How Trade The Path works
No opinions, no newsletter, no guru. Just the same repeatable piece of geometry applied to three years of a stock's own price history.
1. We find the swings, not the noise
A day is a swing high only when its high is the highest across the ten trading days either side of it, and a swing low only when its low is the lowest. That filter throws away day-to-day chop and leaves the turning points that actually mattered.
2. We fit two lines, robustly
The slope of each line comes from a Theil–Sen estimator — the median of the slopes between every pair of swing points. Unlike an ordinary line of best fit, one freak spike cannot tilt it. The line is then pushed out to touch the most extreme swing, so it is a true ceiling and floor rather than an average. Everything is fitted in log-price space, so a channel on a $10 stock and a $1,000 stock mean the same thing proportionally.
3. We project both lines forward 3 months
Each line is fitted on its own: we try every straight line through the swing highs and keep the one that touches the most of them, then do the same for the swing lows. The two are not forced parallel, so a channel that widens or narrows is drawn that way. We try several lookbacks — three years, two, eighteen months, one — and keep the fit the price actually respects, preferring the more recent window when they are close.
The lines keep their slope. We also measure how regularly the highs and lows have repeated, and use that rhythm to estimate when the next low and the next high are due. The forward projection is drawn at half the fitted slope: measured across three years, that is the share of a channel's drift that actually carried into the next six months, so the projected path is the likely one rather than the flattering one.
4. We put a number on it
We do not buy inside the channel. Replayed over the stored history, entries anywhere in the lower part of a channel reached 8% within three months about as often as entries anywhere else — the edge is not there. What does work is an overshoot: price that has dropped below its own floor and has already turned back up. Those entries reached 8% roughly four times in five, in about five weeks, with price typically going less than 6% against them on the way.
So the buy level sits just below the projected floor, and we only call it once price has closed above where it was five and ten sessions ago — the difference between buying a bounce and catching a falling knife.
The target is not a fixed number. We replay every comparable entry this stock has had and measure, for 8%, 10%, 12% and up, how often each arrived inside three months, how long it took, and how far price fell against the entry first. The published target is the largest gain that still lands often enough — on this stock's own record blended with the record across every stock we track, since the setup is rare enough that one stock's handful of dips proves little — and that the projected ceiling leaves room for. The sell date is how long this stock has typically taken to get there; the call is judged over the full three months regardless.
If the fit is not clean enough, if the stock has never overshot and recovered, or if no target above 8% is likely enough, we publish no numbers and send no alerts.
Every call is then published, hit or miss, on the track record.
5. We watch it for you
Free: the chart, the lines, the levels, the watchlist. Paid: a Telegram message the moment price actually reaches one of those levels, or breaks out of the channel entirely.