Its $126K Peak vs Every Past Cycle
- More reference tables in the crypto cycles topic hub
- Browse all SimianX market reference tables
References
- Bitfinex — public candles API used for all price history
- OKX — ETH-BTC spot market, used for cross-validation
- CoinGecko — independent ETH and BTC price reference
- Ethereum.org — official Ethereum documentation and roadmap
- Etherscan — Ethereum network explorer and on-chain statistics
- Investopedia — relative strength and ratio analysis explained
