Suppose there is a person named Xiaoni. On the close of August 22, $NEIRO closed at 0.00009186. He set himself a rule: only look at completed daily K-lines. The day’s low must be lower than the lows of the previous 9 consecutive daily K-lines; ties do not count, meaning a new low must be set across those 10 candles. The day must also be a bullish candle, with the close higher than the open; ties do not count. The close must also be higher than yesterday’s close; if equal, it does not count. If these conditions are met, he buys at the close; he sells at the next day’s close. If the day of selling also meets the conditions, he sells and buys back again. He does not add to the position while holding. His colleague Xiaocang bought in on the same day at 0.00009186, converting all 1000 USDT into coins, planning to hold until the end of the period. The following replay does not include fees.
Within these 90 days, there were 5 days that met the rule. Xiaoni bought 5 times, sold 5 times, never bought back again, and ended in cash.
The opening window had 19 empty candles. The highest price in these 90 days appeared on the opening-window day of August 22, reaching 0.00011109, and both started from that day’s close. On September 11, the low was 0.00007956, below the lowest 0.0000812 among the previous 9 candles. The day opened at 0.00008161 and closed at 0.00008291, higher than yesterday’s close of 0.00008159. He bought in, and on September 12 it closed at 0.00008766, making 5.73%, with the account reaching 1057.29, the best of the 5 trades.
Then 3 empty candles. On September 16, the low was 0.00007647, while the lowest among the previous 9 candles was 0.000077. The day closed at 0.00008017. He bought in, and the next day was +4.64%, reaching 1106.35.
Then 38 more empty candles. In between, on September 22, it closed at 0.00009944, 8.25% above the starting price. This was the highest close in these 90 days, and the last time a closing high was set, appearing 4 times in total. He had already sold by then, and was holding 1106.35. Xiaocang’s account that same day was 1082.52, still below his.
On October 26, the low was 0.00006495, below the previous 9-candle low of 0.00006518. The day closed at 0.00006685. He bought in, and the next day, October 27, it closed at 0.00006826, making 2.11%, with the account at 1129.69, the peak of the strategy’s account value.
Then 2 empty candles. On October 30, the low was 0.00005797, below the previous 9-candle low of 0.00006155. The day closed at 0.00006545. He bought in, and the next day lost 2.46%, dropping to 1101.90. This was the only losing trade among the 5 sells, and also the day with the deepest strategy drawdown.
Then 8 empty candles. The last trade was on November 9. The low was 0.00005624, below the previous 9-candle low of 0.000058. The day closed at 0.00005918. He bought in, and the next day gained 0.42%, reaching 1106.55. Then 9 more empty candles, all the way to the end. On November 14, the lowest close in these 90 days appeared at 0.00004748, on the same day as the lowest price of 0.00004518. He had already sold by then, and was still holding 1106.55. Xiaocang’s account that same day was 516.87.
The results for these 90 days (drawdown based on daily closes):
Xiaocang, fully invested the whole time: 544.52 USDT, -45.55%, deepest drawdown 52.25% (from the September 22 highest close of 0.00009944 and account value 1082.52 to the November 14 close of 0.00004748 and account value 516.87); on November 19, it closed at 0.00005002. To return to that highest close, it would need to rise another 98.80%. Over the same period, $BTC fell -19.58% from the August 22 close to the November 19 close. Both started with 1000 USDT.
Xiaoni, holding overnight only on days that “set a 10-day low, close bullish, and close above yesterday’s close”: 1106.55 USDT, +10.66%, deepest drawdown 2.46% (from the October 27 account value of 1129.69 to the October 31 sold value of 1101.90). Among the 5 sold trades, 4 were profitable and 1 was a loss, with no breakeven trades. That +5.73% trade contributed the most; without it, the remaining 4 trades, still starting from 1000 USDT, would end at 1046.59 USDT, +4.66%.
The above is a rule replay based on Binance spot NEIROUSDT and BTCUSDT historical daily candles (UTC daily candles, closed data from 2026-08-22 to 11-19), and does not indicate what will happen in the future. Over these 90 days, these 5 overnight holds summed to +10.66%, with the best one being the night of September 12; Xiaocang reached 1082.52 on September 22 and ended with only 544.52. Would you use “a 10-day new low with a bullish close that day” to decide whether to hold overnight?
Within these 90 days, there were 5 days that met the rule. Xiaoni bought 5 times, sold 5 times, never bought back again, and ended in cash.
The opening window had 19 empty candles. The highest price in these 90 days appeared on the opening-window day of August 22, reaching 0.00011109, and both started from that day’s close. On September 11, the low was 0.00007956, below the lowest 0.0000812 among the previous 9 candles. The day opened at 0.00008161 and closed at 0.00008291, higher than yesterday’s close of 0.00008159. He bought in, and on September 12 it closed at 0.00008766, making 5.73%, with the account reaching 1057.29, the best of the 5 trades.
Then 3 empty candles. On September 16, the low was 0.00007647, while the lowest among the previous 9 candles was 0.000077. The day closed at 0.00008017. He bought in, and the next day was +4.64%, reaching 1106.35.
Then 38 more empty candles. In between, on September 22, it closed at 0.00009944, 8.25% above the starting price. This was the highest close in these 90 days, and the last time a closing high was set, appearing 4 times in total. He had already sold by then, and was holding 1106.35. Xiaocang’s account that same day was 1082.52, still below his.
On October 26, the low was 0.00006495, below the previous 9-candle low of 0.00006518. The day closed at 0.00006685. He bought in, and the next day, October 27, it closed at 0.00006826, making 2.11%, with the account at 1129.69, the peak of the strategy’s account value.
Then 2 empty candles. On October 30, the low was 0.00005797, below the previous 9-candle low of 0.00006155. The day closed at 0.00006545. He bought in, and the next day lost 2.46%, dropping to 1101.90. This was the only losing trade among the 5 sells, and also the day with the deepest strategy drawdown.
Then 8 empty candles. The last trade was on November 9. The low was 0.00005624, below the previous 9-candle low of 0.000058. The day closed at 0.00005918. He bought in, and the next day gained 0.42%, reaching 1106.55. Then 9 more empty candles, all the way to the end. On November 14, the lowest close in these 90 days appeared at 0.00004748, on the same day as the lowest price of 0.00004518. He had already sold by then, and was still holding 1106.55. Xiaocang’s account that same day was 516.87.
The results for these 90 days (drawdown based on daily closes):
Xiaocang, fully invested the whole time: 544.52 USDT, -45.55%, deepest drawdown 52.25% (from the September 22 highest close of 0.00009944 and account value 1082.52 to the November 14 close of 0.00004748 and account value 516.87); on November 19, it closed at 0.00005002. To return to that highest close, it would need to rise another 98.80%. Over the same period, $BTC fell -19.58% from the August 22 close to the November 19 close. Both started with 1000 USDT.
Xiaoni, holding overnight only on days that “set a 10-day low, close bullish, and close above yesterday’s close”: 1106.55 USDT, +10.66%, deepest drawdown 2.46% (from the October 27 account value of 1129.69 to the October 31 sold value of 1101.90). Among the 5 sold trades, 4 were profitable and 1 was a loss, with no breakeven trades. That +5.73% trade contributed the most; without it, the remaining 4 trades, still starting from 1000 USDT, would end at 1046.59 USDT, +4.66%.
The above is a rule replay based on Binance spot NEIROUSDT and BTCUSDT historical daily candles (UTC daily candles, closed data from 2026-08-22 to 11-19), and does not indicate what will happen in the future. Over these 90 days, these 5 overnight holds summed to +10.66%, with the best one being the night of September 12; Xiaocang reached 1082.52 on September 22 and ended with only 544.52. Would you use “a 10-day new low with a bullish close that day” to decide whether to hold overnight?