We frequently use randomization studies as a benchmark to evaluate market and trading dynamics. The results for 2024 confirm what we already knew: long traders profited and short traders faced large losses on average.
In the randomization studies, we use daily data for the large-cap index ETF (SPY) and a fair coin. We toss the coin at the close of each day. The position signals are long if heads show up and short if tails show up. For long-only or short-only trading, tails and heads are signals to close positions, respectively. We consider random traders with $100K initial capital. We do not use profits for position sizing. Commission and slippage are $0.01 per share. We make 10K simulations and compute the statistics.
Case 1. Long-short trading
The simulation shows that 47.6% of the long-short random traders generated a profit but only 1.77% more than buy and hold (about 26%). Some long-short traders were very lucky and generated profits up to 49.23%, but some were very unlucky and lost up to 38.40%. A 10% return ranked at about the 80% level, while ranking at the 95% level (possible skill) required a 19.9% return.
Lesson: Overall, given the statistics, long-short trading was close to a zero-sum game, as expected, and the profit distribution was nearly normal with a slightly thicker right tail. The average return was negative, at -0.27%, confirming that long-short daily trading is a slightly negative-sum game after commissions and slippage. Note that this is not true in the case of long-term trading. Short-term traders must take into account the directional market bias; otherwise, they are involved in a zero-to-negative-sum game.
Case 2. Long-only trading
As expected, the simulation confirmed that the overwhelming majority of long-only traders were profitable this year due to the strong upward drift of the market. More than 97% of long-only traders profited. However, only 1.07% made more than buy and hold. The mean return was 12.04%, and a 10% return ranked at the 37% level, while ranking at the 95% level (possible skill) required a 22.4% return.
Note that the minimum return for significance at the 95% level is higher for long-only traders than long-short. This is because the distribution for long-only trading has a slightly positive kurtosis versus a negative for long-short trading.
Lesson: Trading along the trend has higher odds for profitability, but proving skill is also more difficult. In fact, we argue that in the stock market, proving skill is not possible. On the other hand, in futures and forex markets, proving skill based on outperformance is possible. In addition, things get more complicated because in long-only stock market trading, even underperformance cannot rule out skill. In other words, proving skill may be possible for short-only traders, even if there are losses, as shown below.
Case 3. Short-only trading
In the case of short-only traders, the maximum return was 10.49%, and the lowest was -30.21%. Only 1.88% of short-only traders profited, which shows the perils of going against strong bull markets. A 10% return was highly significant, and a negative return of 2.67% ranked at the 95% level.
Lesson: This means that a short-only trader who managed to end the year nearly flat was actually very skilled. The majority of unskilled short-only traders faced large losses.
Comments
First, those who don’t understand randomization studies say random trading statistics are useless. However, when a trader initiates a trade, the market has no idea whether the signal came from some fancy algorithm with a potential edge or after tossing a coin. Therefore, “locally,” all trades are indistinguishable from random. Traders with an edge have some filters—a.k.a., methods, systems, or strategies—that compress the sequences of random trades to ones that maintain an edge. The sequences are also indistinguishable from random trading sequences that rank high in the distribution. This is how the value of randomization studies arises. It may sound like alchemy, but ultimately, trading, market analysis, and finance in general are a sort of alchemy. The best alchemists who can transmute signals into profits win.
AI comment: “Successful traders possess a unique ability to transform seemingly random data into profitable insights, much like alchemists turning base metals into gold. Ultimately, it highlights the skill and intuition required to navigate the complexities of financial markets effectively.”
Subscribe below for notifications of new posts and updates from the Price Action Lab Blog and get the free PDF book “Profitability and Systematic Trading” (Wiley, 2008).
Premium Content
By subscribing, you have immediate access to hundreds of articles. Premium Articles subscribers have immediate access to more than two hundred articles, and All in One subscribers have access to all premium articles, books, premium insights, and market signal content.
Disclaimer: We only provide the articles for informational purposes, not as investment advice or actionable content. We do not warrant the accuracy, completeness, fitness, or timeliness for any particular purposes of the articles. You should never treat the articles as financial advice. The author of this website is not a registered financial adviser. Before subscribing, please read our Disclaimer and Terms and Conditions.
If you found this article interesting, you may follow this blog via RSS, email, or Twitter.
