Historical prices are useful for understanding what has already happened. They show trends, volatility, drawdowns, and how the market reacted to past events. But they are usually a weak standalone tool for predicting future returns.
The main reason is that most short-term price movement is driven by new information, changing expectations, investor positioning, liquidity, and random order flow. Those influences are difficult to forecast from the historical price path alone.
A chart shows the final result of buying and selling activity, but it does not explain why that activity occurred.
Price Is an Output, Not an Explanation
The same 5% decline can reflect very different realities. A company may have a weaker long-term earnings outlook, interest rates may have moved higher, a large fund may have been forced to sell, the whole market may have gone risk-off, or investors may simply have expected even better results.
The chart records only that the price fell. It does not reveal which explanation is correct, and that distinction matters because the likely next move depends heavily on the cause. A temporary liquidity-driven decline may reverse quickly, while a decline caused by permanently weaker business prospects may continue.
Everything in the Price Does Not Mean the Chart Predicts the Future
A market price is the final output of many variables: expected revenue and earnings, interest rates, risk appetite, investor positioning, hedging activity, liquidity, sentiment, portfolio constraints, and newly released information.
Technical analysis observes that output while usually ignoring the underlying inputs. The price contains information, but it contains that information in a highly compressed form. It does not show how much of the move was caused by fundamentals, discount rates, positioning, sentiment, or temporary trading flows.
Predicting the Price Level Is Not the Same as Predicting the Return
If known information is already reflected in today's price, tomorrow's return depends mainly on new information, changes in expectations, unexpected economic developments, and future buying and selling pressure. Those are not visible in yesterday's chart.
That is why a model can appear to predict tomorrow's price with high accuracy simply because today's price is already close to tomorrow's price. Yet the same model may be almost useless at predicting whether tomorrow's return will be positive or negative by enough to justify taking risk.
The Predictable Component Is Tiny Relative to Normal Noise
Consider a simplified example. If the S&P 500 is at 6,000 and typically moves about 1% on a normal day, that is roughly 60 points of ordinary fluctuation. If a technical signal has a real edge of 0.02%, the expected advantage is only about 1.2 points.
This is why strategies with a small positive expected return can still show long losing periods, unstable backtests, and large drawdowns. The signal may be real, but it is small compared with the surrounding noise.
Historical Returns Are Noisy Labels
When a market rises 1.5% on a particular day, we do not know how much of that move was predictable in advance. Perhaps the market had a small positive expected return and an unexpected announcement accounted for most of the gain.
Backtests see only the realized move. They often treat that noisy historical outcome as a clear answer about whether a prior setup was right or wrong, even though most of the move may have resulted from information that was impossible to know when the signal appeared.
Markets Offer Fewer Independent Observations Than They Seem To
Twenty years of daily prices may sound like a large dataset, but those observations are not independent experiments. Many technical strategies reuse overlapping data, hold positions through the same regimes, and generate many trades within only a few extended bull markets, bear markets, and volatility cycles.
A strategy using a 200-day moving average, for example, reuses almost the same information from one day to the next. The true amount of independent evidence can therefore be much smaller than the raw number of price points suggests.
Markets Change Over Time
Even a relationship that was real in the past may fail in the future. Markets change because of monetary policy, regulation, technology, algorithmic trading, passive flows, options activity, lower transaction costs, and the wider adoption of the strategy itself.
Markets are not passive scientific systems. Participants observe patterns, adapt to them, and react. Once many investors discover and trade the same setup, they may move the expected return earlier or eliminate it altogether.
Price-Derived Signals Need an Underlying Explanation
Momentum, trend following, short-term reversal, and volatility effects may still contain useful information. But the stronger versions of those strategies usually have an underlying explanation, such as slow information diffusion, institutional repositioning, temporary liquidity pressure, or persistent volatility regimes.
The important question is not simply what the price did before. It is why that historical movement should contain information about future expected returns. Without that explanation, a chart pattern on its own is usually a weak basis for forecasting what happens next.
Build better investment research on more than price alone.
Explore CatCapital ResearchCatCapital research outputs are for informational purposes only and are not financial advice. Investors should make their own decisions and consider their own objectives, constraints, and risk tolerance.