# Golden cross on the top-20 coins, 2020–2025: what the backtest showed > A backtest of the SMA 50/200 golden cross on the top-20 coins on Binance spot, 2020–2025: profit on 9 of 19, ahead of holding on 3 of the 12 that rose. Research · Janorium · Published 2026-10-14, updated 2026-10-15 Web page: https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025 On Binance spot in 2020–2025 the daily SMA 50/200 golden cross made money on 9 of 19 large coins and beat simply holding on only 3 of the 12 that rose; 15 of its runs fell more than 80% from a peak. On perpetual futures the same rule did worse on every coin. ![Six headline numbers: 9 of 19 coins in profit on spot, 3 of 12 rising coins beaten, 15 of 19 drawdowns deeper than 80%, 44 of 124 trades in profit, 19 of 19 coins worse on perpetuals with funding, 47,219 USDT of funding on the ETH perpetual](https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025/scorecard.en.svg 'One rule, 20 coins, six years — every number is from the runs below') ## Setup - **Rule:** long while the daily SMA 50 is above the daily SMA 200 — entered at the golden cross, or on the first bar once the averages are warmed up if 50 is already above 200 — and flat after the death cross. No stop, no target, one position. - **Coins:** the 20 coins with the highest Binance USDT-M turnover in the 30 days before 2021-01-01, picked by the platform as of that date. 19 of them have spot history on the platform; EOS, delisted since, does not. - **Market and period:** Binance spot, 4h candles from which the daily averages are built, 2020-01-01 → 2026-01-01. For comparison, the same rule on Binance USDT-M perpetuals at 1x. - **Costs:** 95% of equity per trade, 0.05% fee and 0.05% slippage on each side, fills at the next bar's open, 10,000 USDT per coin. Funding is charged on perpetuals. - **Runs:** one strategy — spot #159–#177, perpetuals with funding #119–#138 and with funding off #139–#158. Nothing was tuned. ## Golden cross vs Buy & Hold ![Dot plot of the golden cross return against Buy & Hold for each of the 19 coins on spot, log scale](https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025/returns.en.svg 'Teal — the cross, indigo — holding the same coin over the same run') Holding won on 9 of the 12 coins that rose. The cross got ahead on ADA, ETC and THETA, each on the back of one trade opened in 2020 — THETA's, opened on 2020-07-19, made +2,088.74%. On coins that collapsed it lost less than holding on 6 of 7: FIL -85.79% against -99.20%. **Backtest: [Golden Cross 50/200 daily — top-20 study](https://janorium.com/backtests/159)** BTCUSDT · 4h · Spot · 2020-01-01 → 2026-01-01 | Return | Max drawdown | Sharpe | Win rate | Trades | | --- | --- | --- | --- | --- | | +529.12% | 57.11% | 0.96 | 66.7% | 6 | *BTC on spot: +529.12% for the cross, +859.56% for holding* **Backtest: [Golden Cross 50/200 daily — top-20 study](https://janorium.com/backtests/168)** BNBUSDT · 4h · Spot · 2020-01-01 → 2026-01-01 | Return | Max drawdown | Sharpe | Win rate | Trades | | --- | --- | --- | --- | --- | | +2,312.15% | 76.23% | 1.12 | 57.1% | 7 | *BNB, the best run: +2,312.15% — against +4,795.10% for holding* **Backtest: [Golden Cross 50/200 daily — top-20 study](https://janorium.com/backtests/173)** THETAUSDT · 4h · Spot · 2020-01-01 → 2026-01-01 | Return | Max drawdown | Sharpe | Win rate | Trades | | --- | --- | --- | --- | --- | | +472.64% | 90.59% | 0.80 | 33.3% | 6 | *THETA, the one big win over holding: +472.64% against +7.47%* ## When it made money ![Timeline of all 124 spot trades of the 19 runs, coloured by outcome](https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025/timeline.en.svg '13 of 19 coins held nothing from February to December 2022') 44 of 124 trades closed in profit, and the big ones came from one window: trades opened in 2020 won 12 of 16 — the five largest of the study among them. Trades opened in 2021 won 1 of 22, in 2022 — 1 of 9, and from 2023 on — 30 of 77: the averages turned late both ways. ## Drawdowns ![Max drawdown of each of the 19 spot runs, with a line at 80%](https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025/drawdown.en.svg 'The shallowest drawdown was BTC, 57.11%') The trend filter did not keep the equity safe: 15 of 19 runs fell more than 80% from a peak. A 200-day average turns after the fall — by the death cross, much of a rally is already gone. ## On perpetual futures ![Each perpetual run with funding off and on, with the funding paid in USDT](https://janorium.com/articles/golden-cross-top-20-crypto-2020-2025/funding.en.svg 'Hollow — the same run with funding off, filled — with funding') The same rule on USDT-M perpetuals did worse than on spot on all 19 coins. A long on a perpetual pays funding while the market runs hot — exactly when the cross is in: funding cost more than fees and slippage together on all 20 coins and turned XRP and THETA into a loss. With funding off the perpetual runs mostly land close to spot — ETH +751.15% against +755.49% on spot. **Backtest: [Golden Cross 50/200 daily — top-20 study](https://janorium.com/backtests/120)** ETHUSDT · 4h · Futures · 2020-01-01 → 2026-01-01 | Return | Max drawdown | Sharpe | Win rate | Trades | | --- | --- | --- | --- | --- | | +358.00% | 74.47% | 0.74 | 66.7% | 6 | *ETH perpetual with funding: +358.00%, 47,219 USDT of funding paid* **Backtest: [Golden Cross 50/200 daily — top-20 study](https://janorium.com/backtests/140)** ETHUSDT · 4h · Futures · 2020-01-01 → 2026-01-01 | Return | Max drawdown | Sharpe | Win rate | Trades | | --- | --- | --- | --- | --- | | +751.15% | 62.47% | 0.93 | 66.7% | 6 | *The same ETH perpetual run with funding off: +751.15%* > **What this test does not show** > > - **"Top-20" is Binance futures turnover on 2021-01-01, not market cap.** The platform holds every perpetual since > 2020, delisted ones included, so the ranking does not depend on which spot pairs were loaded; but the date is a year > into the window, so 2020 is seen knowing which coins became popular by its end. > - **The averages need about 200 days of history inside the window.** The first trades come in July–August 2020, and > much later for coins listed in late 2020 (GRT's first is in November 2021); Buy & Hold counts from the first candle, > those months included. > - **Spot and perpetual history do not always start together.** THETA's spot history here starts on 2020-01-01, its perpetual > from May 2020, so the two runs of one coin can cover different months. > - **One rule, one set of parameters, one period** — a record of what happened, not a test for an edge. ## All 20 coins | Coin | Spot | Spot Buy & Hold | Max drawdown | Trades | Perpetual, with funding | Funding paid, USDT | | ----- | ---------------------------: | --------------: | -----------: | -----: | ---------------------------: | -----------------: | | BNB | [+2,312.15%](https://janorium.com/backtests/168) | +4,795.10% | 76.23% | 7 | [+1,795.23%](https://janorium.com/backtests/129) | 9,727 | | TRX | [+422.74%](https://janorium.com/backtests/175) | +1,527.29% | 87.26% | 6 | [+216.39%](https://janorium.com/backtests/136) | 10,476 | | ETH | [+755.49%](https://janorium.com/backtests/160) | +1,170.25% | 62.35% | 6 | [+358.00%](https://janorium.com/backtests/120) | 47,219 | | BTC | [+529.12%](https://janorium.com/backtests/159) | +859.56% | 57.11% | 6 | [+200.45%](https://janorium.com/backtests/119) | 22,119 | | XRP | [+14.83%](https://janorium.com/backtests/161) | +824.74% | 90.84% | 9 | [-53.06%](https://janorium.com/backtests/121) | 7,297 | | ADA | [+265.50%](https://janorium.com/backtests/167) | +168.81% | 84.86% | 6 | [+52.59%](https://janorium.com/backtests/127) | 45,321 | | BCH | [+62.79%](https://janorium.com/backtests/164) | +166.70% | 81.60% | 8 | [+11.88%](https://janorium.com/backtests/124) | 4,774 | | XLM | [-56.14%](https://janorium.com/backtests/165) | +100.12% | 88.48% | 8 | [-88.59%](https://janorium.com/backtests/125) | 7,568 | | ETC | [+122.90%](https://janorium.com/backtests/176) | +90.90% | 88.38% | 7 | [+51.56%](https://janorium.com/backtests/137) | 9,360 | | LTC | [-51.82%](https://janorium.com/backtests/162) | +81.22% | 92.55% | 8 | [-82.13%](https://janorium.com/backtests/122) | 10,499 | | LINK | [-9.95%](https://janorium.com/backtests/163) | +53.61% | 85.88% | 7 | [-75.44%](https://janorium.com/backtests/123) | 12,333 | | THETA | [+472.64%](https://janorium.com/backtests/173) | +7.47% | 90.59% | 6 | [-12.73%](https://janorium.com/backtests/134) | 37,025 | | AAVE | [-59.79%](https://janorium.com/backtests/172) | -70.98% | 77.12% | 6 | [-66.67%](https://janorium.com/backtests/133) | 968 | | UNI | [-87.22%](https://janorium.com/backtests/170) | -81.36% | 92.04% | 10 | [-91.41%](https://janorium.com/backtests/131) | 1,342 | | COMP | [-62.13%](https://janorium.com/backtests/174) | -85.16% | 91.25% | 6 | [-75.06%](https://janorium.com/backtests/135) | 3,734 | | DOT | [-85.10%](https://janorium.com/backtests/169) | -94.68% | 90.76% | 6 | [-93.33%](https://janorium.com/backtests/130) | 1,996 | | GRT | [-72.29%](https://janorium.com/backtests/166) | -95.45% | 83.75% | 4 | [-78.26%](https://janorium.com/backtests/126) | 1,147 | | EGLD | [-85.38%](https://janorium.com/backtests/171) | -96.19% | 92.52% | 4 | [-90.02%](https://janorium.com/backtests/132) | 1,006 | | FIL | [-85.79%](https://janorium.com/backtests/177) | -99.20% | 87.66% | 4 | [-88.38%](https://janorium.com/backtests/138) | 722 | | EOS | — | — | — | — | [-83.71%](https://janorium.com/backtests/128) | 3,597 | Each return links to its run. Max drawdown and trades are for spot; EOS has no spot history on the platform. ## Conclusion - As a buy signal on spot, the golden cross beat holding on 3 of the 12 coins that rose, each time thanks to a single trade opened in 2020. - What it did was step aside from collapses — it lost less than holding on 6 of 7 falling coins — at the price of giving back most of every rally. - On perpetuals the same rule did worse on every coin: for a long-only trend rule, funding was a larger cost than fees. How a backtest like this is put together: [what backtesting is](https://janorium.com/guides/what-is-backtesting).