What the Logbook is, and what a real entry will look like
The Logbook is the part of Midas Lab where I share the bot’s actual track record. It’s separate from the regular Posts section, which is mostly engineering and research. The Logbook is just the receipts.
Why split it out
Most retail trading content publishes the wins and quietly forgets the losses. I want this archive to be the opposite — a record that includes everything, so that over time the honest signal-to-noise of the bot’s performance is visible to anyone who looks.
Splitting it from the main Posts section also helps the legal framing. Engineering writing is engineering writing. Performance numbers are something separate, and they come with a different set of disclaimers attached. Keeping them in their own section makes it easy to be careful about what each one says.
What a real entry will include
Each future Logbook entry will cover:
- A defined period. Typically a calendar month, but sometimes a specific experiment window.
- A specific strategy or experiment. Not “the bot in general,” but “the breakout strategy in May” or “the volatility filter A/B test.”
- The numbers. What the bot did, what it would have done if I hadn’t intervened, win/loss counts, average sizes, drawdowns. Where I have a benchmark (S&P 500, equal-weight) I’ll include it.
- What I think happened. This is the part most published trading content skips. Was it the strategy, or was it the market regime? Was it skill, was it luck, or do I genuinely not know yet?
- What changed afterward. If the entry led me to tune something, retire a strategy, or add a guardrail, that’s noted.
What the entries will not include
- Live signals. I won’t be publishing anything you could trade off of in real time. By the time you read a Logbook entry, the trades are months old and the conditions that drove them aren’t around anymore.
- Specific parameter values. General shape of the strategy, yes. The exact numbers, no — those are the proprietary part.
- Promises. The bot is a research project. It might keep working. It might stop. Any entry that reads like a prediction is one I’d want to take back.
Coming up
The first real entry will cover the bot’s behavior over April 2026 — a month where the breakout strategy had its best results to date and the mean-reversion strategy had its worst. Which is convenient narrative-wise. I’ll post that when I’ve finished the analysis.
In the meantime, if you want to follow along, the RSS feed covers both Posts and Logbook entries.
Reminder, because it needs to be reminded: this site is for entertainment and educational purposes only. It is not investment advice. Always consult a licensed financial professional before making any trade. Past performance does not predict future results. None of the strategies described here have been independently audited and you should assume all numbers reflect a personal hobby project with all of the limitations that implies.