Trading edge vs luck: why a setup must work year after year

Trading edge vs luck: why one green month proves little, the 'green in at least half its years' test, and what to do with setups that aren't proven yet.

2026-11-02

Your new setup had a great month. Fifteen trades, mostly green, and you're thinking about sizing up. Before you do, ask the harder question: is this a trading edge, or was it luck in a market that happened to suit it?

The honest test is time. A real edge keeps earning across different kinds of markets: quiet and wild, trending and choppy. A lucky setup earns in one stretch and gives it back in the next. One green month proves very little. Green across several years proves a lot more.

Why one month proves so little

Markets change mood. Some months trend for days on end. Some chop back and forth inside a range. Some explode around Fed meetings and inflation reports, then go quiet for weeks. Traders call these moods regimes, and they can last weeks or months.

Each setup has regimes it loves. A breakout setup tends to shine when markets trend. A fade at the edges of a range tends to shine when they don't. Either one can look brilliant in the month that suits it and ordinary, or worse, in the month that doesn't.

So a green month can mean two things: your setup has an edge, or the month happened to favor it. With 15 trades you can't tell which. Add the small-sample problem (a pure coin-flip setup goes 7 for 10 or better about one time in six) and a single month is mostly noise. More on that in what is a trading edge.

A green month tells you the setup fit that month. Green year after year tells you it fits you.

Same total, different stories

Here are two hypothetical setups, both up $1,200 over a year:

Setup A is one great month carrying a slightly losing year. Take that month away and it's red. Setup B earns a little, most of the time. B is far more likely to keep working, yet A is the one most traders size up on, because its big month is the one they remember.

The same logic works over years. A setup that made all its money in one year and lost a bit in the others has the same shape as Setup A, just stretched out. That's why the test below uses years.

Why the hot setup feels like an edge

Your memory isn't a fair judge. The trades you remember are the recent ones and the big ones, so a setup that just had a great month feels stronger than one that has quietly earned for two years. That's why the numbers have to come from your journal, not from how the setup feels.

There's a second trap: you naturally trade a setup more when it's hot. More trades in the good month, fewer in the bad one, and your sample ends up tilted toward the regime it loves. Looking at each year separately, and at per-trade results rather than totals, softens that tilt.

The 'green in at least half its years' test

A simple rule for calling a setup an edge over your career:

• 30+ trades, so the per-trade number isn't just noise.
• Earning per trade overall, after costs.
• Green in at least half of the years you've traded it.

Years work well because a calendar year usually holds several market moods. A setup that's green in most of them is surviving different conditions, not riding one. It's a strict bar, and it should be: it stops you from calling a hot quarter a career edge.

One honest limit: if you've only traded for one year, the test can only tell you so much. A setup that's green in its only year passes, but it hasn't been through a second year yet. Treat a one-year edge as provisional, and size it like one.

A sample: Breakout vs Pullback

A sample trader in our demo has 28 months of trading, spread across three calendar years:

Breakout passes clearly. A big sample, a solid per-trade result, and green in two of its three years. One red year is fine. Every real edge has bad stretches, and expecting a setup to be green every single year is how traders talk themselves out of good ones.

Pullback passes, but only just. It earns per trade and it's green in exactly half its years, so it meets the rule. But one good year is carrying one losing year, and +$7.95 per trade leaves little room for higher fees, extra slippage or a change in the market. It deserves its normal size, not more, and a close look at what was different in the red year.

Passing the test by the smallest margin is a reason to watch, not a reason to size up.

What to do with 'Not proven yet'

Most setups in most journals land here: fewer than 30 trades, or earning overall but red more years than green. That's normal, especially in your first year or two. What you do next matters more than the label.

• Keep it small. Minimum size until it earns the label. A single micro contract is enough to collect data.
• Write the rules down. A setup with fuzzy rules can't be measured, because you end up counting different trades as the same setup. Write it as a Playbook play.
• Note the conditions. Trending or ranging day, news day, time of day. If it only works in one of them, that's worth knowing.
• Don't keep tweaking. Changing the rules every 10 trades resets the count. Pick a version and give it 30 trades.
• Set a review date. Look again after 30 trades or at the next quarter, whichever comes later.

And if a setup is losing per trade over 30 or more trades, it isn't 'Not proven yet' anymore. It's a leak. Stop it or rewrite it. Trading expectancy explained shows how to spot which of your setups are leaking and why stopping one is often the best improvement available.

Other ways to tell edge from luck

• Split it in halves. Is the setup green in both the first half of its trades and the second?
• Remove the best trade. If one huge winner is the whole profit, the result is luck-shaped.
• Check by market condition. Tag trending and ranging days and compare the setup on each.
• Check by instrument. An edge on MNQ isn't automatically an edge on MES, and the reverse.
• Look at the red year. What was different: the market, your size, your rules? A red year with a clear reason is less worrying than one you can't explain.

See it for your own setups

In Market Matrix, My Trading Journey runs this test for you. Your edge, in My edge & milestones, lists every setup from your tags and Playbook plays across your whole career, with each year's result next to it. It labels a setup Edge only after 30+ trades, earning per trade and green in at least half the years traded. Losing over 30+ trades is a Leak. Everything else is Not proven yet.

Test your setups in My Trading Journey →

It's a Pro feature, and every new member gets 14 days of everything free, no credit card. The point isn't the label. It's to size up only on setups that have survived more than one kind of market, and to stay patient with the rest.

Start your free journal →

This article is for education only and isn't financial advice. Trading futures involves substantial risk of loss.