•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Most investors assume the market is chaotic—unpredictable and impossible to “solve.” But TradeSmith CEO Keith Kaplan argues that markets may follow repeatable patterns, similar to how a complex puzzle can be solved through consistent sequences.
In a Friday Digest takeover, Kaplan describes an AI-powered trading system built to identify high-probability trade setups by analyzing millions of potential trades. He says the system is designed not to predict outcomes, but to flag “signals” that have historically produced consistent results.
Kaplan says each signal in the new system is constructed from a specific combination of factors, including technical indicators, price patterns, and market conditions. When the factors align, the signal “fires.”
He describes the approach as a way to capture relationships that are difficult for human analysts to identify, given the large number of variables involved in stock price movement.
Kaplan says TradeSmith’s engineers spent the last 12 months developing the system. He reports that it evaluates 2.09 million potential trades every day to find alignments across different market environments.
According to the company, when certain combinations of factors align, the resulting trade setups show 90% or better historical accuracy. Kaplan adds that these alignments repeat across bull markets and bear markets, as well as during crashes and recoveries.
Kaplan cites a one-year backtest in which a portfolio of signal-based trades outperformed the S&P 500 by roughly 3-to-1.
Invesco Ltd. (IVZ): Kaplan says two factors had to exceed 80 for the signal to fire: Bollinger Percent B and the Money Flow Index. He reports the IVZ signal produced an 18.8% gain in 11 days.
Lam Research Corp. (LRCX): Kaplan says the signal required two conditions: the stock had to close above its 200-day moving average, and that close had to occur two trading days before a market holiday. He reports that after the signal fired on August 28, 2025 (two trading days before Labor Day), LRCX gained 11.4% in 15 days, with a historical accuracy rate of 86%.
Kaplan also discusses results from a trader, Mike, who previously worked in code for the U.S. Air Force and in cryptography for the Pentagon, and later co-managed mutual funds and high-net-worth accounts worth up to $200 million. Kaplan says Mike holds the Chartered Market Technician designation, held by fewer than 5,000 people worldwide.
Kaplan reports that of the top 100 trades Mike posted, the average gain was 2.6% over nine trading days, while the S&P 500 rose 0.4% over the same period—described as roughly a 7X return versus the market.
Kaplan adds that the average 2.6% return over nine trading days is described as equivalent to a 73% gain across a full year.
Kaplan lists reported outcomes when trading signals with options:
Kaplan says he plans to demo the system at an upcoming launch event, describing a test in which Mike would invest real money in whatever stock the signals system flagged on a given morning, without researching the company or checking earnings or news. Kaplan says the activity was recorded and will be shared with attendees.
Kaplan says the event will explain how the signals system works, including the factors it tracks and the trades it flags, and he frames the upcoming weeks as a potentially favorable environment for signals trading.
The event is scheduled for Wednesday, April 22, at 10 a.m. Eastern. Kaplan also states that registration includes immediate beta access to the signals software, with the ability to explore signals across 2,467 stocks before the event.
Kaplan describes TradeSmith as a financial technology firm based in Baltimore. He says he runs a team of 65 people with an annual budget of $8 million to develop hedge fund-level analytical systems for self-directed investors.
Kaplan reports that more than 134,000 people in 86 countries use TradeSmith software to manage over $29 billion in assets. He also references existing products including TradeStops, which he says is designed to help investors determine ideal times to sell stocks, as well as tools that identify hidden seasonality patterns, find undervalued options plays, and use AI to forecast stock moves up to 21 trading days.
Kaplan concludes by noting that a small group of beta users tested the system ahead of launch, including one user, Edward V., who reported a perfect success rate on every closed position, and another, John M., who called it a “game-changer.”
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
