Suppose there’s a person named Xiaoxing who works evening shifts at a supermarket in a county town. On August 17, $XLM opened at 0.1563. His coworker Lao Liang said to buy and hold, and not stay up all night staring at candlesticks. Xiaoxing set himself a different rule: if the day’s low made a new low compared with the previous 10 completed daily candles, but the close recovered to at least halfway between that day’s high and low, he would buy 1,000 USDT at the close. Then, after the next 5 completed daily candles, he would sell at that day’s close.
I tested this rule using Binance spot XLM/USDT completed daily candles from August 17 through November 14 (UTC), 90 candles in total. The November 15 candle had not closed yet, so it wasn’t included. The first 10 candles served as the baseline, leaving 80 days on which the rule could be applied.
Of those 80 days, 18 made a new low compared with the previous 10 days. On 11 of those, the close failed to recover halfway through the day’s range, so he didn’t buy. Of the remaining 7 days that did recover, 2 came while the previous trade was still open (October 18 and November 4), so he skipped them as the rules required. He made 5 trades in total: 2 winners and 3 losers, with no two trades back-to-back. After putting the 1,000 USDT through 5 round trips, he ended with 1,194.54, a compounded gain of 19.45% and an average gain of 3.95% per trade. The first 2 trades were both profitable, compounding to +29.26% and bringing the account to 1,292.62; the last 3 all lost, giving back 7.59%. The best trade was buying at the September 16 close of 0.1839 and selling at the September 21 close of 0.2155, for +17.18%. The worst was buying at the October 24 close of 0.1716 and selling at the October 29 close of 0.1623, for -5.42%.
If Lao Liang had bought at the August 17 open and held until the November 14 close of 0.1355, his 1,000 USDT would have become 866.92, a loss of 13.31%. Over the same period, $BTC was up 1.99% from the August 17 open to the November 14 close.
The turning point came later. Xiaoxing sold his second trade on September 21. After that, the September 28 close of 0.2325 was the highest close of the 90-day period, and Lao Liang’s account would have reached 1,487.52 that day (+48.75%). On September 29, the intraday high reached 0.2371. Xiaoxing was out of the market on both days. From that peak to the November 14 close, the price fell 42.85%; it would need to rise another 74.98% to get back there. The two points were 46 days apart.
The intraday low on November 14 was 0.13, also the lowest of the 90-day period. The close was 0.1355, down 7.32% from the November 13 close of 0.1462. This day also made a new 10-day low, but the close recovered only 32.93% of the day’s range, so the rules said not to buy. Even if he had bought, the close 5 candles later was not available yet.
Lao Liang’s account was worth more at the end of September, but he later gave back those gains and fell below his starting capital. Xiaoxing stayed out of the market on most days, entering only when the price made a new low and then recovered at least halfway, and holding for 5 days. These are historical results based on completed candles. They don’t indicate what will happen next, and they are not trading advice. When a new low is made, would you wait to see if the price recovers, or hold all the way through?
I tested this rule using Binance spot XLM/USDT completed daily candles from August 17 through November 14 (UTC), 90 candles in total. The November 15 candle had not closed yet, so it wasn’t included. The first 10 candles served as the baseline, leaving 80 days on which the rule could be applied.
Of those 80 days, 18 made a new low compared with the previous 10 days. On 11 of those, the close failed to recover halfway through the day’s range, so he didn’t buy. Of the remaining 7 days that did recover, 2 came while the previous trade was still open (October 18 and November 4), so he skipped them as the rules required. He made 5 trades in total: 2 winners and 3 losers, with no two trades back-to-back. After putting the 1,000 USDT through 5 round trips, he ended with 1,194.54, a compounded gain of 19.45% and an average gain of 3.95% per trade. The first 2 trades were both profitable, compounding to +29.26% and bringing the account to 1,292.62; the last 3 all lost, giving back 7.59%. The best trade was buying at the September 16 close of 0.1839 and selling at the September 21 close of 0.2155, for +17.18%. The worst was buying at the October 24 close of 0.1716 and selling at the October 29 close of 0.1623, for -5.42%.
If Lao Liang had bought at the August 17 open and held until the November 14 close of 0.1355, his 1,000 USDT would have become 866.92, a loss of 13.31%. Over the same period, $BTC was up 1.99% from the August 17 open to the November 14 close.
The turning point came later. Xiaoxing sold his second trade on September 21. After that, the September 28 close of 0.2325 was the highest close of the 90-day period, and Lao Liang’s account would have reached 1,487.52 that day (+48.75%). On September 29, the intraday high reached 0.2371. Xiaoxing was out of the market on both days. From that peak to the November 14 close, the price fell 42.85%; it would need to rise another 74.98% to get back there. The two points were 46 days apart.
The intraday low on November 14 was 0.13, also the lowest of the 90-day period. The close was 0.1355, down 7.32% from the November 13 close of 0.1462. This day also made a new 10-day low, but the close recovered only 32.93% of the day’s range, so the rules said not to buy. Even if he had bought, the close 5 candles later was not available yet.
Lao Liang’s account was worth more at the end of September, but he later gave back those gains and fell below his starting capital. Xiaoxing stayed out of the market on most days, entering only when the price made a new low and then recovered at least halfway, and holding for 5 days. These are historical results based on completed candles. They don’t indicate what will happen next, and they are not trading advice. When a new low is made, would you wait to see if the price recovers, or hold all the way through?