Of 15 popular trading beliefs we tested with real Micro Nasdaq data, 12 didn't hold up and the rest only half did. None of them turned out to be a reliable edge after costs. Each one links to the full study, with how it was measured and its margin of error.
Same method every time: a clear question, years of 1-minute data, a control (a random entry or an arbitrary level) and the result after commission and slippage. If a belief were true, it would beat the control. Here's what happened.
1. “Price hunts the liquidity above highs and below lows”
False. Asian, London, overnight and previous day highs and lows get reached exactly as often as a level with no name at the same distance (0.0 ± 1.5 percentage points). Fading the overnight sweep: +$4.09 ± $7.88 per trade, no proven edge. See the data →
2. “Price always comes back to VWAP”
False. It came back 3.0 percentage points less often than to the mirror level. Fading back to VWAP lost −$5.09 per trade in 7 of 8 years. It only seems to come back because the VWAP moves toward the price. See the data →
3. “Round numbers are support and resistance”
False. At a round 100 like 18,000, price bounced 10 points before breaking 10 points 51.5% of the time; at an arbitrary level like 18,037, 51.0%. A coin flip both times. See the data →
4. “61.8% is the golden Fibonacci retracement”
False. From 61.8%, the first-hour move resumed 47.9% of the time, right on the same curve as 55% and 70%. Fibonacci effect: −0.5 ± 1.7 percentage points. See the data →
5. “A good risk/reward ratio gives you an edge”
False. With random entries, 1:1 won half the time, 1:2 a third and 1:3 a quarter. Every combination lost between −$1.47 and −$2.52 per trade after costs. The edge has to come from the entry. See the data →
6. “RSI over 70 means sell, under 30 means buy”
False. On 5-minute bars, price tended to keep going after the signal. Trading both signals lost −$7.25 per trade after costs (±$4.00), in 7 of 8 years. See the data →
7. “The opening gap always fills”
Half true. Small gaps filled 93% of the time; gaps bigger than a normal day, 19%. Fading them averaged −$2.66 a day after costs: no edge. See the data →
8. “The first hour tells you where the day is going”
Half true. The size, yes: a busy first hour means a busy day (1.23×). The direction, barely: following it made $12 per contract per day with a margin of error of ±$14. See the data →
9. “Quad witching days go crazy”
False. Quarterly expiration Fridays moved 1.09× a normal day, versus 1.04× on any Friday. The famous last hour doesn't show up. See the data →
10. “Order flow bubbles tell you when to enter”
False. With 308 million Nasdaq ticks, aggression bubbles and effort boxes won as often as, or less than, the same entry in a random direction. Useful to read the market, not to pull the trigger. See the data →
11. “The opening range breakout is a reliable edge”
Half true. Two versions made over $22,000 in seven years with one contract, but one made five times more in the years it was designed on, and both blew up almost half of the 50K accounts. See the data →
12. “Popular TradingView strategies can pass a prop firm evaluation”
False. The best of the 13 most popular passed a Lucid 50K 11% of the time and blew it 36%. Combining the top four: 3% passed, 63% blown. See the data →
13. “Smaller timeframes, more opportunities”
False. On 1-minute bars, costs eat 9.8% of the average bar; on 1-hour bars, 1.3%. In our data the same strategies did worse the smaller the bar. See the data →
14. “A 90% win rate backtest means a great strategy”
False. Our own search found one with 90% wins, a 5.3 profit factor and 0% drawdown. It was a bug in deciding who won when the stop and the target fell inside the same bar. See the data →
15. “Optimize until the backtest looks good”
False. Test 100 strategies with no edge at all and the best one will look good 9 times out of 10. An optimization tests hundreds without you noticing. See the data →
What all of them have in common
- Every line looks like it works on a chart. The eye remembers the bounces and forgets the breaks; only counting tells them apart.
- Costs turn a coin flip into a loss. A rule that wins half the time at 1:1 loses money once you pay commission and slippage.
- If it's free and obvious, the market has already taken it.
Educational content based on our own experience. Not investment advice. Firm rules change: check them on each firm's website. This article contains no affiliate links.