Dip Accumulators
- Asset
- SOXL — 3x leveraged semiconductors
- Capital
- $1,000
- Phase 1
- Volatility harvest, 80/20 rebalance — abandoned; needed a +46% run to bank $73
- Phase 2
- Dip accumulator — buy 50% of cash on a −7% daily drop, ratcheted sells above cost basis
- Outcome
- Phase 2 replaced Phase 1 and is the template the later bots were built on.
With the boom of AI, data centers, and companies beefing up on capex to build the infrastructure, I’ve been closely watching the performance of semiconductors. The next bot would trade SOXL as the underlying asset. SOXL is a 3x leveraged ETF tracking the daily return of the NYSE Semiconductor Index, which consists of 30 of the largest semiconductor companies listed in the U.S.
Phase 1 – Volatility Harvest
- Strategy – 80/20 equity/cash rebalancing
- Buy SOXL with 80% of cash while holding 20% as cash reserve
- Rebalance when 80/20 drifts 5% in either direction
- Capital - $1,000
| SOXL Volatility Harvest | |||
|---|---|---|---|
| SOXL Price | % Equity | % Cash | Notes |
| $240.00 | 80% | 20% | Position Opened |
| $264.00 | 81% | 19% | 10% increase in SOXL price |
| $290.40 | 83% | 17% | 10% increase in SOXL price |
| $319.44 | 84% | 16% | 10% increase in SOXL price |
| $351.38 | 85% | 15% | 10% increase in SOXL price |
Each price movement in the table above represents a 10% increase in SOXL. At this rate, SOXL would have to be around $351, representing a 46% increase from the entry price for the first rebalancing to occur. In that scenario, the banked profit would be $73. This was far too large an upswing needed to recognize gains and was not a sustainable approach.
The strategy needs to be revised so capital is actively deployed. Trades need to be executed to properly gauge the bot’s performance. Instead of rebalancing positions, I could buy on daily drops and sell based on the average cost basis.
Phase 2 – Dip Accumulator
- Strategy
- Buy on -7% daily drops with 50% of allocated cash
- Ratcheted sells when returns are a certain level above the average cost basis
| Gain above Avg Cost | Sell % |
|---|---|
| +5% to 15% | 20% |
| +15% to 25% | 30% |
| +25% to 40% | 45% |
| +40%+ | 50% |
The -7% daily drop could seem like a significant movement at first, but SOXL has experienced high volatility due to strong quarterly performances with semiconductors, and profit-taking. I had initially started with -3% or -4% drops, but it was depleting my cash reserves in a single trading day, and there would not be any dry powder remaining for subsequent downturns.
The ratcheted sells were designed to capture gains while maintaining exposure. I thought of this as trimming my positions, but the more important factor was to continue holding a position that could benefit from an upward trajectory.
The dip accumulator bot was scheduled to run on a 300 second (5 minute) interval, where Claude would pull SOXL’s price and calculate the return based on my average cost basis. During each process, if SOXL was at least 5% above my cost basis, it would sell a percentage of my total position based on the bands mentioned above.
If SOXL faces headwinds and remains below my cost basis, the bot incorporates a B&H strategy which holds any underwater positions.
Using the same SOXL price figures from the volatility harvest table, the dip accumulator tells a different story. With $1,000 in capital and an entry at $240 on a -7% daily drop, the bot deploys 50% ($500) into 2.082 shares and holds $500 as dry powder.
| SOXL Dip Accumulator | |||||
|---|---|---|---|---|---|
| After Trade | Returns | ||||
| SOXL Price | Action | Shares | Cash | Realized P&L | % Return |
| $240.00 | Buy 50% of cash | 2.082 | $500.00 | – | – |
| $264.00 | +10% gain, sell 20% | 1.664 | $610.00 | $10.00 | 5.0% |
| $290.40 | +21% gain, sell 30% | 1.164 | $755.18 | $35.18 | 9.4% |
| $319.44 | +33% gain, sell 45% | 0.645 | $922.91 | $76.91 | 12.8% |
| $351.38 | +46% gain, sell 50% | 0.318 | $1,035.64 | $112.64 | 14.8% |
By the time SOXL reaches $351 – the same point where the volatility harvest would execute its first rebalance for $73 – the dip accumulator has already executed four sells, realized $112.64 in profit, and returned 14.8% on its deployed capital. The bot progressively locks in gains while still holding 0.318 shares for further upside.
What it looks like in a chop
That run-up flatters the strategy by never testing it. A clean 46% climb is the one scenario where a dip accumulator cannot buy anything – there are no dips – and plain buy-and-hold would have returned the full 46% while the ladder returned 14.8%. The ladder is not there to beat a rally. It is there to earn its keep when the price grinds sideways, which is most of the time.
Here is the same $1,000 through three -7% days followed by a partial recovery:
| SOXL Dip Accumulator – Chop | |||||
|---|---|---|---|---|---|
| SOXL Price | Action | Shares Held | Cash | Avg Cost | Realized P&L |
| $240.00 | -7% day, buy 50% of cash | 2.083 | $500.00 | $240.00 | – |
| $223.20 | -7% day, buy 50% of cash | 3.203 | $250.00 | $234.13 | – |
| $207.58 | -7% day, buy 50% of cash | 3.806 | $125.00 | $229.92 | – |
| $241.42 | +5% over cost, sell 20% | 3.044 | $308.75 | $229.92 | $8.75 |
| $264.41 | +15% over cost, sell 30% | 2.131 | $550.25 | $229.92 | $40.25 |
The three dips pull the average cost from $240.00 down to $229.92, and that is the whole mechanism. By the time SOXL recovers to $264.41 the bot is worth $1,113.75 against $1,101.72 for buy-and-hold over the same path – a thin edge, and it comes from the buying, not the selling. Take away the dips and the ladder loses to buy-and-hold every time.
Terms used above are defined in the glossary.