Buying SOXL Near the Top: How My Bot Lost 3 Trades in a Week
Buying SOXL Near the Top: How My Bot Lost 3 Trades in a Week
"Buy the dip" is one of the most repeated phrases in investing. It usually works ??until it doesn't, and the moment it stops working is almost always the same: when the dip isn't a pullback in an uptrend, but the first crack in a top. My automated SOXL bot walked straight into this trap during one week of June 2026, buying the dip again and again as the 3x semiconductor ETF rolled over, and getting stopped out three times for double-digit losses. Here's the trade-by-trade record of how a good rule fails at the wrong time.
Disclosure: Personal, small-scale experiment for educational purposes only. Not investment advice. Leveraged ETFs carry extreme risk.
The Setup: A Perfect Run, Then the Turn
The week started brilliantly. On June 3, the bot closed its best trade ever ??selling at roughly $282 for +76.94% as RSI hit an extreme 80.4. SOXL had gone parabolic. That should have been the warning sign. After a move like that, the smart posture is caution. Instead, the bot's entry logic saw the subsequent drop as a buying opportunity ??a "pullback in an uptrend" ??and started buying again.
The Trades That Followed
Here's what happened over the next eight days:
| Date | Action | Price | RSI | Result |
|---|---|---|---|---|
| Jun 3 | SELL | $282.56 | 80.4 | +76.94% |
| Jun 5 | BUY | $219.75 | 57.3 | ?? |
| Jun 6 | SELL | $194.00 | 51.0 | -11.76% |
| Jun 9 | BUY | $212.36 | 54.6 | ?? |
| Jun 10 | SELL | $179.66 | 48.0 | -14.97% |
| Jun 10 | BUY | $208.22 | 53.5 | ?? |
| Jun 11 | SELL | $179.34 | 47.7 | -13.99% |
Read it top to bottom. After selling the top at $282, the bot bought at $219, got stopped at $194. Bought again at $212, stopped at $179. Bought again at $208, stopped at $179. Each "dip" it bought turned out to be another step down a staircase, not a bounce.
Why the Rule Failed
The entry logic wasn't broken ??it was context-blind. "Buy the pullback when RSI dips into the mid-50s during an uptrend" is a perfectly reasonable rule. The problem is the phrase "during an uptrend." After June 3, there was no uptrend anymore. SOXL had peaked and was declining. But the RSI readings (57, 54, 53) still looked like healthy pullback levels in isolation, so the bot kept buying what it thought were dips.
This is the single most expensive lesson in the entire seven-month log: the same entry signal means opposite things depending on whether the larger trend is intact. A mid-50s RSI dip is a gift in a young uptrend and a trap in a topping market. The bot couldn't tell the difference, so it bought the top three times.
The Fix the Data Points To
The losses here weren't caused by the exits ??the stops actually worked, capping each loss near -12% to -15% instead of letting them run to -40%. The fix has to happen at entry. Specifically:
- Add a regime filter. Before buying any dip, confirm the larger trend is still up (e.g., price above a longer moving average, or no recent parabolic exhaustion signal).
- Stand down after a blow-off top. When an RSI-80 exit just fired after a parabolic run, that's precisely when not to re-enter aggressively. The bot did the opposite.
- Respect the staircase. Three consecutive stopped-out dip buys is itself a signal: the "pullbacks" aren't pullbacks. A smarter system would halt dip-buying after two failures in a row.
What Every Investor Should Take From This
You don't need a bot to fall into this trap ??humans do it constantly, and worse, because they don't have a stop-loss. The lesson generalizes cleanly:
- "Buy the dip" requires an intact uptrend to be valid. Without that condition, you're just catching a falling knife.
- The most dangerous dips look identical to the best ones. The only difference is the larger context, which is exactly what's easiest to ignore when a stock has been going up.
- After a euphoric top, the first few "dips" are usually distribution, not opportunity.
The Bottom Line
In one week, my bot bought the dip four times and lost three trades for a combined double-digit hit ??not because dip-buying is wrong, but because it was dip-buying in a market that had just topped. The stops contained the damage; the entry logic caused it. If there's a single line to remember: a dip is only a buying opportunity if the trend it's dipping within is still alive.
Disclosure: Personal experiment, educational purposes only. Not financial advice. Leveraged ETFs carry substantial risk of loss.