Suppose there’s a person named Xiaomin. At the close on August 19, $PEOPLE closed at 0.00811. He set himself a rule: look only at completed daily candles, and only at that day’s own high and low. The opening price must be in the bottom quarter of the day’s range (position = (open - low) / (high - low), no more than 25%; exactly 25% counts), while the closing price must be in the top quarter of the day’s range (position at least 75%; exactly 75% counts). Days when the high equals the low, meaning the range is zero, don’t count. If the conditions are met, he buys at the close and sells at the next day’s close. If the conditions still hold on the day he sells, he sells and buys back in. He doesn’t add to his position while holding. His colleague Xiaocang converted all 1,000 USDT into the coin at 0.00811 on the same day, planning to hold until the end. The replay below excludes fees.

Over these 90 days, there were 6 days that met the rule. Xiaomin bought 6 times, sold 6 times, sold and bought back in 2 times, and ended with cash.

His turn came on the very first day of the window. On August 19, the open was 0.00757, at a position of 22.47%; the close was 0.00811, at a position of 83.15%. He bought at that close and sold the next day for a 29.72% gain, bringing his account to 1,297.16—the biggest gain among his 6 completed trades. The conditions still held on the day he sold: on August 20, the opening position was 5.28% and the closing position was 75.95%, just over the 75% threshold. He sold and bought back in at 0.01052. The next day, he gained 7.98%, bringing his account to 1,400.74. On August 21, PEOPLE touched its highest price of the 90-day period, 0.01315, intraday, and also set its highest close at 0.01136, up 40.07% from the starting point. Xiaocang’s account was at 1,400.74; Xiaomin happened to sell at that candle. There were only 3 new closing highs over these 90 days, and August 21 was the last one.

Then he stayed out for 20 candles. On September 11, he bought at 0.00782 and gained 1.15% the next day, bringing his account to 1,416.86. After staying out for 4 candles, he bought at 0.00834 on September 17 and gained 8.27% the next day, bringing his account to 1,534.08—the second-biggest gain among his 6 completed trades, and the highest his account reached under the strategy. He bought back in at 0.00903 on the day he sold, then lost 0.33% the next day, bringing the account to 1,528.99.

After staying out for another 19 candles, he made his final trade: on October 9, he bought at 0.00819, then lost 0.37% the next day, bringing his account to 1,523.39. This was the biggest loss among his 6 completed trades. He then stayed out for another 37 candles, all the way to the end of the period.

Results over these 90 days (drawdown calculated using daily closes):
Xiaocang, fully invested the whole time: 628.85 USDT, -37.11%, with a maximum drawdown of 57.48% (from the August 21 highest close of 0.01136, when his account was at 1,400.74, to the November 14 close of 0.00483, when his account was at 595.56). On November 16, the close was 0.0051, 55.11% below the highest close; it would take a further gain of 122.75% to return to that high. Over the same period, $BTC fell 6.42% from its August 19 close to its November 16 close. The lowest close in these 90 days was on November 14. Both started with 1,000 USDT.
Xiaomin, holding overnight only on days when the open was low enough and the close high enough: 1,523.39 USDT, +52.34%, with a maximum drawdown of 0.70% (from his account value of 1,534.08 on September 18 to 1,523.39 after selling on October 10). Of the 6 completed trades, 4 were profitable and 2 lost money. The first trade, at +29.72%, contributed the most. Excluding it, the remaining 5 trades, still starting from 1,000 USDT, would have ended at 1,174.40 USDT, or +17.44%.

The figures above are a replay of one rule using Binance spot PEOPLEUSDT and BTCUSDT historical daily candles (UTC daily candles; completed data from 2026-08-19 to 11-16). They don’t indicate what will happen in the future. Across these 90 days, the 6 overnight trades added up to +52.34%. The second trade’s selling price happened to coincide with the highest close on the August 21 candle. For most of the days afterward, he stayed out and didn’t ride the subsequent drawdown. Would you use a low enough open and a high enough close on the day to decide whether to hold overnight?