Split the window in half

Cut the stretch in two and count each half on its own — the cheapest way to catch a made-up result.

Why you need it

Any result counted over one long stretch is a blend of different market moods. Inside there can be one lucky month and six months of nothing, and what comes out is a smooth, pleasant number.

That result lies twice over. It doesn't just overstate the money — it also makes the rule feel sturdy when it isn't.

How it's done

Split the stretch into two equal halves by time. Count each one by itself, and only then put the results side by side.

The rule passes if both halves come out similar. Not identical — similar in sign and in size. If the first half makes a lot and the second makes nothing, there's nothing left to discuss: most likely the first half just had the right market.

Count the halves before you look at the combined result. Otherwise you start to feel the pull to move the cut until both halves look respectable.

What usually comes out

The most common picture: the whole result was made by one short stretch. It's especially easy to fall into this when you ran the measurement right after a good month — that month ends up both inside the sample and the reason you decided to measure at all.

The second common picture: the rule worked while the market was rising and stopped when it turned. That isn't always a death sentence, but you want to know it up front rather than find out with money.

What this does not tell you

Two halves are not a substitute for testing the rule in a different market mood. If the whole stretch sits inside one rise, both halves will agree with each other and both will be useless.