How the scan works
Every rule behind every level in the letter.
FinVigilante is a nightly list of chart pattern setups in US stocks, found by code and reviewed by a person before it goes out. This page is the whole method. Nothing in the letter is discretionary beyond the final review, so if you understand this page you understand every level we print.
Classical charting, in one paragraph
The patterns we scan for are the classical ones: head and shoulders, double bottoms, rectangles, and breakouts from a base. The idea behind all of them is the same. A pattern is a stretch of time where buyers and sellers fought to a standstill inside a boundary. The boundary matters because it is where the losing side gave up before; when price closes through it, the balance has changed and the people trapped on the wrong side become fuel for the move. The measured move (the height of the pattern added to the breakout) is the classical estimate of how far that fuel carries. Everything else here, the trigger, the stop, the negation level, is about proving the breakout is real before risking money on it, and knowing exactly when it has failed.
Where this comes from
None of these rules are ours. The specific borrowings:
Defined the patterns, the measured-move target, and the idea that a breakout needs a close a few percent past the boundary to count. Their 3% rule is the ancestor of our trigger.
Gave us the long-term moving average as a trend filter and the rule that you only buy breakouts from a base when the stock is in, or entering, an uptrend. Our 200-day (40-week) filter is his 30-week average, one notch slower.
Made bases and pivot points the vocabulary of stock traders. Our base breakout scan is that idea with a mechanical definition.
Measured the classical patterns statistically: how often each one breaks out, how far it goes, how often it fails. His work is the reason we follow every setup until it reaches its target or its negation level, so we can report how often each pattern actually works, not just show the ones that did.
Turned charting into a trading discipline: horizontal boundaries, no sloping lines, the Last Day Rule for the stop, and the patience to let the market prove the breakout rather than anticipate it.
Applies that discipline to global stocks every week with a consistency we admire, and is the closest model for what this letter tries to be for US names, nightly.
Gave us the trend template: price above the 50-, 150- and 200-day averages, the averages in order and rising, well off the 52-week low, near the 52-week high, and strong relative strength. We score every name 0 to 5 on it each night and show the score with each setup. It never adds or removes a name; it tells you which of two identical patterns sits in the stronger stock.
Where we depart: we confirm with half an average true range rather than a fixed 3%, because a fixed percentage treats a utility and a biotech the same. We only print the stop once the breakout exists. And we find the patterns with code, which means every rule is written down, applied the same way to every stock, and listed on this page.
What we scan
About 2,000 US-listed stocks, every night after the close, on daily bars. Four setups:
Inverse head and shoulders
Read on weekly bars. Three lows where the middle one is clearly the lowest, shoulders within 10% of each other, a horizontal neckline at the higher of the two reaction highs. The pattern has to be at least 12% tall relative to the neckline and the right shoulder has to have held. Patterns that follow a decline are bottoms; the same shape inside an uptrend is a continuation, and we label it that way.
Double bottom
Weekly bars. Two lows within a few percent of each other, several weeks apart, a horizontal ceiling at the high between them.
Base breakout
Daily bars. Price spent at least a few months under a prior high and is now pressing on it.
Breakout re-test
Daily bars. A stock cleared a multi-month high on volume in the last few weeks, pulled back to that level without going more than 3% through it, and is holding.
The line, the trigger, the states
Every setup has a line: the neckline, the ceiling, the prior high. Price has to clear every high the pattern made, so the line is horizontal and sits at the highest of them. We do not draw sloping necklines.
The trigger is a daily close at least half an average true range (10-day) past the line. A fixed percentage treats a utility and a biotech the same; half an ATR scales with how the stock actually moves.
Each name is in one of three states. Approaching the line: the last close is at or under the line. Over the line, not confirmed: closed over the line but not yet past the trigger. Confirmed: closed past the trigger. We drop a name from the nightly list once its confirmation is more than five sessions old; after that it is a position, not a setup.
Stops: the Last Day Rule
The stop goes under the low of the last day that traded entirely inside the pattern before the first close over the line, Peter Brandt's Last Day Rule as Kibar applies it. That day does not exist until the breakout happens, so a name that is still approaching shows no stop and no reward-to-risk. The stop appears the night the stock closes over the line and does not move after that.
Targets and negation
The target is the measured move: the height of the pattern added to the line. The shaded band on each chart is the target area, 80 to 100 percent of that move. A breakout that reaches the band without a pullback usually keeps going; one that stalls under it is where you start protecting profit.
Negation is the level that says the pattern failed: the right shoulder low for an inverse head and shoulders, the second low for a double bottom. A close under it ends the setup whether or not it ever triggered.
What gets into the letter
A name appears when it passes all of these, and only then:
- It is approaching, over the line, or confirmed within the last five sessions.
- The last close is within 6% of the line. Further away is a pattern, not a setup.
- The 200-day average is rising, or flat and reclaimed. Nothing under a falling 200-day.
- Reward to risk is at least 2 to 1 where a stop exists.
- The pattern has not already broken out once. A second trip through the same line is a re-test of old resistance, not a fresh breakout, and we leave it out.
- For breakouts, the stop is no more than 10% from the trigger.
Every name that passes goes to paid members with a weekly chart, a daily chart, and the levels. There is no top-ten cut; some nights there are three names, some nights twelve.
The list has memory
Each night's post, for paid members, opens with what changed: names that confirmed, names that reached the target, names that were negated and the level they closed under, and names removed with the reason (trigger went stale, trend turned down, pattern no longer detected). A name has to miss three nights before it is called removed, so a stock hovering near the cutoff does not flicker on and off. The Sunday recap, free to everyone, shows how the week's setups resolved.
Charts
Reversal patterns are shown on weekly bars first, because that is the timeframe the scan reads them on, with the daily behind a toggle for entries and stops. Breakouts and re-tests lead with the daily. The gold line is the 200-day average (40-week on the weekly). Click any chart to enlarge it.
What this is not
Not advice. A setup is a place where the odds and the risk are defined, nothing more. Levels are from the last close and can be gone by the open. Earnings dates are printed under each chart because a report inside the trade changes everything; we flag anything inside ten sessions. Size for the stop, not the target.