August 1, 2026 – August 31, 2026
August 2026 rebalancing bot results: +4.52% vs BTC +23.07%
| Month | Max drawdown (month) | Cumulative at month end | |
|---|---|---|---|
| Rebalancing bot | +4.52% | -4.83% | -7.54% |
| BTC 100% hold | +23.07% | -3.93% | +22.17% |
| QQQ 100% hold | +4.07% | -3.91% | -3.52% |
Versus the same 25/25/50 start with no rebalancing: -1.26 pts
August was a good month for my rebalancing bot, though not a good month next to bitcoin. The portfolio finished up 4.52% for the month. BTC rose 23.07% and QQQ rose 4.07%. So the bot slightly beat the Nasdaq-100 ETF and trailed bitcoin by a wide margin. That is what I should expect from a portfolio that holds only 25% in BTC spot, 25% in a 1x long SOXL perpetual, and 50% in USDT on Bybit. It is built to be smoother than a pure bitcoin position, and in a strong bitcoin month that costs a lot of upside.
The longer view is less comfortable. After August, the bot's cumulative result is still -7.54%. BTC's cumulative figure is +22.17% and QQQ's is -3.52%. A good month did not erase the earlier damage, and I would rather say that plainly than let one green month change the story. If you are searching for rebalancing bot results, this is the honest picture: positive this month, still under water overall.
The best day was August 4, when the bot gained 5.74% while BTC gained 0.51% and QQQ gained 2.82%. The worst day was August 18, when the bot lost 3.65% while BTC rose 0.86% and QQQ fell 1.89%. I have no notes attached to either day explaining what drove the moves, so I won't invent a story. I can only say that on both days the bot moved much more than bitcoin did, which tells me the SOXL leg, not the BTC leg, probably mattered more for day-to-day swings. That is an inference from the shape of the numbers, not a confirmed cause. The month had 18 up days and 13 down days.
On rebalancing itself, the bot made 8 trades in total: 4 buys and 4 sells. Three were in BTC and five were in SOXL. The rule is simple: check hourly and rebalance when any weight drifts 2 percentage points from target. By the end of the month the weights were 25.71% BTC, 24.25% SOXL and 50.04% USDT, so it stayed close to the 25/25/50 plan. The important result is that rebalancing did not beat buy-and-hold this month. The rebalanced portfolio came in 1.26 percentage points behind a buy-and-hold portfolio that started at the same weights. BTC went from 63,270.7 to 77,862.1, and every time the bot trimmed a rising BTC position back to target, it gave up some of that run.
This is where Shannon's demon comes in. The idea is that a portfolio rebalanced between volatile assets can earn a return from the volatility itself, by selling what has risen and buying what has fallen. It works best when prices bounce around without a strong trend. This month BTC trended hard in one direction, and SOXL slipped from 121.45 to 113.65. Rebalancing between a strongly rising asset and a weaker one is a drag, not a benefit. One month says little about whether the approach works over time, but it is a clear example of where it hurts.
On drawdown, the bot's worst peak-to-trough drop within the month was -4.83%. BTC's was -3.93% and QQQ's was -3.91%. So even with half the portfolio in USDT, the bot had the deepest dip of the three, even though it did not rise as much as bitcoin. That is the cost of the SOXL leg: a leveraged-semiconductor ETF-style exposure adds sharp swings, and holding only 1x long on a perpetual does not make it calm. The USDT cushion helps, but it did not make this month smoother than simply holding QQQ.
The lesson I take from August is that a month of positive returns can hide a weaker month underneath. I came out ahead of QQQ, behind bitcoin, behind my own buy-and-hold baseline, and with a larger drawdown than both benchmarks. I would not read that as proof the strategy is broken, and I would not read the 4.52% as proof it is working. I will keep watching whether the rebalancing premium shows up over many months, rather than judging it on one. This is a record of what my bot did, not a recommendation for anyone else.
Daily entries this month
How rebalancing works: Shannon's Demon →
Methodology
Total equity is logged by the bot every hour at :59. Deposits are removed at the hour they arrive and returns are chained as a time-weighted return (TWR). Daily return compares the last snapshot of the day with the previous day's last snapshot. Benchmarks use Bybit BTCUSDT spot and QQQUSDT perpetual hourly closes.
This site is a personal record, not investment advice. Past performance does not guarantee future results.