Suppose there’s a person named Xiao Ze who works the night shift at a cold-storage warehouse south of the city and can’t clock out until he’s checked all the cargo manifests in the early hours. People tell him that if he likes an asset, he should just keep holding it. But he gets anxious watching days with big swings, so he set a rule for $ZKP , using 10,000 USDT as his starting capital: based on UTC daily candles, if the day’s high minus its low is strictly less than the range of every day in the previous 7 days (that is, it’s the narrowest of the most recent 8 closed daily candles), he buys at that day’s close. Then he counts 3 more closed daily candles and sells at the close of the third one. If another day with the same narrow range occurs while he’s holding, he skips the buy. If he’s still holding when the window ends, the position is valued at the final close.
We recalculated this using Binance spot daily candles (UTC, August 24 to November 21, 2026; 90 closed daily candles. To determine the narrowest range over 8 days, we also use closed candles from before those 90 days—the earliest one used is from August 17):
The first trade: bought at the August 26 close of 0.0417 (the day’s range was 0.0018, narrower than the narrowest range in the previous 7 days), then sold at the August 29 close of 0.0502. That was a 20.38% gain in 3 days, bringing the cash balance to about 12038.37.
He made 4 more trades in September. Bought at 0.0488 on the 5th and sold at 0.0479 on the 8th, losing 1.84%; bought at 0.0431 on the 11th and sold at 0.0446 on the 14th, gaining 3.48%; the 14th also met the rule, so he bought back at the same close immediately after selling, then sold at 0.0435 on the 17th, losing 2.47%; the 17th also triggered another buy, and he sold at 0.0476 on the 20th, gaining 9.43%, with about 13050.08 in cash. A week after he sold, on September 27, the price closed at 0.0522—the highest close in these 90 days, 9.66% above his selling price. He was already out of the market.
He bought back in on October 3 at 0.0491, 3.15% higher than his September 20 selling price. The price closed at 0.0517 on the 5th, and his account briefly reached about 13741.12 on paper. But the rule required him to wait until the close of the third candle: he sold at 0.0494 on the 6th, leaving a gain of just 0.61% on this trade and about 13129.81 in cash. On October 19, he bought again at 0.0410 and sold at 0.0413 on the 22nd, for a gain of just 0.73%.
The first two trades in November both went badly. He bought at 0.0402 on the 2nd and sold at 0.0365 on the 5th, losing 9.20%; he bought at 0.0328 on the 12th, and the price closed at 0.0292 on the 14th, bringing his account down to about 10690.56 on paper. He sold at 0.0300 on the 15th, losing 8.54%, with about 10983.45 in cash. Finally, he bought at 0.0301 on the 16th and sold at 0.0328 on the 19th, gaining 8.97%. Ten buys and ten sells; at the end of the window he was in cash, with about 11968.68, a gain of 19.69%. On November 21, the price closed at 0.0305, 7.01% below what he’d sold it for two days earlier. He was already out.
· If he had invested the full 10,000 at the August 24 close of 0.0411 and held until November 21 at 0.0305, he would have had about 7420.92 left, a loss of 25.79%—about 4547.76 less than with his strategy. The maximum drawdown while fully invested occurred on November 14: 44.06%. His strategy’s maximum drawdown was 22.20%, also on November 14, when his account was worth about 10690.56 on paper;
· On November 21, the price was 41.57% below the highest close of 0.0522; it would need to rise 71.15% to recover;
· Over the same period, $BTC went from 78992.75 to 60215.77, a decline of 23.77. The same 10,000 would have become about 7622.95. Applying the same rule—buy on a narrow-range day and hold for 3 days—would have turned that 10,000 into about 9292.03, a loss of 7.08%.
Xiao Ze finished checking the figures and stared at the line for September 20. “The highest close came the week after I sold. On October 5, my account hit its high, but on the third day I sold at a lower price. Those two trades in mid-November gave back some of what I’d made earlier.”
I didn’t tell him he shouldn’t have sold on September 20, or that he shouldn’t have bought again in November. These are all historical, closed daily candles; the latest one is from November 21, and no one knows what happens next. In this window, he ended up with about 4547.76 more than if he’d invested everything at once. But from the highest account value on paper, about 13741.12, down to 11968.68, he’d already given back about 1772.44. Choose a different window, and both 19.69% and -25.79% will change. Without that 20.38% gain from August 26 to 29, the earlier figure wouldn’t have been the same either.
Have you ever had a moment like this: you choose to enter on the quietest days, the highest close happens during a week when you’re out of the market, and a few days later you buy back in because the range has narrowed again?
We recalculated this using Binance spot daily candles (UTC, August 24 to November 21, 2026; 90 closed daily candles. To determine the narrowest range over 8 days, we also use closed candles from before those 90 days—the earliest one used is from August 17):
The first trade: bought at the August 26 close of 0.0417 (the day’s range was 0.0018, narrower than the narrowest range in the previous 7 days), then sold at the August 29 close of 0.0502. That was a 20.38% gain in 3 days, bringing the cash balance to about 12038.37.
He made 4 more trades in September. Bought at 0.0488 on the 5th and sold at 0.0479 on the 8th, losing 1.84%; bought at 0.0431 on the 11th and sold at 0.0446 on the 14th, gaining 3.48%; the 14th also met the rule, so he bought back at the same close immediately after selling, then sold at 0.0435 on the 17th, losing 2.47%; the 17th also triggered another buy, and he sold at 0.0476 on the 20th, gaining 9.43%, with about 13050.08 in cash. A week after he sold, on September 27, the price closed at 0.0522—the highest close in these 90 days, 9.66% above his selling price. He was already out of the market.
He bought back in on October 3 at 0.0491, 3.15% higher than his September 20 selling price. The price closed at 0.0517 on the 5th, and his account briefly reached about 13741.12 on paper. But the rule required him to wait until the close of the third candle: he sold at 0.0494 on the 6th, leaving a gain of just 0.61% on this trade and about 13129.81 in cash. On October 19, he bought again at 0.0410 and sold at 0.0413 on the 22nd, for a gain of just 0.73%.
The first two trades in November both went badly. He bought at 0.0402 on the 2nd and sold at 0.0365 on the 5th, losing 9.20%; he bought at 0.0328 on the 12th, and the price closed at 0.0292 on the 14th, bringing his account down to about 10690.56 on paper. He sold at 0.0300 on the 15th, losing 8.54%, with about 10983.45 in cash. Finally, he bought at 0.0301 on the 16th and sold at 0.0328 on the 19th, gaining 8.97%. Ten buys and ten sells; at the end of the window he was in cash, with about 11968.68, a gain of 19.69%. On November 21, the price closed at 0.0305, 7.01% below what he’d sold it for two days earlier. He was already out.
· If he had invested the full 10,000 at the August 24 close of 0.0411 and held until November 21 at 0.0305, he would have had about 7420.92 left, a loss of 25.79%—about 4547.76 less than with his strategy. The maximum drawdown while fully invested occurred on November 14: 44.06%. His strategy’s maximum drawdown was 22.20%, also on November 14, when his account was worth about 10690.56 on paper;
· On November 21, the price was 41.57% below the highest close of 0.0522; it would need to rise 71.15% to recover;
· Over the same period, $BTC went from 78992.75 to 60215.77, a decline of 23.77. The same 10,000 would have become about 7622.95. Applying the same rule—buy on a narrow-range day and hold for 3 days—would have turned that 10,000 into about 9292.03, a loss of 7.08%.
Xiao Ze finished checking the figures and stared at the line for September 20. “The highest close came the week after I sold. On October 5, my account hit its high, but on the third day I sold at a lower price. Those two trades in mid-November gave back some of what I’d made earlier.”
I didn’t tell him he shouldn’t have sold on September 20, or that he shouldn’t have bought again in November. These are all historical, closed daily candles; the latest one is from November 21, and no one knows what happens next. In this window, he ended up with about 4547.76 more than if he’d invested everything at once. But from the highest account value on paper, about 13741.12, down to 11968.68, he’d already given back about 1772.44. Choose a different window, and both 19.69% and -25.79% will change. Without that 20.38% gain from August 26 to 29, the earlier figure wouldn’t have been the same either.
Have you ever had a moment like this: you choose to enter on the quietest days, the highest close happens during a week when you’re out of the market, and a few days later you buy back in because the range has narrowed again?