|
Search Google for “SPY trading system” and you will quickly encounter a strategy called the “3 Lines of Code” system from a site called Relaxed Trader. The headline statistic is hard to miss. 82% winning trades. The comments section shows the usual reaction. Traders ask how to get the code, whether it works on leveraged ETFs, and how to implement it on platforms like TradingView. It feels like the discovery of something powerful. High win rates and simple rules have a way of creating that impression. After spending years building and testing SPY trading systems myself, I was curious how this one performs when replicated using clean data and a dedicated backtesting environment. The answer is instructive. Not because the system is terrible. It is not. The real lesson is how easily a single headline statistic can hide the metrics that actually determine whether a strategy is attractive. The StrategyThe rules are genuinely simple. In TradeStation syntax they amount to three conditions. Only buy when SPY is above its 66 day moving average Enter when today’s close is the lowest close of the last three days Exit when today’s close is the highest close of the last nineteen days There are no stops, no short selling, and no position sizing rules. Conceptually the system is straightforward. Uptrend, buy the dip, sell the rebound. This is not a ridiculous idea. Short term mean reversion within longer term uptrends is a real market behavior and many legitimate strategies exploit exactly this dynamic. So the issue here is not the strategy itself. The issue is how the results are presented. What the Page ShowsThe presentation highlights a few numbers. • 82 percent win rate • 503 trades • A steadily rising equity curve • Average winners and losers reported in dollar terms At first glance it looks excellent. However several important metrics are missing. • Maximum drawdown (a big one) • Annualized return • Risk adjusted performance such as Sharpe ratio • Percentage based average win and loss • Time based equity curve • Benchmark comparisons These are not advanced statistics. They are the basic tools used to evaluate systematic trading strategies. Without them you cannot see the full risk profile of the system. The equity curve presented on the site looks like this: Wow. That has got to be one incredible system. What the Numbers Actually Look LikeRunning the strategy on clean SPY data from 1993 through 2024 (as tested on the site) produces results roughly like this. My equity curve takes on a different feel than the one presented above. Agree? Trades: 371 Win Rate: 83% Avg Winner: +1.44% Avg Loser: −2.84% Annualized Return: 8.8% Maximum Drawdown: −30.5% Sharpe Ratio: 0.84 One characteristic stands out immediately. The average loss is roughly twice the size of the average win. That is the tradeoff behind the 82 percent win rate. You win often, but when the system loses the losses are materially larger. This structure is not unusual. Many profitable systems operate this way. The key point is that the headline statistic does not reveal this relationship. The Capital Efficiency QuestionOne particularly revealing metric is Profit Per Bar, this is a metric built in to my Wealth-Lab platform that assumes a $5,000 position, then divides the sum of trade profits by the sum of the number of bars held by the trades. In simple terms, when your money is at risk, how hard is it working? Measured that way: System: about $2.86 per bar This comparison matters. A timing system enters the market only during specific conditions. The expectation is that those conditions represent better than average opportunities. If you are choosing when to participate in the market, the days you select should ideally be more productive than the average day of simply owning the index. Otherwise the timing filter is not improving exposure. It is simply rearranging it. But here the opposite relationship appears. During the days when the strategy has capital deployed, it generates less profit per day of exposure than passive buy and hold. That does not mean the strategy cannot make money. Clearly it does. What it suggests is that the system is not concentrating capital into unusually productive periods. Instead it is selecting trading windows that are, on average, less efficient than the market baseline. That changes the narrative considerably. Why the Equity Curve Looks So SmoothAnother subtle detail involves how the equity curve is presented. The chart on the site is plotted by trade number, not by calendar time. This compresses time dramatically. A six month drawdown appears no wider than a two week dip if both contain a similar number of trades. Periods when the system is completely out of the market simply disappear. A time based equity curve shows a more realistic path. The moving average filter does keep the system out of extended bear markets, which is exactly what it is designed to do. However the strategy still experiences meaningful drawdowns while it is active in otherwise favorable conditions. When plotted against real calendar time those periods become much easier to see. The Win Rate TrapHigh win rates are psychologically powerful. An 82 percent win rate feels safe. It suggests you are almost always right. But win rate by itself tells you very little. What matters is expectancy. Expectancy = (Win Rate × Avg Win) − (Loss Rate × Avg Loss) For this system: (0.83 × 1.44%) − (0.17 × 2.84%) ≈ 0.71 percent per trade That is positive expectancy, which explains why the strategy ultimately makes money. However spread across roughly a dozen trades per year it produces moderate annual returns while still exposing the account to drawdowns exceeding 30 percent. Once expectancy becomes visible, the emotional appeal of the win rate begins to fade. The One Number Traders Need to KnowPerhaps the most important statistic missing from the original presentation is maximum drawdown. Drawdown determines whether a trader can realistically stay with a strategy. Experiencing a 30 percent drawdown in real time with real capital is psychologically brutal. Many traders abandon systems during exactly those periods, often locking in losses just before recovery begins. If a system historically experiences drawdowns of that magnitude, that information should be clearly visible. Without it, traders have no framework for what the strategy will actually feel like to trade. Why These Metrics MatterWhen designing systematic strategies, these are the numbers that determine whether a system is actually tradable in the real world. A strategy may show profits in a backtest but still be unattractive if drawdowns are large, capital is deployed inefficiently, or risk adjusted returns are weak. In my own SPY research, the design process always starts with those constraints. The objective is not simply to produce profits in a historical test. The goal is to build strategies where the return profile, drawdown behavior, and capital efficiency make sense relative to the benchmark they are trying to improve upon. That is why a full statistical picture matters. Without it, traders are heading in to a very difficult game blinded by ignorance. That is not a trait helpful in this difficult game. The Bigger LessonThe Relaxed Trader strategy is not fraudulent. The rules are clear, the win rate is real, and the core idea is logically sound. However its popularity illustrates a broader issue. Many publicly presented trading systems emphasize the statistics that look impressive, while omitting the statistics that determine whether a strategy is actually viable. I encourage you to dig deeper in to that site and see if you can find other examples like I highlighted here. When evaluating any trading system, whether from Google, YouTube, or a newsletter, several metrics should always be visible. • Annualized return If those numbers are missing, the appropriate response is not necessarily skepticism. But it should at least be curiosity. Because without them, you're only seeing part of the picture. Trade carefully out there. If you ever have questions about a particular system, don't hesitate to reach out. I always do my best to reply back promptly. PS - I have 2 open slots left in the Bear Hunter Beta. That system triggered today at the open and had a good run. Have a Great Night! Dave Johnson - Quantitative System Designer at via Trading Time Machine https://ift.tt/1xSAa0j
0 Comments
Leave a Reply. |
Dave JohnsonI'm Dave Johnson, a former investment advisor and quantitative system developer with over 30 years of experience building and trading mechanical systems. These days I focus on rules-based research, honest backtests, and sharing what the data actually shows. Archives
June 2026
Categories |
RSS Feed