Do Support and Resistance Levels Actually Hold? Tested, Not Eyeballed

The Anecdote Problem

Every S/R tutorial shows you the same thing: a chart where price hit a line and bounced perfectly. That's not evidence — it's a screenshot chosen after the fact. You don't see the dozens of times price punched through that same level without pausing. Teaching by cherry-picked example is how entire trading communities develop misplaced confidence in something that may not hold up under scrutiny.

We ran 660,005 out-of-sample backtests across 903 assets on four timeframes — 1-Hour, 4-Hour, Daily, and Weekly — to test this properly. The question is simple: do price levels have measurable, repeatable edge when you commit to them before price gets there, not after?

What Systematic Level-Based Trading Actually Looks Like

The honest way to backtest S/R isn't to draw lines by eye and call it a strategy. It's to use a pivot indicator: a formula that computes specific price levels from prior bars and generates signals when price reaches those levels. Fibonacci Pivots and Camarilla Pivots both work this way. Two people running the same indicator on the same data get the same levels. That repeatability is what makes a strategy testable.

Eyeballed S/R can't be backtested honestly — the levels change depending on who's drawing them and when. If you can't define the rule before the trade, you can't measure the rule's real-world performance.

Where Pivot Indicators Actually Ranked

In Stocks, Fibonacci Pivots finished as the top-ranked indicator across 22 assets. Camarilla Pivots ranked first across 16 stocks. In Crypto, Fibonacci Pivots appeared in the top five across 4 assets and Camarilla Pivots across 3. These two indicators showed up consistently across two very different asset classes, which is a meaningful pattern from a large sample.

The result isn't 'S/R always works.' It's more precise than that: mathematically defined, pre-set levels carry measurable signal in specific contexts. Stocks and crypto showed that signal more clearly than other asset classes in our data.

The High Win-Rate Trap

Now here's where the data gets uncomfortable for a large part of retail TA. SMC: Liquidity Sweep — a popular Smart Money Concepts approach built entirely around S/R-like liquidity zones — carried a 71.2% median win rate across our sample. That sounds like a winning strategy. But only 8% of assets where it was tested actually beat buy-and-hold. Holy Grail Confluence showed a 73.3% median win rate, with the same 8% beat rate. Murrey Math Lines hit 74.3% median wins and 11% beat rate.

A high win rate combined with failing to beat a passive benchmark is a specific failure mode: the strategy captures small, frequent winners and surrenders larger losses, often with unfavorable reward-to-risk. You feel right more often than you're wrong, but the math doesn't add up. No SMC-based indicator beat buy-and-hold across any asset in our test set. That's not a knock on the community — it's data.

What to Take From This (Backtests Only, Not Advice)

The gap in our data isn't between 'level trading works' and 'level trading doesn't work.' It's between pre-defined, systematic levels and post-hoc, interpretive levels. Fibonacci and Camarilla Pivots are in the first category. Eyeballed S/R, SMC zones, and liquidity sweeps tend to fall in the second.

Asset class matters too. Pivot indicators topped Stocks and Crypto but didn't dominate in Forex or ETFs, where different indicator families showed edge. Every indicator has a home; checking the data for your specific market is the only honest approach.

All results on this site are hypothetical backtests with realistic trading costs applied. They do not represent real account returns and are not financial or investment advice. Past backtest performance does not guarantee future results. Trade at your own risk.