Define the test
A proper comparison uses the same start and end dates and the same currency. CoinMarketCap calls it an altcoin season when at least 75% of its eligible top 100 coins have outperformed BTC over the trailing 90 days. That is a breadth test—not a prediction and not a claim that every altcoin will win.
Define the test
CoinMarketCapBitcoin share of total crypto market
CoinGecko records Bitcoin market share falling from 87.2% to 38.9% during 2017 and from 69.5% to 38.2% during 2021. Those are striking shifts in the structure of the market. Market share is not a direct investment return: new token issuance, valuations and falling BTC prices can also move the denominator. Treat these dates as historical context, not a reconstructed top-100 backtest.
Only ETH/BTC · not a top-100 index · CoinGecko
2021: four periods, with separate sources and dates
For 2021 there is a direct price comparison. CoinGecko reports ETH +403% for the calendar year against BTC +62.6%. CoinMarketCap cites a separate February–May 2021 interval in which large-cap altcoins returned about 174% against 2% for BTC. The intervals, baskets and sources differ; they both show that outperformance existed, but neither says a randomly selected coin beat BTC.
January–March
CoinGecko’s fixed top-five non-stablecoin group as of 1 January offers a useful same-quarter comparison. BTC rose 103%, ETH 159%, DOT 294%, ADA 555% and BNB 710%. Three named altcoins decisively beat both BTC and ETH. Fixing membership at the start helps avoid selecting only winners in hindsight; it still leaves out most tokens and trading costs. The growth of alternative smart-contract networks was already visible before the May mania.
February–May
CoinMarketCap’s retrospective comparison of large-cap altcoins reports approximately +174% against BTC +2% over this different interval. This is the closest broad group comparison here, but its definition and membership are not interchangeable with the Q1 group or the full 90-day index. The figures show a powerful rotation in a defined group, not an entry date or a guaranteed return for a portfolio bought at the peak.
April–September
The trend was uneven. CoinGecko reports Q2 DOGE +366% while BTC fell 40%; DOGE is a selected extreme, not a portfolio return. BTC was 55% below its April peak at one point in June, and monthly spot exchange volume fell about 60% from May to June. After that sell-off, Q3 BTC recovered 25% and ADA gained 53%. So the same calendar year contained speculative acceleration, a sharp reversal and a new relative winner.
Year-end and aftermath
Across the calendar year ETH returned +403% versus BTC +62.6%, and BTC’s market share ended near 38.2% from 69.5% at the start. CoinGecko documents DeFi market capitalisation growing from roughly $20 billion to $150 billion in 2021, another sign of capital flowing beyond BTC. Still, a calendar-year winner could have had a very different result when bought in November. The subsequent 2022 market decline is a reminder that outperformance did not protect against later losses.
2021: two distinct price comparisons · Q1 2021
February–May 2021
Q2 2021
Q3 2021
2021
ETH/BTC at the UTC daily close
Our second lens uses 993 matching Binance Spot daily closes for BTCUSDT and ETHUSDT from 1 January 2024 through 19 September 2026. We divide ETH by BTC at each UTC close, then set the first ratio to 100. A rising line means ETH gained relative to BTC, even if both fell in dollar terms. This is one major altcoin, not the broad market.
How to read these charts
- The first chart starts at 100 on 1 January 2024. Below 100 means ETH lost ground against BTC since that date.
- The second chart compares ETH with BTC over the previous 90 days. Above 0% means ETH did better. For example, if ETH fell 5% and BTC fell 10%, ETH was stronger even though both lost value in dollars.
- Each daily 90-day window shares most dates with the next. A cluster is one extended market regime, not repeated independent evidence.
Only ETH/BTC · not a top-100 index · Binance Spot
ETH versus BTC · trailing 90 days
For each date after the first 90 days, we compare the ETH/BTC ratio with its value 90 calendar days earlier. ETH led BTC in 249 of 903 overlapping windows, or 27.6%. The longest uninterrupted run of positive 90-day comparisons lasted 132 days, from 26 June to 4 November 2025. Over the complete calendar years ETH/BTC still fell 33.0% in 2024 and 4.6% in 2025; from 1 January through 19 September 2026 it fell 4.2%. Short intervals and whole-year results answer different questions.
| Period | Bitcoin | Ethereum | ETH/BTC |
|---|---|---|---|
| 2024-01-01 → 2024-12-31 | +111.81% | +41.91% | -33.00% |
| 2025-01-01 → 2025-12-31 | -7.34% | -11.57% | -4.56% |
| 2026-01-01 → 2026-09-19 | -8.54% | -12.37% | -4.18% |
What the charts cannot prove
A falling BTC dominance line does not prove that most coins produced a positive USD return. A strong ETH/BTC chart does not prove that the other 99 coins outperformed. These 90-day observations also overlap heavily, so 249 positive windows are not 249 independent trades. Token lists change through time; judging yesterday’s market with today’s survivors would bias a historical test.
A practical reading order
Start with the question: individual coin, fixed basket, or broad market? Then match dates, currency, token eligibility and holding period. Read a breadth index alongside relative price, and look at absolute returns and drawdowns before calling any period investable.
Historical outperformance does not predict the next cycle. This research is educational and does not rank or recommend any token.
Original sources
Our daily source is Binance Spot BTCUSDT and ETHUSDT, using UTC daily closes. Relative 90-day return = (ETH/BTC today ÷ ETH/BTC 90 days earlier − 1) × 100. The windows overlap. The 2017/2021 dominance and 2021 price facts come from the linked publications, not from the Binance sample. CoinMarketCap’s 75% threshold is quoted as a definition; we did not reconstruct its full historical index.