Which one paid? The ATR can't tell you.

Here's the truth: that dial never picks the winner. There are days it goes in the other direction entirely.

What it tells you, every single time, is the size of the swing on the way — and how much room your plan's gonna need. A stop inside that breath gets clipped by noise. Win or lose.

Here's the ride, not the odds. That is the whole of the number: it measures the trip, it does not vote on the ending. Two names can breathe at nearly the same rate and finish in opposite places — these two did, and historically plenty of pairs have.

In brief: take the day's travel, high to low. Count the overnight gap as well, when the stock opened away from the day before's close. Average the last fourteen of those, divide by the price, and you have a percent — the share of its own price the name travels on an ordinary day.

The same arithmetic, written out step by step, is in One paid. One died. The ATR didn't pick either..

Same scanner — and that is the part of the test that isn't a trick. Both names came off the same board, and the number on each is a column that is there for every name: ATR % (daily), the average true range over fourteen daily bars as a percent of the close. The blind test was two rows of that one column.

From there the same number does two more jobs. It draws the entry zone, one ATR either side of today's centre — the room the plan starts in. And when you mark a signal as one you're in, it lays out the stop levels below your fill as multiples of that name's own ATR. Which is how you stop parking a stop inside the breath by accident.