A free weekly letter for self-directed investors. It covers one idea per issue in 5 to 8 minutes: a written process, machine learning with clear limits, and risk rules set before any trade.
Most investors are either on autopilot with no way to adjust for risk, or chasing headlines with no flight plan. This letter gives you a process instead of another prediction.
Trades start from headlines or gut feel, with no written rule for entry, exit or size.
Tools hint that they predict the market, and nobody explains what the model actually does.
Nice numbers with hidden assumptions and no worst case shown.
A new strategy feels risky when you can't see the limits before you start.
Every decision checks market regime, model evidence and portfolio risk first. The reasoning is logged before you act.
The model scores whether today looks like conditions where a setup worked before. It can block a trade. It can never start one.
Every trade, price and assumption is published, along with the worse result next to the better one.
Size caps, cash reserves and loss cuts are set by rule, not by willpower on a bad morning. See below.
It's the right question to ask. Here's what keeps a single bad trade or a bad month from turning into a disaster. Each rule is applied by the system, not left to discipline.
No single position can take more than one third of the account.
Exposure is capped, so the account is never all-in on one idea.
A minimum reserve always stays in cash, and it grows when volatility rises.
A short option is closed once its loss reaches twice the credit received, so small losses don't become big ones.
A trade must clear every gate, including trend, volatility, regime and ML confidence. Eleven entries in 5.5 years: most weeks, nothing needs doing.
After a losing stretch, the system takes less risk automatically instead of trying to win it back.
A long LEAPS call can never lose more than what you paid for it. You know your maximum loss on that position before you enter.
QQQ holds 100 companies, so there's no single-company blowup, and pricing is tight with easy exits.
TradeMind never connects to your broker, places orders or holds your money. You decide and place every order.
QQQ Defensive needs no options approval, and the first month is refundable, so you can watch the process before committing.
| Worst drop, shown openly | Max drawdown | Time to bottom |
|---|---|---|
| TradeMind QQQ LEAPS, model backtest 2021 to 2026 | -14.5% | About 2 months |
| TradeMind, 15-month real-quote test | -30.4% | About 6 weeks |
| QQQ buy and hold, same 2021 to 2026 window | -35.6% | 14 months |
We publish the number that hurts, not just the one that flatters. These are hypothetical, backtested or simulated results, not live accounts. The next drawdown could be deeper. Options can lose the entire amount invested.
Together they explain what this letter believes: process over prediction, models with limits, and numbers you can audit.
Problem: another forecast doesn't fix a missing plan. You get: what a written process should include before any trade.
Read the sample issue →Problem: AI is sold as a crystal ball. You get: where the model helps (ranking and vetoes) and where it never goes (starting trades).
Read Issue 6 →Problem: results are shown without assumptions. You get: a checklist for reading ours and any other backtest.
Read Issue 7 →Confirm your email and you get 30% off a one-year TradeMind subscription, valid for 90 days after you confirm. Unsubscribe anytime.
Subscribe freeTradeMind is software for self-directed investors, not investment advice and not a broker-dealer. All newsletter content is educational. Options involve substantial risk and are not suitable for every investor; you could lose some or all of your invested capital. Performance figures are hypothetical or simulated, have inherent limitations, were not achieved by any actual account and do not guarantee future results. You make every decision and place every order at your own broker. Risk Disclosure · Terms · Privacy