One paid. One died. The ATR didn't pick either.
Short · Risk & sizing · September 29, 2026
Olympia publishes educational content for learning purposes only. Past performance is not indicative of future results. Nothing here is investment advice.
Two flags — one paid, one died. The ATR number didn't call either one. What it tells you is how rough the ride gets and how much room the trade needs: small breath, tighter room; big breath, more room. Size to the breath, not to the dream.
The lesson in writing
Here we have two charts that our models flagged. They don't always win — sometimes they do, sometimes they don't. One of them took its time and ground out the win. The other never reached it.
Now the breath of each. When they were flagged, one was moving about seven percent a day. The other was moving about the same. Nearly the same breath, opposite endings.
So here's the lesson about that number. It's the ATR — the average true range. It isn't telling you which one wins. It tells you something else.
What it is
What it tells you is how rough the ride is going to get. How the fight is going to look, and how bloodied each opponent will be at the end of that fight. And about how much room that trade needs to move and succeed.
A small breath, small swings, tighter room. A big breath, big swings — and a stop parked inside that breath, you are a target waiting to be shot out of the sky.
That is the whole of it. The breath is a measurement of the ride, not a verdict on the ending. Two names can breathe at the same rate and finish in opposite places, and historically they often have — these two did. What the number does tell you, every single time, is the size of the swings on the way, and how much room your plan needs to survive them.
How it's worked out
Every stock has a normal day. The average true range is a measurement of how big that normal day is.
Start with one day: the high minus the low — how far the price travelled between its best and its worst.
“True” means the gap counts too. If it opened well above or below the day before's close, that jump is part of the day's move, so the range is measured from that earlier close whenever that stretch is the wider one. A stock that gapped and then sat still did not have a quiet day, and the true range says so.
Do that for each of the last fourteen days and average them. That average is the ATR, and it comes out in dollars.
Divide it by the price and you get a percent. About seven percent means that name travels about seven percent of its own price on an ordinary day. The percent is what makes two names comparable — a dollar of daily travel means one thing on a ten-dollar stock and something else entirely on a three-hundred-dollar one.
Where Olympia calculates it
You don't have to work any of it out. Olympia measures the breath of every name it scans, and then uses it in three places.
On the scanner it is a column of its own — ATR % (daily): the average true range over fourteen daily bars, as a percent of the close. That is the arithmetic above, run on every name on the board, so you can sort the calm ones from the wild ones before you do anything else.
On a signal it draws the entry zone. The zone is one ATR either side of the centre — the room the plan starts in — and it is measured on the same clock the signal itself runs on, which is not always the daily column.
And when you mark a signal as one you're in, it lays out the stop levels below your fill. Each one is a multiple of that name's own ATR — one ATR down, two, and on up — shown as a price and as the distance below your entry; put in your own size and it carries through to what that stop would put at risk. Your numbers, its arithmetic. That is the moment the breath stops being a metaphor: you can see, in dollars, how much of the stock's ordinary breathing your stop is sitting inside of.
Size to the breath, not to the dream. The breath decides the room.
The pair: Olympia's Death of a Trade card, 26 September 2026 — every signal tracked to its finish, wins and losses both. The ATR figures: average true range over fourteen daily bars, split-adjusted, measured on each name's own flag date. Historical, not predictive.
