Confluence Strategies Beat Buy-and-Hold Only 26% of the Time: Which Combos Help and Which Filter Out Good Trades
Adding a second indicator sounds like risk management — but across 660,005 backtests, most combinations underperform simply holding the asset.
Stacking Indicators Costs More Than It Buys
The logic behind confluence is sound on the surface: if two signals agree, the trade should be higher quality. The problem is that every additional condition you add shrinks your trade sample, and that filter doesn't distinguish between low-quality setups and the profitable ones. It removes both.
Across 660,005 backtests covering 903 assets on 1-Hour, 4-Hour, Daily, and Weekly timeframes, only 26% of indicator combinations beat buy-and-hold after realistic transaction costs. Three out of every four confluence setups underperformed a passive hold. The default outcome of adding a confirmation filter is not better trades — it's fewer returns.
Win Rate Is a Trap
Confluence almost always improves win rate. That's exactly why it feels like progress. But win rate and edge against buy-and-hold are different things, and the data shows a sharp gap between them.
Holy Grail Confluence — a strategy designed around the premise of stacking signals — posts a median 73.3% win rate across tested assets. It beats buy-and-hold on just 8% of them. Murrey Math Lines hits a median 74.3% win rate and beats buy-and-hold on 11%. RSI Mean-Reversion reaches 71.7% wins and beats buy-and-hold on 10%. The pattern is consistent: the filter cleaned up the entry signal while stripping out enough of the large winners to destroy the strategy's edge against passive holding.
The moves that drive most of an asset's return tend to look messy on a chart — conflicting signals, choppy lead-up, no clean confirmation. Confluence filters those out first. You end up with a high win rate on trades that don't matter much and miss the ones that do.
The Combos That Do Work Are Asset-Class Specific
The 26% of combinations that do beat buy-and-hold are not universal. Every top-performing setup in our data is tied to a specific asset class. There is no two-indicator combination in our results that consistently wins across stocks, forex, crypto, and commodities.
In forex, the Fisher Transform leads the results. In stocks, Fibonacci Pivots and Projection Bands top the list. Crypto favors MA Envelope and Fibonacci Pivots. Commodities point toward Keltner Mean-Reversion and Laguerre RSI. Taking a combination that dominates in one market and applying it to another is not adding confirmation — it's adding unrelated noise.
No Smart Money Concepts indicator beat buy-and-hold in our dataset. SMC: Liquidity Sweep — one of the more widely used SMC setups — posts a median 71.2% win rate and beats buy-and-hold on 8% of tested assets. The narrative around SMC as a high-probability confluence framework is not supported by the backtest record.
These Are Hypothetical Backtests — Not Advice
Every figure in this article comes from hypothetical backtests on historical price data across 1-Hour, 4-Hour, Daily, and Weekly timeframes, with realistic transaction costs included. Out-of-sample testing was used to reduce in-sample bias, but backtested results can still reflect historical patterns that no longer hold.
This is not financial advice. No combination in this data guarantees future profit. The point of publishing these numbers is to give you a more honest baseline than forum posts or backtests that ignore costs — not to suggest any specific setup is worth trading live without your own verification on your specific market and conditions.
Questions, answered
Why did adding the 200 EMA to my strategy make results worse?
Because the 200 EMA filtered out some of the trades that drove your baseline edge, not just the ones you wanted to avoid. Every confirmation condition you add reduces trade count — and does it indiscriminately. The filtered-out trades may have looked imperfect on a chart but contributed to positive expectancy across many occurrences. Fewer trades with 'confirmation' is not automatically better trades.
Which indicator combinations actually beat buy-and-hold?
The ones that hold up in our data are asset-class specific. Fisher Transform for forex. Fibonacci Pivots and Projection Bands for stocks. MA Envelope and Fibonacci Pivots for crypto. Keltner Mean-Reversion for commodities. There is no universal combination in our data that consistently beats buy-and-hold across all markets. You can check what worked on your specific asset on the <a href="/assets">asset results page</a>.
Does a 70%+ win rate mean a strategy is profitable?
Not necessarily. Several strategies in our data post median win rates above 70% while beating buy-and-hold on only 8–11% of tested assets. Win rate measures how often you close a trade positive — it says nothing about whether the size of those wins outpaces the losses and missed asset appreciation. A strategy that wins most small reversion trades but misses large trending moves can easily lose to a passive hold over time.
Are these real trade results or simulated?
Simulated. All results come from backtests on historical price data with realistic transaction costs applied. No real money was traded to produce these figures. Backtested results — even out-of-sample — do not guarantee the same performance in live markets, where execution, slippage, and changing market structure all affect outcomes.
Every figure here comes from our own out-of-sample backtests, costs included — not a course or a guess. Educational information only — not investment advice. Hypothetical backtested results; past performance does not guarantee future results. Trading involves risk of loss.
Keep reading
Get the weekly edge report
The best-performing indicator per asset, what changed this week, and the honest caveats — straight to your inbox.