Why I Automated My Trading with Claude AI | Thinking in Systems (Pt 1)

Summary

The presenter explains that he automated his trading to reduce the time and emotional strain of discretionary trading while balancing a full-time job, an investment-property renovation, and personal commitments. He says he had traded discretionarily for about 3.5 years and was earning payouts through prop firms, but found that watching charts and making rushed decisions interfered with his free time and affected his mood.

Despite having no programming background, he used Claude to help build automated systems. The systems execute trades according to predefined entry criteria, risk settings, stop-losses, and take-profits. He receives trade notifications by email and on his phone or watch.

He says he now spends around 30–60 minutes a day on trading if he chooses, rather than needing to monitor charts. Previously, he sometimes spent 6–8 hours waiting for setups.

Systems and Reported Results

The presenter says his dashboard, called Arc Algos, tracks accounts, profit factor, win rate, trade count, and other statistics. He reports 212 live trades across two systems:

• “15-minute fair value gap” — described as a reversal strategy from zones.
• “Flipper” — described as more trend-continuation oriented.

He reports approximately 55% winning trades and a 1.26 profit factor for the systems. His earlier discretionary win rate reportedly ranged from 50% to 80%, depending on the account and his emotional state. He says he did not keep sufficiently complete records to estimate it accurately.

He also reports that he:

• Received a payout after four profitable days on a $50,000 funded account.
• Had a profitable day on another account.
• Passed a $25,000 assessment in about three weeks.
• Passed a $50,000 assessment over several weeks.

He says the automated accounts and assessments were run at what he considers minimal risk, without moving to a higher-risk approach. No payout amounts are provided.

These are the presenter’s own reported results; the video does not independently verify them.

Methodology and Trading-System Lessons

The presenter’s main lessons include:

• Translate discretionary ideas into specific, measurable rules. Vague judgments such as “the candle looked impulsive” need operational definitions—for example, candle-body-to-wick proportions, consecutive candles, or a reaction to a fair-value gap.
• Define the relevant details of a setup, such as the gap’s timeframe, whether it is an inversion gap, and its size.
• Favor simple systems with fewer conditions. He says two or three criteria worked better for him than requiring around ten.
• Use fixed risk, stop-loss, take-profit, and entry rules; do not take a trade if the criteria are not met.
• Test strategies on historical data and gather enough trades to assess them. He says he developed more than 50 strategies, of which only about four or five worked out.
• Expect that some intuitively appealing discretionary ideas will fail when tested or converted into explicit rules.
• Consider combining systems with negative correlation, so one may perform better when another struggles.
• Track results systematically. The presenter says automation made outcomes more consistent under the same conditions and easier to measure.

Performance Metrics and Cautions

The presenter argues that a very high profit factor is not necessary for an automated strategy. He cites roughly 1.10–1.20—and, in some prop-firm contexts, 1.10—as potentially sufficient, while acknowledging that commissions and slippage matter.

He recommends testing thousands of trades across multiple years rather than relying on impressive results from a small sample. He says a year with a 1.05 profit factor can occur within a three- or four-year period. These thresholds are his opinions, not universal standards or guarantees of profitability.

He contrasts discretionary and automated trading: discretionary trading may make more in the short term when the trader varies risk, while automation may offer more consistency and less emotional burden over the long term. Neither approach is presented as guaranteed to make money.

He also emphasizes that prop-firm trading depends on the account’s rules and the math of money in versus money out. Traders should choose a strategy suited to their financial circumstances. Algorithmic systems cannot capture vague contextual judgments unless those judgments are converted into precise rules.

Assets, Firms, Tools, and Disclosures

• Instruments and assets: The subtitles discuss futures and trading generally but name no specific ticker or contract.
• Tools and services: Claude; PickMyTrade, which the presenter says he used to connect or execute the system; and the Arc Algos dashboard.
• Prop firms named: Apex, My Funded Futures, Topstep, TradiFi, and Lux.
• Presenter and sources: The presenter is not named in the subtitles; the channel is Fullview Trading. He refers viewers to his Fullview Trading posts on X/Twitter for performance records and mentions Claude and PickMyTrade as tools used in the process.

The presenter explicitly says, “This is not financial advice,” and frames the video as a description of what he did. Futures and prop-firm trading can involve substantial risk, and the video provides no guarantee of future performance.