Trading myths vs. actual data

Your trading stats are probably wrong.

We test popular trading ideas on 500+ US stocks and explain what the data says, in plain English, with a bit of humour. Everything gets compared to the most boring strategy there is: just holding the S&P 500.

Warning label

This is not an optimistic website.

Most money content runs on optimism: the strategy that works, the stock that will fly, the course that will change your life. This site is where that optimism comes to die. Think of it as a graveyard for good-sounding ideas, with neat headstones and the exact cause of death on each one.

It's less fun than a trading guru. It's also much cheaper. Every idea that dies here is one you won't have to test with your own money, and in the long run the savings in nerves are even bigger than the savings in cash.

Possible side effects: fewer exciting trades, more sleep, and an irritating habit of asking "compared to what?" at family barbecues.

New here? Here's the deal.

Every day, somebody on the internet promises a trading strategy that "just works". Some of them are selling a course. Some of them are your relatives.

We take those promises, turn them into exact rules, and run them on years of real market data from hundreds of US stocks. Then we tell you what actually happened. No course, no secret indicator, no Lambo in the thumbnail.

What you'll find

Myths, tested properly

Golden crosses, buying the dip, AI that reads the news. Every test ends with a one-sentence verdict at the top, so you can skip to the end like everyone else does.

No maths degree

Statistics for noobs

If you can follow a story about a man at a barbecue, you can follow this. Each article has one box that explains one stats idea. Zero formulas. We checked.

Lost? Click.

A glossary for when the jargon hits

See a word with a dotted line under it, like survivorship bias? Click it. Every term gets a plain-English explanation and a silly example. Browse the glossary.

The cast

Meet the family

Uncle Gary has had a sure thing since 2008. Aunt Linda has a mood range you could set a stop by. Cousin Kevin asks a chatbot before every trade. None of them are financial advisers. That's the point.

Our mission

Most trading content sells hope. We'd rather tell you an idea doesn't work before the market tells you, at a much higher price.

So far, almost every idea we've tested has failed, including dozens of our own. If that sounds depressing, think of it as money you didn't lose.

On Friday, Uncle Gary told me at the barbecue that when the 50-day crosses the 200-day, you buy. He said it with the confidence of a man who has been waiting since 2008 to break even.

So we checked. 72 versions of the moving-average cross. 577 US stocks. Sixteen years of data.

How many beat a random stock bought on the same day, once you account for luck? None.

Gary is still grilling. The market is still doing its own thing. Read the full test.

Latest

Myth #8 · Stop-loss

Does a Stop-Loss Protect Your Money? We Tested Tight and Wide Stops

1.5 million simulated entries, 577 US stocks. Stops changed how the losses looked, not the average. A random stock with the same stop did just as well.

Read the test →
Myth #7 · Chart levels

Do Support Levels and Fibonacci Work? We Drew Random Lines Too

Seven families of chart levels, about 580 US stocks, 16 years. Buying at support did no better than buying at a line shifted to a random nearby price.

Read the test →
Myth #6 · Track records

Do Trading Strategies Work? Here Are All 430 Versions We Tested

Every course shows the strategy that worked. We show our whole drawer: nine studies, 430 versions, zero survivors after correcting for how many we tried.

Read the test →
Myth #5 · Earnings

Should You Buy Stocks After Good Earnings? Post-Earnings Drift, Tested

12,497 earnings reports, 531 US stocks. The best surprises did no better than a random day. The famous drift lived on the losers, and it was fading anyway.

Read the test →
Myth #4 · Volume

Does Volume Confirm the Trend? Sometimes It Contradicts It

Three classic volume rules, 490 US stocks, 16 years. Loud trading told us nothing the price didn't already say, and the selling climax pointed the wrong way.

Read the test →
Myth #3 · Pullbacks

Does Buying the Dip Work? We Checked Which Dips

30 versions, including a champion trader's method. Strong stocks that never dipped made the same money. The trend did all the work.

Read the test →
Myth #2 · AI

Can AI Predict Stock Prices From the News? We Asked It 599 Times

599 real news stories, three different questions, one very confident AI. The link with the next price move: zero.

Read the test →
Myth #1

Does the Golden Cross Work? We Tested 72 Strategies

72 versions, 1.5 million signals, 577 US stocks. Buying on the cross did no better than buying a random stock on the same day.

Read the test →

Statistics in real life

No trading, just the everyday places where statistics quietly fools all of us. Uncle Gary appears in most of them, usually holding a drink.

Probability

How Long Does It Take to Form a Habit? Why Your Routine Lasts Four Days

Linda's new routine lasted four days. Gary predicted it. He didn't know Linda, he knew a formula.
Memory

Who Does More Housework? Why Couples Add Up to 140%

Gary says he does 70% of the housework. Linda says she does 70%. Kevin says 30%. The dishes are still in the sink.
Corporate · Measurement

Productivity Up 40%. Nobody Knows What We Produce.

Kevin's team closed 40% more tickets per person. It solved 20% fewer problems. There was cake.

All real-life stories (bad-luck streaks, missing socks, hangover cures, wedding bets and more) and office stories in Corporate life (KPIs, dashboards, meetings about meetings)

Lost in your stats class?

You're not bad at statistics. You were probably taught it formulas first and ideas never. Statistics From Zero is a free twenty-two-lesson course that flips that: the idea first, a real-life example, a small exercise, and a translation table for the words your lecturer uses.

Start with Lesson 1: What Is Statistics Actually For?

Coming soon

Stats for noobs

536,000 Monkeys With Darts

Why, if you test enough strategies, some of them will always look brilliant by pure luck.

How we test

Follow along