Fisher Transform Scanner
- Indicators
-
Evgenii Sutulov
Premium components
1. https://www.mql5.com/en/market/product/201632 - Rotation Manager Console - Version: 1.3
- Activations: 20
Sharp turns, found across your whole watchlist before you have finished reading the first chart. Fisher Transform Scanner runs Ehlers' transform over up to twenty eight symbols and four timeframes, grades each turn by how far into the extreme it began, and keeps the oscillator under the board so a graded cell can be checked by eye.
The price increases by $10 every 10 purchases.
Fisher Transform Scanner is two products in one file. The first is the Fisher Transform itself, in this chart's own sub window: the line, its trigger, and the turn histogram.
The second is a board that runs the same calculation across a list of symbols on several timeframes at once. The board says which instruments have signalled and how recently; the chart under it shows you what produced the signal, so you do not have to attach a second indicator to act on what the table found.
The board
A row per symbol, a column per scanned timeframe, up to four columns and up to twenty eight symbols. It ships scanning fifteen symbols and is raised to twenty eight from one setting. The symbol list is comma separated, or the board can take the Market Watch list instead. Broker prefix and suffix are detected from the chart you are on, or given by hand. Out of the box the board scans four timeframes: M15, H1, H4 and D1.
A cell reads like ^14b G2: which side the signal was on, how many bars ago it fired, and its grade. The word flat means the calculation ran and found no direction. Three dots mean the cell has not been measured yet, and hovering it explains why in words once tooltips are switched on, which they are not by default. A signal older than the freshness threshold keeps its colour but is drawn quieter, so hue says direction and weight says how much it still matters. A row carrying a fresh signal anywhere in it gets a wash of the accent colour.
The AGREE column is the reason to use a board rather than eight charts. For each symbol it takes the visible timeframe columns, finds the majority side, and prints something like v3/4: the direction, and how many of the visible timeframes stand on it. When every visible timeframe agrees, the cell is drawn bold in that side's colour. A tie prints a dash and is deliberately not a verdict. With a single timeframe visible the column is removed entirely rather than printing 1/1, which would only restate the cell beside it.
Column headers are buttons. Click a timeframe header to sort by that column, freshest first, rows with nothing in it at the bottom; click it again to reverse. Click AGREE to sort by agreement, full agreement first. The active header carries the accent colour and a caret, so the sort state lives on the control itself. A separate Sort button cycles freshest, grade and name for a whole-board order.
A Show button cycles all, buys, sells and fresh. Buys and sells follow the AGREE verdict and not any single cell, so a row shown under buys always reads an up glyph in AGREE, and a tie appears under neither. This filters the view and not the calculation: a hidden row is still measured and can still alert.
The legend above the table decodes the notation in one line, including the grade half, which it prints only because this port grades. It can be switched off. Timeframe buttons show and hide a column without stopping the calculation behind it.
The visible row count is fitted to the chart pane and the row setting is a maximum. No page buttons are drawn when everything fits; when it does not, previous and next appear with the page number. Because the board never draws more rows than the pane holds, the page buttons are always on screen and always reachable, however tall the row setting is.
Two status lines are always on show: symbols ready of total, how long since the last update, signals in the lookback window; then either the watchdog line or a sentence explaining why nothing is showing.
Diagnostics
A second panel, off by default, answers the question "why is there no signal": symbol coverage, which series are short of bars, how many signals were found in the window, and how many symbols this indicator switched on in Market Watch to do its work. A cell that could not be measured is explained rather than left blank.
The chart indicator
In its own sub window: the Fisher line and its trigger, and the turn histogram between them, coloured by whether the turn is expanding or fading. Divergence lines are drawn on the Fisher line and, if you want them, between the matching pivots on the price chart as well.
The settings sheet
Values covers the timeframe, the window price is normalised over, nine by default, the extreme level of 1.5 which grades a turn two instead of one, and the three divergence rows. Styles covers the palette, the line and its trigger, the line width, and the turn histogram. Every Values row has steppers, a range and a sentence of help.
Alerts covers the master switch, the per-event switches, the four channels, the cooldown and a test button.
Three colour schemes, including a colour-blind safe one that uses blue and orange rather than red and green. Panel position, scheme, the sort, the filter, which columns are visible and every value changed on the sheet persist per chart across a timeframe change and a terminal restart. The interface is English.
Alerts
Off out of the box. A master switch sits on the panel itself, next to a Test alert button that fires through every channel currently on.
The events are a fresh BUY and a fresh SELL, declared separately for each scanned timeframe, plus data health, which speaks when a symbol loses or regains its data. There is a minimum grade setting, and because this port grades its cells the setting does real work here.
The channels are a popup window, a sound, a push to the mobile terminal and e-mail. The cooldown is kept per cell and per event, thirty minutes by default. Telegram is not part of this product.
Cost
The work is budgeted: a set number of symbols is refreshed per timer tick, so a long list spreads its load instead of stalling the terminal. The first pass is progressive and the panel reports the percentage measured.
Reading it from an EA
The chart indicator publishes eight buffers, all readable through iCustom and CopyBuffer. Buffer 0 is the Fisher line and buffer 1 is the trigger, which in Ehlers' own definition is the Fisher line itself shifted one bar back. Buffer 2 is the turn histogram, the distance between the line and its trigger, and buffer 3 is the colour index that paints it. Buffer 4, Signal, is +1 on a BUY bar, -1 on a SELL bar and 0 otherwise. Buffer 5 is the grade of that signal, buffer 6 carries the divergence marks, and buffer 7 is the zero line.
This product has no separate Buy and Sell price buffers, because there is no price to put an arrow at in a sub window. The signal is buffer 4 and it is the contract: filled on every pass whatever the Styles tab says, and present in the Data Window. Hiding a plot is done by taking its colour away and never by emptying a buffer, so every index above stays readable with the display switches off.
This map is the scanner's own. The free single chart indicator publishes the same eight indices and adds an Override buffer at index 8, which the scanner has no use for. Read the list above rather than assuming the two are interchangeable.
The board itself is a panel and not a buffer. It is read by eye and by alert.
The single chart indicator behind this one, Fisher Transform MTF, is free and carries the same engine. If you follow one symbol, take that one; this scanner exists for watching many at once.
Provenance
This is a port of the Ehlers Fisher Transform. John Ehlers' mathematics is public and the indicator is built into TradingView, which makes a port by formula admissible. Two deviations are declared: the normalising window is progressive on the first bars where Pine returns nothing and both recursions start from zero, which is what Ehlers' own nz does; and on the very first bar the trigger equals the line, so the histogram there is zero rather than a jump.
The product is called Fisher Transform Scanner because that is the name of the method, not because authorship of the method is claimed here. The original work belongs to its author.
Honest limits
The board reads closed bars and counts a cell's age in closed bars. The chart indicator under it can show a signal on the bar still forming, and such a signal can be gone by the close. What has been published on a closed bar does not change afterwards.
The oscillator is live on the forming bar. A cross that exists mid-bar can be absent at the close, and the panel line and the grade go with it. Closed bars do not change.
The transform exaggerates by design. Extreme readings are what it is built to produce, and their frequency is a property of the transform rather than a measure of how unusual the market is.
The grade describes where the turn began, not how likely it is to hold. A grade two turn started from further out. That is all it says.
The normalising window moves every turn the product will ever mark. A short window turns constantly, a long one rarely, and neither of the two settings is the correct one.
AGREE counts timeframes, it does not estimate probability. Four timeframes agreeing means four correlated readings of one instrument agreed. A tie produces no verdict on purpose and is hidden by both the buys filter and the sells filter.
Scanning costs. Twenty-eight symbols on four timeframes is a hundred and twelve cells and will be felt on a slow machine; the budget setting is what you turn down.
Symbols your broker does not carry, and series with fewer bars than the settings need, cannot be measured, and the board says which and why rather than leaving the cell blank.
No profit figure, win rate or backtest result is claimed here.
