The premise
Cross-sectional characteristics — cheapness, recent relative strength, profitability, low volatility — have historically been associated with return differences that persist across decades and geographies. Constructing a portfolio tilted toward them is entirely mechanical.
The uncomfortable part
Factor premia disappear for a decade at a time. Value underperformed for roughly ten years before reasserting; participants who abandoned it in year eight experienced the full loss and none of the recovery. The strategy's edge and the investor's endurance are separate things.
Why it suits automation better than a person
Rebalancing on schedule regardless of how the last quarter felt is the entire discipline, and it is the thing humans reliably fail at. A machine rebalancing into what has just underperformed is doing the strategy; a human doing it is fighting themselves.