Calling Tops and Bottoms

The entire structure of market timing rests on a single assumption: you can identify when the market has reached a top (peak) or bottom (valley) and act on it before it bounces. In practice, this is harder than it appears. Tops and bottoms are only recognizable in hindsight. In real time, they feel like ordinary days in a continuing trend.

Quick definition: A market top occurs at the peak price before a decline; a market bottom occurs at the lowest price before a rally. They are unrecognizable in real time and typically only identified days, weeks, or months after they've occurred.

Consider the March 9, 2009 bottom of the financial crisis. The S&P 500 fell 57% from peak to trough. On March 9 itself—the exact bottom—the market closed down that day, having fallen another 3% in the preceding two weeks. There was no announcement saying "This is the bottom." In fact, the media was filled with warnings that further declines were possible. Bank stocks were in free fall. Housing prices were still falling. Unemployment was rising.

An investor with perfect information about the economy would have still struggled to know that March 9 was the bottom at the time. The bottom was only apparent weeks or months later, after markets had rallied and confidence began to return. By that time, much of the move was already priced in.

Key Takeaways

• Market tops and bottoms are unidentifiable in real time; they're only clear in hindsight
• The worst days and best days often occur before the psychological or technical signals that would trigger a trade
• Research shows that traders who attempt to call bottoms are typically wrong—they call bottoms repeatedly on falling markets, buying progressively on the way down
• Attempting to identify reversals with technical analysis fails because price patterns that worked in the past are not predictive of future patterns
• Even if you catch the right day, you still face the problem of knowing when to re-enter (for tops) or exit (for bottoms)—two decisions, both of which must be correct
• Confirmation bias causes traders to see the signals they expect, explaining why systems can feel accurate in retrospect but fail prospectively

The Two-Decision Problem

Market timing isn't a single decision—it's two. You must decide when to exit and when to re-enter. Miss on either decision, and the strategy fails.

Consider an investor who successfully exits the market on January 15, 2020, just before the COVID crash. Excellent! They avoid the 34% decline. But then they face the harder problem: when to re-enter? The market has fallen 34%, fear is everywhere, financial media is discussing depression-era comparisons. The psychological urge to buy is minimal. When do they actually invest?

• If they buy March 9 (the exact bottom): They're right, but how would they know it's the bottom?
• If they buy March 23 (two weeks later): The market has already rallied 10% from the bottom. They've given up 10% of the move.
• If they buy April 2020 (another week later): The market has rallied 15%. They've given up 15%.
• If they buy June 2020: The market has rallied 35%. They've given up 35%.
• If they buy August 2020: The market has rallied 50%. They've given up 50%.

The problem becomes clear: even if you get the exit decision right, the re-entry decision is nearly impossible to get right at the same time. The moment that feels psychologically safe to buy (after the market has already recovered 20–30%) is far from the moment that's optimal (at the actual bottom, when fear is highest).

The Challenge of Real-Time Identification

A mountain peak is unmistakable when you're standing on top of it. A market top is not. When the S&P 500 reached its all-time high on January 3, 2022, at 4,765 points, nothing about that day indicated it was a top. The market was strong. Earnings were good. The Federal Reserve was still providing liquidity. There was no sign in real time that this was the last day of the bull market for 10 months.

A top only becomes apparent when the market has fallen 10%, then 15%, then 20%. By then, the opportunity to sell "at the top" is gone. You can sell into a decline, but that's not selling at the top—that's selling after significant losses have already occurred.

The same applies to bottoms. When the market fell 50% in 2008, the bottom was a particular day (March 9, 2009). But on that day, there was no certainty it was the bottom. The previous bottom in late 2008 had been followed by further declines. Investors who bought what they thought was a bottom in November 2008 got another chance to be wrong in January 2009 when prices fell further. Calling a bottom on the first attempt is almost impossible.

The Technical Analysis Problem

Many traders attempt to call tops and bottoms using technical analysis—identifying chart patterns that supposedly predict reversals. The promise is appealing: if you can spot a head-and-shoulders pattern, a double bottom, or some other formation, you can time your trades to reversals.

The evidence is clear: this doesn't work on live data at sufficient frequency to overcome fees and taxes. A 2020 analysis by researchers at Tulane University tested 29 common technical trading strategies on daily data from 1960 to 2020. After accounting for transaction costs, all 29 underperformed buying and holding. The reason: patterns that work in backtests fail prospectively because:

1. Overfitting: Historical data can be mined for patterns, but those patterns are often coincidences. When you test 1,000 patterns on historical data, some will appear predictive purely by chance.
2. Market Adaptation: If a pattern were truly predictive, traders would exploit it, and the pattern would disappear. The moment a pattern becomes known, it's traded away.
3. Look-Ahead Bias: Backtests can incorporate information unavailable in real time. A chart pattern that looks clean in a historical chart might be ambiguous in real time.

The Anchoring Problem

Investors who attempt to call bottoms often fall victim to anchoring. They anchor to the previous peak price and assume that as a "fair value." When the price falls 40% from the peak, they feel it's cheap and buy. Then it falls another 40%, and they feel betrayed—this wasn't supposed to be cheap.

A 2010 study found that investors who experienced losses were more likely to hold losing stocks not because of fundamental reasons but because they were anchored to their purchase price. They'd call bottoms prematurely, become psychologically attached to their analysis, and then refuse to sell even as fundamental deterioration occurred.

During the tech bubble's collapse from 2000 to 2002, this was rampant. Investors who bought at $50 per share and watched stocks fall to $10 felt they'd found value and were buying bottoms. But the stocks continued to $2 or $0.50, proving that there was no meaningful bottom—only further deterioration.

Real-World Examples

Case 1: The Repeated Bottom Calling. An investor identifying what they perceived as bottoms called the following "buys" over a 24-month period:

• September 2008: "The market has fallen 25%, this is a bottom." (Market fell another 32% over the next 6 months.)
• January 2009: "Now we've fallen 50%, this is definitely a bottom." (Market fell another 7% in the next month.)
• February 2009: "Surely this is it." (One week later, market bottomed.)

By waiting and holding capital, they would have invested at the true bottom. By trying to call multiple bottoms, they deployed capital inefficiently, anchored themselves to failed predictions, and likely became overconfident on a later call when luck was with them.

Case 2: The 2018 Volatility Panic. In late 2018, a trader convinced that the market had broken used technical analysis to identify what they called "lower lows" and sold in fear of further declines. The market fell 19%, validating their call. They stayed on the sidelines for January 2019 as the market rallied 10%, then February as it rallied another 8%, then March as it rallied 7%. By the time they re-entered, they'd given up 25% of the recovery move.

Case 3: The COVID Tragedy. An investor successfully avoided the February-March 2020 crash, staying in cash through the 34% decline. But then they waited for "confirmation" that the crisis was over. They didn't invest until May 2020, by which time the market had already recovered 25% of its losses. Their successful exit was almost entirely undermined by their poor re-entry timing.

Common Mistakes

Mistake 1: Assuming the Present Resembles the Past. Technical traders often assume that if a head-and-shoulders pattern preceded a crash in 2008, seeing it again in 2023 means a crash is coming. But market structure changes. Algorithms, ETFs, and index funds have altered the mechanics of price discovery. Patterns that worked 20 years ago don't work today.

Mistake 2: Mistaking a Relief Rally for a Real Bottom. A market can bottom, rally 10–15% (a relief bounce), and then fall further. Traders interpreting the relief rally as confirmation of the bottom often sell into further declines, crystallizing losses twice.

Mistake 3: Forgetting That Bottoms Often Feel Scary. The psychological reality is that market bottoms feel terrifying. If a bottom felt comfortable and obviously attractive, it wouldn't be a bottom—it would have already been bought and the market would have rallied. True bottoms are scary enough to trigger panic, which is why most investors sell at them rather than buy.

FAQ

Q: Can't you use moving averages or other technical indicators to identify reversals? A: No, not reliably. A 2013 study by MIT found that 99% of published trading rules perform worse in live trading than in historical backtests. The reason: all technical indicators incorporate past price information. If they were predictive, everyone would use them, they'd be traded away, and they'd stop working.

Q: What about using multiple signals—if you need several conditions before making a trade, doesn't that reduce false signals? A: It reduces frequency of trades, but not the accuracy of trades. A study of traders using complex rule-based systems found that adding more conditions to the rules increased complexity without improving out-of-sample performance. The extra conditions often just added lag, causing traders to miss reversals or enter after moves had begun.

Q: Isn't there a difference between calling a top/bottom and simply using a rebalancing rule to trim winners and buy losers? A: Yes, and that's a crucial distinction. A rebalancing rule is systematic and doesn't require timing. You've decided in advance to trim when positions drift 5% above target, not to time market peaks. That's far more defensible and successful than discretionary calls.

Q: What about calling an intermediate-term bottom, like within the next 6 months? A: This is still timing, and the evidence shows it fails. A 2015 study found that investors attempting to identify 6-month turning points underperformed buy-and-hold investors, even when their directional calls were correct 55% of the time. The reason: the opportunity costs and reduced compounding during out-of-market periods exceeded the benefits of avoiding declines.

Q: If I've successfully called bottoms before, doesn't that mean I have skill? A: Probably not. A 2018 study found that investors who had successfully called one or two reversals were not more successful on subsequent calls. The earlier successes were likely luck, and the investor's overconfidence from those successes often led to worse performance on future attempts.

Q: Doesn't even calling a bottom on the third attempt at least beat sitting in cash? A: Only if you invested in cash that earned nothing. If you'd stayed invested and earned the market return over the same period, even a successful third-attempt bottom call often fails to make up for missed compounding during your three attempts.

Related Concepts

• Mean Reversion: The tendency for extreme prices to revert toward average; however, knowing when reversion will occur is the unsolved problem
• Support and Resistance: Price levels that technical traders believe act as barriers; however, research shows these are often self-fulfilling prophecies that fail in real trading
• Overfitting: Fitting a model to historical data so precisely that it captures noise rather than signal, causing it to fail prospectively
• Hindsight Bias: The tendency to see past events as having been predictable, making past tops and bottoms seem obvious when they were actually unknowable
• Slippage: The difference between expected execution price and actual execution price, which often exceeds expected gains from successful timing calls

Summary

Tops and bottoms are invisible in real time. By definition, they're only recognizable in hindsight. An investor attempting to call them faces not one challenge but two: identifying the reversal point and then knowing when to re-enter the market at a favorable moment. Missing on either decision undermines the entire strategy.

Technical analysis, which promises to solve this problem, fails systematically once transaction costs are factored in. The evidence is conclusive: traders using chart patterns and technical indicators underperform buy-and-hold investors even when their directional calls are correct 55% of the time.

The path to wealth is not paved with perfect timing calls. It's paved with consistent investment, patience, and the willingness to let compounding work across decades, even through declines.

Next

We'll examine a fascinating real case study—an investor who attempted to time every major market move for over 40 years and whose results provide a cautionary tale about the cumulative cost of even occasional timing errors.