75+ Strategies | The Algorithmic Trader who Trades just 10 Minutes a Day - Laurens Bensdorp

ELI5 / TLDR

Laurens Bensdorp runs 75+ automated trading strategies across stocks, ETFs, and commodities, spending roughly zero minutes a day on execution. His core insight is counterintuitive: don’t look for one great strategy — build a portfolio of mediocre-looking strategies that each shine in different market conditions. Equal-weight them, keep the rules simple, think in risk first, and let compounding do the heavy lifting at 15-20% annual returns with drawdowns under 10%.

The Full Story

From White-Water Rafting to Systematic Trading

Bensdorp was running an adventure tourism company in Mexico when his father called in 2000 — the family’s boutique venture capital firm was hemorrhaging money thanks to crooked accountants and bad advisers. He came in to help, liquidated the venture capital side, and got hooked on the stock trading portion. Zero formal education in finance. He read the Market Wizards books, noticed how many successful traders were systematic, and decided that was his path.

By 2007 he had four mean reversion strategies — two long, two short. Buy oversold stocks, wait for them to snap back. Short overbought stocks, wait for the drop. High win rate, short holding periods, two to five days. The 2008 bear market was kind to his short-side strategies. Easy start.

The 2011 Lesson That Changed Everything

Then August 2011 hit. The market dropped 15-20% in a week. His long mean reversion strategies all lost money simultaneously — doesn’t matter if you hold 30 different stocks when the whole market panics in lockstep. His short strategies weren’t even in the market because they needed pullbacks to short, and this was straight momentum down.

“That made me realize, okay, it’s all cool that I have a combination of different systems, but I do need momentum systems as well, both on the long and on the short side.”

This was the inflection point. He began building strategies with different purposes for different market regimes: volatile, non-volatile, choppy, strong momentum. Each one designed to perform during a specific condition.

The Portfolio-of-Strategies Philosophy

The key mental shift: a strategy doesn’t need to look good on its own. In fact, you want to see a weird equity curve — big spikes during the exact periods your other strategies are struggling, then flat or losing money the rest of the time.

“You actually want to see a very weird equity curve. You want to see an equity curve that excels in those time periods of choppiness, but at other times where your trend following strategies are making a lot of money, then I would expect that new strategy actually to lose money.”

He starts every new strategy by identifying a weakness in his existing blend. What market scenario hasn’t he planned for? Where did things go wrong in 2018, or during COVID, or July 2024? Was the flaw random, or was it a conceptual gap he could fill with a purpose-built strategy?

Thematic opportunities matter too. Geopolitical tension? Build a trend-following strategy on aerospace and defense stocks. Commodity supercycle? Create a simple system on gold, silver, and crude oil ETFs using Bollinger Bands or Keltner channels. That commodity strategy might show only 8% annual returns with a 40% max drawdown in isolation — looks terrible. But during the 2008 bear market, the 2022 Russia-Ukraine crisis, and the 2025 precious metals run, it was making money while everything else bled.

Capital Allocation: The Equal-Weight Rule

“What I’ve seen what happens a lot is that people have a blend of strategies. What they do is the strategy that tests the best, they allocate the most money to it. Don’t do that. Don’t do that.”

The logic is clean: if your best backtested strategy underperforms going forward, it does the most damage to your account because it has the most capital. Equal allocation across all 75 strategies is the default. Two exceptions — ETF-based strategies get slightly more because they lack individual corporate risk, and penny stock strategies get slightly less because the volatility is higher.

New strategies get full allocation immediately. No paper trading period, no gradual ramp-up. If the strategy was designed for volatility spikes and doesn’t fire for three years, that might just mean volatility has been calm. The homework was done beforehand.

Keeping It Simple

“Don’t have 17 different rules and five different exit mechanisms and stuff like that, but try to keep it simple.”

For trend following, the job description is short: find a trend, stay in until it bends. Period. His lookback periods range from 20-day Keltner channels (fast entry — would have caught the COVID bounce early) to 250-day channels (slow entry — late to the party but more reliable). Everything runs on end-of-day data. No intraday strategies currently, though you can build day-trading systems off daily data.

For the precious metals trade in 2025, he scaled into long positions in thirds: 50-day channel plus one ATR, 100-day plus one ATR, 150-day plus 1.5 ATR. Trailing stops at different distances. On the short side, an intraday order shorts when the commodity drops below its 10-day exponential moving average minus one ATR — a quick risk reducer after parabolic moves.

Risk as Religion

“I think risk is the most important part to be aware of in trading, by far.”

At 15% compounded annual growth, you double your money every 4.7 years. Keep drawdowns below 10% and you sleep well. He cites Tom Basso from the New Market Wizards — good bucks and a great life beats huge bucks and a terrible life.

The math works differently as accounts grow. A 35% drawdown on a small account is unpleasant. On $5 million, it’s existential. Someone approaching retirement with that kind of drawdown might need to go back to work. So Bensdorp plays defense — targets lower returns with much lower volatility, and lets compounding handle the rest.

The Lifestyle Design

Everything runs on a virtual server. Custom software generates orders, executes them through the broker’s API, and sends confirmation emails. He doesn’t even need to open his broker platform daily — and considers that an edge in itself.

“Having a look at your broker platform and seeing that P&L on the left side of your Interactive Brokers account and you see it going up and down like that… it’s not necessarily that helpful anymore. That’s actually your biggest enemy, I think.”

He lives in three places worldwide. Full automation enables that. The team includes a business partner handling the tech side and three additional programmers managing nuances like netting orders across strategies (when strategy A says buy 100 shares of Apple and strategy B says sell 20, the broker needs to see a single order for 80 shares long).

On AI in Trading

He uses AI extensively for research and idea generation, and his programmers use it for coding. But he’s skeptical of AI as a price predictor — same skepticism he had about neural networks in the ’90s.

“AI does not know what the price is going to be of the stock tomorrow, but for increasing your processes… it’s phenomenal.”

The important caveat: AI wants to please you. Bad prompts get confident-sounding garbage. Same principle applies to strategy development — the better you define what you want, the better the output.

Claude’s Take

Score: 7/10

Solid interview with practical substance, though it stays at the philosophy level more than the implementation level. Bensdorp’s core ideas are battle-tested and genuinely useful: build a portfolio of strategies designed for different market regimes, equal-weight them, keep individual strategies simple, think in risk first. The 2011 drawdown story is a good illustration of how real experience shapes trading philosophy.

What elevates this above a generic systematic trading interview is the emphasis on wanting ugly equity curves for individual strategies. That’s counterintuitive enough to be sticky, and it’s the kind of insight that only comes from actually running dozens of strategies simultaneously.

The interview doesn’t go deep on specific rules, parameters, or code — which is understandable given it’s a podcast, not a masterclass. If you already think in terms of strategy correlation, regime-dependent performance, and portfolio-level risk, you won’t find much new here. But the framing is clean and the mental models are well-articulated.

The AI discussion is the weakest section — well-meaning but surface-level. The lifestyle angle is interesting but a bit self-promotional. Overall, a good reference for anyone thinking about moving from single-strategy to multi-strategy portfolio construction.