OPEN-SOURCE SCRIPT
Statistical Mapping - True + Midnight Open
⚠️⚠️ IMPORTANT — READ BEFORE USING ⚠️⚠️
This indicator is an educational and analytical tool. It is not financial advice, not a signal service, and not a trading system. It does not tell you when to buy or sell.
Every number it shows is a HISTORICAL FREQUENCY measured on past data. It is not a probability, not a forecast, and not a guarantee. Markets change; a level that was reached on 70% of the last 90 days may be reached far less often over the next 90. Past behaviour never guarantees future behaviour.
Do not size positions off these levels alone. Do not treat a "Reach" percentage as an edge. Use this tool to understand context — how far this market normally travels — and combine it with your own analysis, your own risk management, and your own testing.
You are solely responsible for your trading decisions and any losses that result from them.
════════════════════════════════════════
WHAT THIS IS, IN ONE PARAGRAPH
Every trading day has a shape. Price opens, usually pokes a little way in the wrong direction, then travels in the direction it is actually going to close. Statistical Mapping measures both of those distances across the last N days and draws them on your chart as five levels around today's open. It tells you, before the day develops, roughly how far this market normally pulls back and roughly how far it normally runs.
════════════════════════════════════════
THE FIVE LEVELS
Working from the top of the screen down:
+D Upside distribution objective
-M Upside manipulation area
O The anchor open
+M Downside manipulation area
-D Downside distribution objective
"Distribution" (D) is travel in the direction the period closed. "Manipulation" (M) is travel against it, before the period resolved.
So on a day that closes UP:
- the run from the open up to the high is DISTRIBUTION
- the dip from the open down to the low is MANIPULATION
And on a day that closes DOWN, the mirror image.
That is why the levels are not symmetrical, and why -M sits above the open while +M sits below it. +D and -D are objectives. +M and -M are the areas where a move typically fakes out before doing what it was going to do anyway.
════════════════════════════════════════
CALCULATION METHODS:
BOTH MEDIAN & MEAN
MEAN ONLY
MEDIAN ONLY
---------------
TIMEFRAME EXAMPLES:
1 WEEK
4 HOUR
1 HOUR
---------------
OPTION TO CHOOSE THE NEW YORK MIDNIGHT OPEN AS THE DAILY OPENING PRICE
════════════════════════════════════════
HOW IT IS CALCULATED
For each of the last N completed periods (default 90 days), the script records:
Direction = up if close > open, down if close < open
If the period closed UP:
Distribution sample = high - open
Manipulation sample = open - low
If the period closed DOWN:
Distribution sample = open - low
Manipulation sample = high - open
It then takes the mean or the median of each set of samples and projects those two distances from the CURRENT period's open:
+D = today's open + distribution
-M = today's open + manipulation
+M = today's open - manipulation
-D = today's open - distribution
Mean is the arithmetic average — it is pulled around by outlier days such as CPI, FOMC or gap opens. Median is the middle value — it ignores those outliers and is usually the tighter, more realistic number.
"Both" mode draws a shaded zone spanning from the mean to the median instead of a single line. The WIDTH of that zone is itself information: a wide zone means the sample is skewed by a handful of violent days; a narrow zone means the market has been behaving consistently.
════════════════════════════════════════
THE STATISTICS TABLE — AND WHY IT MATTERS
For each level the table reports:
Level the level name
Price where the level currently sits
Dist how far that is from the anchor open, in price
Reach see below
Hit whether the current period has already traded through the level
In **Both** mode the chart draws a ZONE from the mean to the median, and the table reports the **near edge** of that zone — whichever of the two sits closer to the anchor open. That is the first price of the zone price actually reaches, so it is the number that matters in practice. Price, Dist, Reach and Hit all use that same near edge, so every column describes the same price, and the far edge stays visible on the chart as the other side of the band. Alerts use it too.
Hover any column header for a full explanation of that column. A compact footer row shows the mapping period, anchor mode, method, how many periods were actually usable, and how the sample splits between up-closing and down-closing periods.
IF YOU RUN BOTH MAPPING PERIODS AT ONCE
The table describes ONE mapping period at a time. Two periods have entirely different levels, distances and statistics, and interleaving them would produce a table nobody could read.
By default that is MAPPING PERIOD 1 — the first of the two timeframe slots. You can point it at period 2 instead with "Show Statistics For" in the statistics table settings.
If the period you pick is not on screen — its levels are hidden because your chart timeframe is not lower than it — the table falls back to the other one rather than showing you nothing.
The footer's first cell always names the period the table is describing, right next to the anchor mode and the calculation method, so you can confirm which one you are reading at a glance. The chart draws both sets of levels regardless; only the table is limited to one.
WHAT "REACH" MEANS, PLAINLY
Reach answers one question: out of the days in your lookback, how many of them actually got this far?
Worked example. Lookback is 90 days. +D sits 419 points above today's open, and Reach shows 25.6%.
That means: on 23 of the last 90 days, price traded 419 points or more above THAT day's open at some point during the day. On the other 67 days, it never got that far.
Nothing more than that. It is a count of past days, expressed as a percentage.
Why it is useful: a level on its own is just a line. Reach tells you whether that line marks something ordinary or something rare.
Low Reach (say 15-25%) — price rarely gets here. An extended target. If price
is already here, the day has done unusual work.
High Reach (say 70-80%) — price gets here on most days. Routine. Reaching it
tells you very little on its own.
The footer shows how many periods were actually usable and the up/down split, so you can always see the sample the numbers rest on.
Reach is a count of what happened on past days. It is not a probability of it happening today, and it is not a forecast.
════════════════════════════════════════
SAMPLE SELECTION — WHICH PAST PERIODS GET MEASURED
This is the setting that decides what the statistics actually describe, and it matters more than any other.
MATCHED (the default)
Only periods from the SAME SLOT are measured.
On an intraday mapping period that means the same time of day. With a 1H mapping at 09:15, the numbers come from the 09:00-10:00 hour of each of the previous days. At 10:00 the indicator switches to the 10:00-11:00 hour of those same days. On a 1D mapping period it means the same weekday — a Thursday is measured against previous Thursdays.
ROLLING
The last N periods in a row, whatever time of day they happened to be.
WHY THIS EXISTS
Markets do not behave the same at every hour. The New York open and the middle of the Asian session are different animals. Average them together and you get a number that describes neither.
The practical consequence is specific: on a 1H mapping, a rolling average of the last 90 hours is dominated by quiet hours, because most hours are quiet. Project that at 09:30 and the levels sit far too close in — price blows through them in the first ten minutes and the map looks broken. It is not broken; it was answering the wrong question. Matched sampling asks the right one: how far does THIS hour usually travel?
The same applies on a daily mapping. Mondays and Fridays do not behave like Wednesdays.
WHAT IT DOES TO "LOOKBACK"
In Matched mode, Lookback counts OCCURRENCES of the slot rather than consecutive periods. Lookback 90 on a 1H mapping means the last 90 appearances of that hour — roughly 90 trading days, not 90 hours. That is a much longer reach into history, which has one consequence worth knowing about, below.
WHEN IT DOES NOT APPLY
A 1W mapping period contains one of each slot, so there is nothing to match against. The indicator uses Rolling there and says so on the chart rather than pretending otherwise.
HONEST LIMITATION
Reaching back 90 occurrences of a slot means reaching back 90 days of data. On coarser mapping periods — 30m, 1H, 4H, 1D — that fits comfortably. On finer ones, 15m and below, it needs more intraday history than the script is given, so the sample comes up short.
The indicator does not hide this. The footer shows the real count as, for example, "n 46/90", and a notice explains that fewer samples were available and why. The levels remain valid; they simply rest on a smaller sample, and you can decide whether that is enough. Lowering Lookback removes the notice.
The table footer names the active slot, so you can always see exactly which pool the numbers came from — "1H 09:00" rather than just "1H".
ONE THING TO SET ONCE
That slot label has a time zone setting, in the statistics table section. It defaults to New York, which is the reference most index-futures and FX traders keep their charts on. If your chart is set to anything else, change it to match.
This is not laziness — Pine scripts genuinely cannot read TradingView's chart Time Zone setting. TradingView treats it as a display preference and gives scripts no access to it. So if you have changed your chart away from Exchange time, the label has no way of knowing until you tell it.
The practical case: MNQ trades on CME, whose exchange time zone is Chicago. A chart left on Exchange time therefore runs an hour behind New York. Set this to whatever your chart shows, once, and forget it.
It affects the LABEL ONLY. Slot grouping, every level, every statistic and every Reach figure are completely unaffected — changing time zone shifts every bar by the same amount, so exactly the same periods are grouped together either way.
════════════════════════════════════════
WHICH LEVELS GET REACHED ON WHICH KIND OF DAY
This falls straight out of how the levels are built, and it is worth understanding because it is most of what makes the tool useful.
+M and -M sit close to the open, because a typical day's counter-move is small. They get reached on most days — including slow, quiet, range-bound ones. Look at your own Reach column and they will usually be the two highest numbers in the table. That is exactly why price touching +M or -M, on its own, tells you very little. It is the normal texture of a day, not an event.
+D and -D sit much further out, because they represent a full typical day's directional travel. Price only gets there when the day has already moved further from its open than an average day manages. In practice that means TRENDING DAYS and HIGH-VOLATILITY DAYS — expansion sessions, news days, days that pick a direction in the morning and hold it. On a quiet range day price frequently never comes close to either one.
So the two pairs are answering different questions:
Price at +M / -M -> ordinary. The day is doing what days do.
Price at +D / -D -> this day is not ordinary. It has already
behaved like a trend or expansion day.
That second line is the practical one. Reaching a distribution level is itself information about the character of the session, before you form any view about what happens next.
Two honest caveats. First, this is a description of what the levels mean, not a prediction — nothing here says today will be a trend day. Second, "volatile" and "trending" are not the same thing and the tool does not distinguish them: a violent chop that swings 400 points in one direction and back can reach +D just as a smooth trend can. The level tells you the distance was covered, not how or in what order.
════════════════════════════════════════
HOW THIS DIFFERS FROM ADR / AVERAGE DAILY RANGE
ADR takes the average of (high - low) over N days and usually draws a band above and below either the open or the previous close. It answers one question: how big is a typical day?
Statistical Mapping answers a different and, I would argue, more useful set of questions.
WHERE IT GOES FURTHER THAN ADR
1. It compares like with like. ADR averages the last N days as one undifferentiated pool. On an intraday mapping period this indicator averages only the SAME TIME OF DAY — the 09:00 hour against previous 09:00 hours — and on a daily period only the same weekday. No ADR variant does this, and it is the difference between a projection that survives the New York open and one that price walks through in the first ten minutes.
2. It separates the range into direction. ADR gives you one number for the whole candle. Stat Map splits that candle into the part that travelled with the close and the part that travelled against it, and measures them separately. That is the difference between "the day is usually 300 points" and "the day usually pulls back 90 points before running 210".
3. It is asymmetric, and deliberately so. Because up-days and down-days are measured on their own terms, the upside and downside levels are not mirror images. ADR bands almost always are.
4. It offers the median, not just the mean. A single CPI day can inflate an ADR reading for weeks. The median is immune to that. Being able to flip between the two — and to see the gap between them in "Both" mode — is a diagnostic in its own right.
5. It reports how often each level was actually reached. This is the big one. ADR draws a line and stops. Stat Map tells you the historical frequency behind every line it draws.
6. It supports a NY Midnight anchor. For 24-hour markets the exchange's own daily open is often an arbitrary moment. Many traders work from 00:00 New York instead. The script rebuilds whole days around that time and recomputes every statistic from scratch, rather than just shifting a line.
7. It works on any mapping period, not only daily. Set it to 1W and you get the same decomposition for the weekly candle.
WHERE ADR IS THE BETTER TOOL, OR WHERE THIS ONE IS WEAKER
Being straight about this matters more than selling it.
1. It is more complicated. ADR is one number and anyone can use it in thirty seconds. This has five levels with a specific meaning each, and it will confuse a beginner who has not read the definitions above.
2. It needs a directional close to classify a period. A day that closes exactly at its open contributes to neither sample set. This is rare but it means the sample count can be slightly below your lookback setting.
3. The classification is only known in hindsight. A period is labelled up or down by its CLOSE. That is fine for building statistics from finished days, but it means the levels drawn on today's open are built on a mix of past up-days and past down-days — the script does not and cannot know which kind of day today will be. Both sides are drawn precisely because that is unknowable.
4. It is not adaptive within the period. The levels are fixed at the open and do not adjust as volatility develops during the session. ADR-style tools have the same limitation, but it is worth stating.
5. It says nothing about sequence or timing. It tells you how far, not when, and not in what order. A day that runs to +D at 09:45 and a day that grinds there by 15:55 look identical to this tool.
6. Regime changes take time to show up. With a 90-period lookback, a genuine shift in volatility takes weeks to be fully reflected. Shorten the lookback if you want faster adaptation — and accept a noisier, less stable reading in exchange.
Neither tool replaces the other. ADR sizes the day. This maps it.
════════════════════════════════════════
HOW TO USE IT — PRACTICAL
FOR BEGINNERS, START HERE
Put it on a 15m chart with the defaults, set Calculation method to Median, and just watch it for two weeks without trading it. Notice how often price dips to +M early and then turns. Notice how often +D holds as a high for the day. You are building an intuition for how far this market actually moves — which is the single most common thing new traders have no feel for.
There is no single "correct" way to trade this. The levels describe the shape of a period; which part of that shape is useful depends entirely on what you trade. The sections below cover the common approaches, and the range section is as important as the trend one.
INTRADAY / DAY TRADING
The manipulation levels (+M and -M) are where the tool earns its keep. If you are looking for longs and price has come down into +M, you are at the area where up-days have historically found their low. That is a location to look for your own entry trigger — not a signal by itself. The distribution levels (+D and -D) work the other way: they are where you consider taking profit rather than initiating, because price reaching there means the day has already done a typical day's work in that direction.
The anchor open (O) is a simple bias line. Above it, you are on the bullish side of the period; below it, the bearish side.
RANGE, CONSOLIDATION AND MEAN-REVERSION
This is the other half of the tool, and it is easy to miss if you only read the section above.
Most days are not trend days. On an ordinary session price spends its time between +M and -M, oscillating around the anchor open, and never comes close to +D or -D. That is not the tool failing — it is the tool telling you what kind of day it is.
For anyone trading ranges, consolidations, or short mean-reversion, the useful structure is the inner three levels and nothing else:
-M the upper edge of the ordinary daily range
O the middle, and the level price returns to most often
+M the lower edge of the ordinary daily range
Look at the Reach figures for +M and -M on your instrument. They are typically the two highest numbers in the table — commonly 55-80%. That is the whole point: these are levels price reaches on most days, including quiet ones. Approaches built around them are naturally higher-frequency and lower reward-to-risk than approaches built around +D and -D, which is a trade-off, not a flaw. Fading -M back toward the open, or buying +M back toward the open, is a coherent way to use this.
The anchor open is the natural target for that kind of trade, and often the natural invalidation for the opposite one.
TWO HONEST WARNINGS ABOUT THIS
First, and this matters: Reach measures how often price GOT to a level. It does not measure how often price REVERSED there. Those are completely different questions and this indicator only answers the first. A 75% Reach on -M means price traded there on three days in four — it says nothing whatever about what happened next. Do not read a high Reach as a high win rate.
Second, the trades that make range approaches work are the same trades that get destroyed on trend days. The day you fade -M is the day price runs to +D. That is precisely why the distribution levels are on the chart at the same time: if price is pushing through -M with conviction rather than stalling at it, the map is telling you this may not be a range day. Use the whole structure, not half of it.
SCALPING
Use the Reach column as a filter. If price is sitting just past a level with a 20% reach, the market is already in unusual territory for the session and further continuation in that direction has historically been the exception, not the rule. Conversely a level with 70% reach is barely a level at all — price gets there on most days and it is poor evidence of anything.
Also watch the Hit column. Once +D is ticked for the day, the remaining upside to a typical day's extension is spent.
SWING TRADING
Put 1W in mapping period 1, untick period 2, and drop the chart to 1D or 4H. (Slot
1 rather than slot 2 on purpose: alerts only ever fire from slot 1 — see ALERTS
below.) You now get the same decomposition for the weekly candle: how far a week typically pulls back before running, and how far it typically runs. Weekly +M often lines up with the sort of pullback entry swing traders wait for.
You can also run both at once — 1D and 1W together — on a 1H or 4H chart, to see where the daily and weekly structures agree.
CHOOSING MEAN vs MEDIAN
Median for normal conditions and for tighter, more conservative targets. Mean when you want the levels to account for the fat tail — around known event risk, for example. Both, when you want to see how far apart they are, because that gap is a direct read on how outlier-driven the recent sample has been.
════════════════════════════════════════
IMPORTANT BEHAVIOUR YOU SHOULD KNOW ABOUT
THE TIMEFRAME RULE — please read this one, it is the most common confusion
Your chart timeframe must be STRICTLY LOWER than the mapping period.
Mapping 1D -> chart must be 4H, 1H, 15m, 5m, 1m ...
Mapping 1W -> chart must be 1D, 4H, 1H ...
On a 1D chart the 1D levels will NOT appear. This is correct and intended: a period cannot be projected forward across a chart bar that already contains it. The script tells you so in a message at the bottom of the chart rather than failing silently. If you find that message annoying once you understand the rule, you can switch it off in the settings.
NY MIDNIGHT ANCHOR PRECISION
Days are rebuilt from 1-hour data, inside that data's own context rather than from your chart's bars. 00:00 New York falls on an hourly boundary for the futures, forex and crypto markets this mode is intended for, so the reconstruction is exact.
Because the reconstruction never touches chart bars, the anchor and every level are identical on every chart timeframe. A 5m chart, a 1H chart and a 4H chart all show the same prices.
The anchor mode applies to the 1D mapping only. Any other mapping period always uses that period's own true open.
DAYLIGHT SAVING TIME
Handled automatically, and worth explaining because it is a common source of doubt.
The script uses America/New_York, which is a full timezone rule rather than a fixed UTC offset. The anchor therefore tracks local New York clock time all year — EST in winter, EDT in summer — and its position relative to UTC shifts on its own at each changeover. You never need to adjust anything.
The two changeover days are 23 and 25 hours long. Days are rebuilt by watching the New York calendar date change, not by counting a fixed number of bars, so those two days are measured correctly as well: one simply contains one hour less of data, the other one hour more.
WHY THERE IS NO CUSTOM TIMEZONE OPTION
This is deliberate, for three reasons.
First, a technical one. Days are reconstructed from 1-hour bars, which is exact only because midnight New York lands on an hourly boundary. Several timezones are offset by a half or quarter hour — India, Iran, Nepal, parts of Australia — and there midnight falls in the middle of an hourly bar. The reconstruction would be quietly wrong rather than visibly broken, which is the worst kind of wrong.
Second, a conceptual one. The New York midnight open is a specific reference point that a large amount of flow actually keys off. It is not an arbitrary parameter. A free-form timezone box would imply every choice is equally meaningful, and most are not.
Third, an honest one about method. Offering a dial that changes every number in the table invites tuning it until the levels look good on the chart in front of you. That is curve-fitting, and it makes the statistics worse while feeling like it makes them better.
If you want a different anchor, the True Daily Open mode already gives you the exchange's own reference, which is the other genuinely meaningful one.
SAMPLE SIZE
If your data history cannot supply the number of periods you asked for, the table footer shows what was actually used and a message appears on the chart. The statistics are still valid, they are just built on fewer samples. Be more sceptical of a Reach figure built on 20 periods than one built on 200.
WHAT "TRUE DAILY OPEN" MEANS ON YOUR INSTRUMENT
It is the open of the 1D candle exactly as TradingView builds it for that symbol — so it follows each market's own session definition rather than imposing one:
US stocks 09:30 New York (regular session)
Euronext stocks 09:00 local exchange time
CME index futures 18:00 New York, previous day
Other futures that product's own session start, which differs by complex (grains, energy, metals and softs do not all open at the same time)
Forex and CFDs typically 17:00 New York
Crypto 00:00 UTC
One thing worth knowing: for instruments with a pre/post market, the daily candle follows YOUR CHART'S extended-hours setting. Turn extended hours on for a US stock and the daily open becomes the pre-market open rather than 09:30. That is consistent with what you see on the chart, but it does mean two traders looking at the same stock with different session settings will see different levels. If that matters to you, fix your chart's session setting and leave it alone.
If you trade something unusual and want to be sure, put the indicator on a 1H chart and compare the O line against the open of the daily candle on a 1D chart. They should match to the tick.
THE SAME ON EVERY CHART TIMEFRAME
Everything the script draws is read from the mapping period's own context, never rebuilt from chart bars. Put a 1W mapping on a 1D chart, then a 4H, then a 1H, then a 5m: the anchor, all five levels and every statistic are the same prices every time.
That is a deliberate design decision, not a detail. Rebuilding the anchor from chart bars cannot guarantee it, because bar alignment, session definitions, holidays and gaps all differ between timeframes — and a level that moves when you change timeframe is worse than no level at all.
REPAINTING
No level moves once it is drawn. Every level is fixed the moment its period opens and stays there until the next period begins.
Being precise about how that holds, since the script does use lookahead:
1. Every statistic — the means, the medians, the Reach percentages, the
sample counts — is computed from COMPLETED periods only. The forming
period's high, low and close never enter any of them. This is the part
that would leak the future, and it does not.
2. The five levels are built from those statistics plus the current period's
OPENING price, and drawn between its start and end timestamps. All three
of those are known the instant the period begins, so reading them ahead is
not future information. This is the standard, documented way to anchor a
higher-timeframe open.
3. One further value is read: the current period's running high and low. It
feeds exactly one thing — the Hit column, which reports whether the period
SO FAR has traded through a level — and that column is only ever drawn on
the last bar, where "so far" means right now. No level, no statistic and no
alert depends on it, and nothing about it is plotted historically.
Point 3 is worth stating plainly because it is the kind of thing that deserves scrutiny in an open-source script. It is read from the period's own context rather than rebuilt from chart bars for a concrete reason: on a live chart, a script is not guaranteed to calculate over the full period, so counting back through chart bars can silently measure only a recent slice of it — and do so differently in Bar Replay than in real time.
WHEN THERE IS NOT ENOUGH HISTORY
If the symbol does not have as many completed periods as your Lookback asks for — 90 weekly periods is nearly two years, and plenty of symbols do not have that — the script does not hide anything and does not error out.
It uses every period that does exist, shows the real count in the table footer as for example "n 47/90", and puts a notice at the bottom of the chart telling you the sample is smaller than you requested. The levels remain valid; they simply rest on fewer samples, and you can decide whether that is enough for you. Lowering Lookback to a number the symbol can actually supply removes the notice.
The statistics table never disappears because of missing data. If a value genuinely cannot be computed it reads n/a, so you can always see what the script is and is not able to do.
════════════════════════════════════════
SETTINGS
1 — MAPPING PERIODS
Two independent slots, each with its own on/off toggle. Defaults are 1D on, 1W off.
Daily anchor mode: True Daily Open, or NY Midnight Open (00:00 America/New_York).
Lookback: how many completed periods feed the statistics. Default 90. In Matched sampling this counts occurrences of the current slot.
Sample Selection: Matched or Rolling. Defaults to Matched. See the section above.
Calculation method: Mean, Median, or Both. Defaults to Both.
2 — "BOTH" MODE ZONES
Fill colours for the mean-to-median zones. Sits directly under Calculation method because it only has an effect when that is set to Both.
3 — LEVEL LINES, COLOURS & LABELS
Colour, line style and thickness for each of the five levels, listed in the same top-to-bottom order they appear on the chart. Level labels can be turned off.
4 — STATISTICS TABLE
On/off; which mapping period it describes (defaults to period 1); slot label time zone (defaults to New York; set it to match your chart); position (all eight edge and corner slots, including top and bottom centre; defaults to Middle Right); text size (defaults to Normal).
5 — ALERTS
Which levels can fire, and what counts as reaching one.
6 — WARNINGS
On/off for the on-chart notices described above. The warning panel automatically places itself away from the statistics table, so the two never overlap wherever you put the table.
════════════════════════════════════════
ALERTS
Six alert conditions are available:
+D reached
-M reached
Open crossed
+M reached
-D reached
Any enabled level
ALERTS COME FROM MAPPING PERIOD 1 ONLY. Worth stating plainly, because nothing on
screen will tell you otherwise: if you untick mapping period 1 and run only period
2, these six conditions still appear in TradingView's dropdown and you can still
create the alert — it simply never fires. If you want alerts on a particular
period, put that period in SLOT 1.
Settings section 5 controls two things. First, which levels are allowed to fire — untick a level and it will never trigger, even if you created an alert for it. Second, what counts as reaching a level:
Touches the level (wick) — fires as soon as any part of the bar reaches the
level. Earlier and more sensitive.
Closes beyond the level — fires only when a bar CLOSES past it. Later, and
fewer false triggers.
TO CREATE AN ALERT
1. Set the tickboxes in section 5 the way you want them, then press Ok.
2. Right-click the chart and choose Add alert (or press Alt+A).
3. In the Condition dropdown at the top, select "Stat-Map (Gigi)".
4. In the second dropdown, pick the level you want. Use "Any enabled level" if
you would rather have one alert covering all of them.
5. Set Trigger to "Once Per Bar Close" for confirmed signals, or "Once Per Bar"
for intrabar.
6. Press Create. Repeat for each level you want separately.
The levels jumping to new prices at the start of a period never by itself sends an alert. A genuine touch on that opening bar does, though — an opening bar that runs from the anchor open straight into a level is a real event, and both the Hit column and the alerts treat it as one.
════════════════════════════════════════
Open source under the Mozilla Public License 2.0. You are welcome to read, learn from and build on the code.
Feedback and bug reports are genuinely welcome — if you find a symbol or timeframe where something looks wrong, please say so.
— Gigi_Luigino
This indicator is an educational and analytical tool. It is not financial advice, not a signal service, and not a trading system. It does not tell you when to buy or sell.
Every number it shows is a HISTORICAL FREQUENCY measured on past data. It is not a probability, not a forecast, and not a guarantee. Markets change; a level that was reached on 70% of the last 90 days may be reached far less often over the next 90. Past behaviour never guarantees future behaviour.
Do not size positions off these levels alone. Do not treat a "Reach" percentage as an edge. Use this tool to understand context — how far this market normally travels — and combine it with your own analysis, your own risk management, and your own testing.
You are solely responsible for your trading decisions and any losses that result from them.
════════════════════════════════════════
WHAT THIS IS, IN ONE PARAGRAPH
Every trading day has a shape. Price opens, usually pokes a little way in the wrong direction, then travels in the direction it is actually going to close. Statistical Mapping measures both of those distances across the last N days and draws them on your chart as five levels around today's open. It tells you, before the day develops, roughly how far this market normally pulls back and roughly how far it normally runs.
════════════════════════════════════════
THE FIVE LEVELS
Working from the top of the screen down:
+D Upside distribution objective
-M Upside manipulation area
O The anchor open
+M Downside manipulation area
-D Downside distribution objective
"Distribution" (D) is travel in the direction the period closed. "Manipulation" (M) is travel against it, before the period resolved.
So on a day that closes UP:
- the run from the open up to the high is DISTRIBUTION
- the dip from the open down to the low is MANIPULATION
And on a day that closes DOWN, the mirror image.
That is why the levels are not symmetrical, and why -M sits above the open while +M sits below it. +D and -D are objectives. +M and -M are the areas where a move typically fakes out before doing what it was going to do anyway.
════════════════════════════════════════
CALCULATION METHODS:
BOTH MEDIAN & MEAN
MEAN ONLY
MEDIAN ONLY
---------------
TIMEFRAME EXAMPLES:
1 WEEK
4 HOUR
1 HOUR
---------------
OPTION TO CHOOSE THE NEW YORK MIDNIGHT OPEN AS THE DAILY OPENING PRICE
════════════════════════════════════════
HOW IT IS CALCULATED
For each of the last N completed periods (default 90 days), the script records:
Direction = up if close > open, down if close < open
If the period closed UP:
Distribution sample = high - open
Manipulation sample = open - low
If the period closed DOWN:
Distribution sample = open - low
Manipulation sample = high - open
It then takes the mean or the median of each set of samples and projects those two distances from the CURRENT period's open:
+D = today's open + distribution
-M = today's open + manipulation
+M = today's open - manipulation
-D = today's open - distribution
Mean is the arithmetic average — it is pulled around by outlier days such as CPI, FOMC or gap opens. Median is the middle value — it ignores those outliers and is usually the tighter, more realistic number.
"Both" mode draws a shaded zone spanning from the mean to the median instead of a single line. The WIDTH of that zone is itself information: a wide zone means the sample is skewed by a handful of violent days; a narrow zone means the market has been behaving consistently.
════════════════════════════════════════
THE STATISTICS TABLE — AND WHY IT MATTERS
For each level the table reports:
Level the level name
Price where the level currently sits
Dist how far that is from the anchor open, in price
Reach see below
Hit whether the current period has already traded through the level
In **Both** mode the chart draws a ZONE from the mean to the median, and the table reports the **near edge** of that zone — whichever of the two sits closer to the anchor open. That is the first price of the zone price actually reaches, so it is the number that matters in practice. Price, Dist, Reach and Hit all use that same near edge, so every column describes the same price, and the far edge stays visible on the chart as the other side of the band. Alerts use it too.
Hover any column header for a full explanation of that column. A compact footer row shows the mapping period, anchor mode, method, how many periods were actually usable, and how the sample splits between up-closing and down-closing periods.
IF YOU RUN BOTH MAPPING PERIODS AT ONCE
The table describes ONE mapping period at a time. Two periods have entirely different levels, distances and statistics, and interleaving them would produce a table nobody could read.
By default that is MAPPING PERIOD 1 — the first of the two timeframe slots. You can point it at period 2 instead with "Show Statistics For" in the statistics table settings.
If the period you pick is not on screen — its levels are hidden because your chart timeframe is not lower than it — the table falls back to the other one rather than showing you nothing.
The footer's first cell always names the period the table is describing, right next to the anchor mode and the calculation method, so you can confirm which one you are reading at a glance. The chart draws both sets of levels regardless; only the table is limited to one.
WHAT "REACH" MEANS, PLAINLY
Reach answers one question: out of the days in your lookback, how many of them actually got this far?
Worked example. Lookback is 90 days. +D sits 419 points above today's open, and Reach shows 25.6%.
That means: on 23 of the last 90 days, price traded 419 points or more above THAT day's open at some point during the day. On the other 67 days, it never got that far.
Nothing more than that. It is a count of past days, expressed as a percentage.
Why it is useful: a level on its own is just a line. Reach tells you whether that line marks something ordinary or something rare.
Low Reach (say 15-25%) — price rarely gets here. An extended target. If price
is already here, the day has done unusual work.
High Reach (say 70-80%) — price gets here on most days. Routine. Reaching it
tells you very little on its own.
The footer shows how many periods were actually usable and the up/down split, so you can always see the sample the numbers rest on.
Reach is a count of what happened on past days. It is not a probability of it happening today, and it is not a forecast.
════════════════════════════════════════
SAMPLE SELECTION — WHICH PAST PERIODS GET MEASURED
This is the setting that decides what the statistics actually describe, and it matters more than any other.
MATCHED (the default)
Only periods from the SAME SLOT are measured.
On an intraday mapping period that means the same time of day. With a 1H mapping at 09:15, the numbers come from the 09:00-10:00 hour of each of the previous days. At 10:00 the indicator switches to the 10:00-11:00 hour of those same days. On a 1D mapping period it means the same weekday — a Thursday is measured against previous Thursdays.
ROLLING
The last N periods in a row, whatever time of day they happened to be.
WHY THIS EXISTS
Markets do not behave the same at every hour. The New York open and the middle of the Asian session are different animals. Average them together and you get a number that describes neither.
The practical consequence is specific: on a 1H mapping, a rolling average of the last 90 hours is dominated by quiet hours, because most hours are quiet. Project that at 09:30 and the levels sit far too close in — price blows through them in the first ten minutes and the map looks broken. It is not broken; it was answering the wrong question. Matched sampling asks the right one: how far does THIS hour usually travel?
The same applies on a daily mapping. Mondays and Fridays do not behave like Wednesdays.
WHAT IT DOES TO "LOOKBACK"
In Matched mode, Lookback counts OCCURRENCES of the slot rather than consecutive periods. Lookback 90 on a 1H mapping means the last 90 appearances of that hour — roughly 90 trading days, not 90 hours. That is a much longer reach into history, which has one consequence worth knowing about, below.
WHEN IT DOES NOT APPLY
A 1W mapping period contains one of each slot, so there is nothing to match against. The indicator uses Rolling there and says so on the chart rather than pretending otherwise.
HONEST LIMITATION
Reaching back 90 occurrences of a slot means reaching back 90 days of data. On coarser mapping periods — 30m, 1H, 4H, 1D — that fits comfortably. On finer ones, 15m and below, it needs more intraday history than the script is given, so the sample comes up short.
The indicator does not hide this. The footer shows the real count as, for example, "n 46/90", and a notice explains that fewer samples were available and why. The levels remain valid; they simply rest on a smaller sample, and you can decide whether that is enough. Lowering Lookback removes the notice.
The table footer names the active slot, so you can always see exactly which pool the numbers came from — "1H 09:00" rather than just "1H".
ONE THING TO SET ONCE
That slot label has a time zone setting, in the statistics table section. It defaults to New York, which is the reference most index-futures and FX traders keep their charts on. If your chart is set to anything else, change it to match.
This is not laziness — Pine scripts genuinely cannot read TradingView's chart Time Zone setting. TradingView treats it as a display preference and gives scripts no access to it. So if you have changed your chart away from Exchange time, the label has no way of knowing until you tell it.
The practical case: MNQ trades on CME, whose exchange time zone is Chicago. A chart left on Exchange time therefore runs an hour behind New York. Set this to whatever your chart shows, once, and forget it.
It affects the LABEL ONLY. Slot grouping, every level, every statistic and every Reach figure are completely unaffected — changing time zone shifts every bar by the same amount, so exactly the same periods are grouped together either way.
════════════════════════════════════════
WHICH LEVELS GET REACHED ON WHICH KIND OF DAY
This falls straight out of how the levels are built, and it is worth understanding because it is most of what makes the tool useful.
+M and -M sit close to the open, because a typical day's counter-move is small. They get reached on most days — including slow, quiet, range-bound ones. Look at your own Reach column and they will usually be the two highest numbers in the table. That is exactly why price touching +M or -M, on its own, tells you very little. It is the normal texture of a day, not an event.
+D and -D sit much further out, because they represent a full typical day's directional travel. Price only gets there when the day has already moved further from its open than an average day manages. In practice that means TRENDING DAYS and HIGH-VOLATILITY DAYS — expansion sessions, news days, days that pick a direction in the morning and hold it. On a quiet range day price frequently never comes close to either one.
So the two pairs are answering different questions:
Price at +M / -M -> ordinary. The day is doing what days do.
Price at +D / -D -> this day is not ordinary. It has already
behaved like a trend or expansion day.
That second line is the practical one. Reaching a distribution level is itself information about the character of the session, before you form any view about what happens next.
Two honest caveats. First, this is a description of what the levels mean, not a prediction — nothing here says today will be a trend day. Second, "volatile" and "trending" are not the same thing and the tool does not distinguish them: a violent chop that swings 400 points in one direction and back can reach +D just as a smooth trend can. The level tells you the distance was covered, not how or in what order.
════════════════════════════════════════
HOW THIS DIFFERS FROM ADR / AVERAGE DAILY RANGE
ADR takes the average of (high - low) over N days and usually draws a band above and below either the open or the previous close. It answers one question: how big is a typical day?
Statistical Mapping answers a different and, I would argue, more useful set of questions.
WHERE IT GOES FURTHER THAN ADR
1. It compares like with like. ADR averages the last N days as one undifferentiated pool. On an intraday mapping period this indicator averages only the SAME TIME OF DAY — the 09:00 hour against previous 09:00 hours — and on a daily period only the same weekday. No ADR variant does this, and it is the difference between a projection that survives the New York open and one that price walks through in the first ten minutes.
2. It separates the range into direction. ADR gives you one number for the whole candle. Stat Map splits that candle into the part that travelled with the close and the part that travelled against it, and measures them separately. That is the difference between "the day is usually 300 points" and "the day usually pulls back 90 points before running 210".
3. It is asymmetric, and deliberately so. Because up-days and down-days are measured on their own terms, the upside and downside levels are not mirror images. ADR bands almost always are.
4. It offers the median, not just the mean. A single CPI day can inflate an ADR reading for weeks. The median is immune to that. Being able to flip between the two — and to see the gap between them in "Both" mode — is a diagnostic in its own right.
5. It reports how often each level was actually reached. This is the big one. ADR draws a line and stops. Stat Map tells you the historical frequency behind every line it draws.
6. It supports a NY Midnight anchor. For 24-hour markets the exchange's own daily open is often an arbitrary moment. Many traders work from 00:00 New York instead. The script rebuilds whole days around that time and recomputes every statistic from scratch, rather than just shifting a line.
7. It works on any mapping period, not only daily. Set it to 1W and you get the same decomposition for the weekly candle.
WHERE ADR IS THE BETTER TOOL, OR WHERE THIS ONE IS WEAKER
Being straight about this matters more than selling it.
1. It is more complicated. ADR is one number and anyone can use it in thirty seconds. This has five levels with a specific meaning each, and it will confuse a beginner who has not read the definitions above.
2. It needs a directional close to classify a period. A day that closes exactly at its open contributes to neither sample set. This is rare but it means the sample count can be slightly below your lookback setting.
3. The classification is only known in hindsight. A period is labelled up or down by its CLOSE. That is fine for building statistics from finished days, but it means the levels drawn on today's open are built on a mix of past up-days and past down-days — the script does not and cannot know which kind of day today will be. Both sides are drawn precisely because that is unknowable.
4. It is not adaptive within the period. The levels are fixed at the open and do not adjust as volatility develops during the session. ADR-style tools have the same limitation, but it is worth stating.
5. It says nothing about sequence or timing. It tells you how far, not when, and not in what order. A day that runs to +D at 09:45 and a day that grinds there by 15:55 look identical to this tool.
6. Regime changes take time to show up. With a 90-period lookback, a genuine shift in volatility takes weeks to be fully reflected. Shorten the lookback if you want faster adaptation — and accept a noisier, less stable reading in exchange.
Neither tool replaces the other. ADR sizes the day. This maps it.
════════════════════════════════════════
HOW TO USE IT — PRACTICAL
FOR BEGINNERS, START HERE
Put it on a 15m chart with the defaults, set Calculation method to Median, and just watch it for two weeks without trading it. Notice how often price dips to +M early and then turns. Notice how often +D holds as a high for the day. You are building an intuition for how far this market actually moves — which is the single most common thing new traders have no feel for.
There is no single "correct" way to trade this. The levels describe the shape of a period; which part of that shape is useful depends entirely on what you trade. The sections below cover the common approaches, and the range section is as important as the trend one.
INTRADAY / DAY TRADING
The manipulation levels (+M and -M) are where the tool earns its keep. If you are looking for longs and price has come down into +M, you are at the area where up-days have historically found their low. That is a location to look for your own entry trigger — not a signal by itself. The distribution levels (+D and -D) work the other way: they are where you consider taking profit rather than initiating, because price reaching there means the day has already done a typical day's work in that direction.
The anchor open (O) is a simple bias line. Above it, you are on the bullish side of the period; below it, the bearish side.
RANGE, CONSOLIDATION AND MEAN-REVERSION
This is the other half of the tool, and it is easy to miss if you only read the section above.
Most days are not trend days. On an ordinary session price spends its time between +M and -M, oscillating around the anchor open, and never comes close to +D or -D. That is not the tool failing — it is the tool telling you what kind of day it is.
For anyone trading ranges, consolidations, or short mean-reversion, the useful structure is the inner three levels and nothing else:
-M the upper edge of the ordinary daily range
O the middle, and the level price returns to most often
+M the lower edge of the ordinary daily range
Look at the Reach figures for +M and -M on your instrument. They are typically the two highest numbers in the table — commonly 55-80%. That is the whole point: these are levels price reaches on most days, including quiet ones. Approaches built around them are naturally higher-frequency and lower reward-to-risk than approaches built around +D and -D, which is a trade-off, not a flaw. Fading -M back toward the open, or buying +M back toward the open, is a coherent way to use this.
The anchor open is the natural target for that kind of trade, and often the natural invalidation for the opposite one.
TWO HONEST WARNINGS ABOUT THIS
First, and this matters: Reach measures how often price GOT to a level. It does not measure how often price REVERSED there. Those are completely different questions and this indicator only answers the first. A 75% Reach on -M means price traded there on three days in four — it says nothing whatever about what happened next. Do not read a high Reach as a high win rate.
Second, the trades that make range approaches work are the same trades that get destroyed on trend days. The day you fade -M is the day price runs to +D. That is precisely why the distribution levels are on the chart at the same time: if price is pushing through -M with conviction rather than stalling at it, the map is telling you this may not be a range day. Use the whole structure, not half of it.
SCALPING
Use the Reach column as a filter. If price is sitting just past a level with a 20% reach, the market is already in unusual territory for the session and further continuation in that direction has historically been the exception, not the rule. Conversely a level with 70% reach is barely a level at all — price gets there on most days and it is poor evidence of anything.
Also watch the Hit column. Once +D is ticked for the day, the remaining upside to a typical day's extension is spent.
SWING TRADING
Put 1W in mapping period 1, untick period 2, and drop the chart to 1D or 4H. (Slot
1 rather than slot 2 on purpose: alerts only ever fire from slot 1 — see ALERTS
below.) You now get the same decomposition for the weekly candle: how far a week typically pulls back before running, and how far it typically runs. Weekly +M often lines up with the sort of pullback entry swing traders wait for.
You can also run both at once — 1D and 1W together — on a 1H or 4H chart, to see where the daily and weekly structures agree.
CHOOSING MEAN vs MEDIAN
Median for normal conditions and for tighter, more conservative targets. Mean when you want the levels to account for the fat tail — around known event risk, for example. Both, when you want to see how far apart they are, because that gap is a direct read on how outlier-driven the recent sample has been.
════════════════════════════════════════
IMPORTANT BEHAVIOUR YOU SHOULD KNOW ABOUT
THE TIMEFRAME RULE — please read this one, it is the most common confusion
Your chart timeframe must be STRICTLY LOWER than the mapping period.
Mapping 1D -> chart must be 4H, 1H, 15m, 5m, 1m ...
Mapping 1W -> chart must be 1D, 4H, 1H ...
On a 1D chart the 1D levels will NOT appear. This is correct and intended: a period cannot be projected forward across a chart bar that already contains it. The script tells you so in a message at the bottom of the chart rather than failing silently. If you find that message annoying once you understand the rule, you can switch it off in the settings.
NY MIDNIGHT ANCHOR PRECISION
Days are rebuilt from 1-hour data, inside that data's own context rather than from your chart's bars. 00:00 New York falls on an hourly boundary for the futures, forex and crypto markets this mode is intended for, so the reconstruction is exact.
Because the reconstruction never touches chart bars, the anchor and every level are identical on every chart timeframe. A 5m chart, a 1H chart and a 4H chart all show the same prices.
The anchor mode applies to the 1D mapping only. Any other mapping period always uses that period's own true open.
DAYLIGHT SAVING TIME
Handled automatically, and worth explaining because it is a common source of doubt.
The script uses America/New_York, which is a full timezone rule rather than a fixed UTC offset. The anchor therefore tracks local New York clock time all year — EST in winter, EDT in summer — and its position relative to UTC shifts on its own at each changeover. You never need to adjust anything.
The two changeover days are 23 and 25 hours long. Days are rebuilt by watching the New York calendar date change, not by counting a fixed number of bars, so those two days are measured correctly as well: one simply contains one hour less of data, the other one hour more.
WHY THERE IS NO CUSTOM TIMEZONE OPTION
This is deliberate, for three reasons.
First, a technical one. Days are reconstructed from 1-hour bars, which is exact only because midnight New York lands on an hourly boundary. Several timezones are offset by a half or quarter hour — India, Iran, Nepal, parts of Australia — and there midnight falls in the middle of an hourly bar. The reconstruction would be quietly wrong rather than visibly broken, which is the worst kind of wrong.
Second, a conceptual one. The New York midnight open is a specific reference point that a large amount of flow actually keys off. It is not an arbitrary parameter. A free-form timezone box would imply every choice is equally meaningful, and most are not.
Third, an honest one about method. Offering a dial that changes every number in the table invites tuning it until the levels look good on the chart in front of you. That is curve-fitting, and it makes the statistics worse while feeling like it makes them better.
If you want a different anchor, the True Daily Open mode already gives you the exchange's own reference, which is the other genuinely meaningful one.
SAMPLE SIZE
If your data history cannot supply the number of periods you asked for, the table footer shows what was actually used and a message appears on the chart. The statistics are still valid, they are just built on fewer samples. Be more sceptical of a Reach figure built on 20 periods than one built on 200.
WHAT "TRUE DAILY OPEN" MEANS ON YOUR INSTRUMENT
It is the open of the 1D candle exactly as TradingView builds it for that symbol — so it follows each market's own session definition rather than imposing one:
US stocks 09:30 New York (regular session)
Euronext stocks 09:00 local exchange time
CME index futures 18:00 New York, previous day
Other futures that product's own session start, which differs by complex (grains, energy, metals and softs do not all open at the same time)
Forex and CFDs typically 17:00 New York
Crypto 00:00 UTC
One thing worth knowing: for instruments with a pre/post market, the daily candle follows YOUR CHART'S extended-hours setting. Turn extended hours on for a US stock and the daily open becomes the pre-market open rather than 09:30. That is consistent with what you see on the chart, but it does mean two traders looking at the same stock with different session settings will see different levels. If that matters to you, fix your chart's session setting and leave it alone.
If you trade something unusual and want to be sure, put the indicator on a 1H chart and compare the O line against the open of the daily candle on a 1D chart. They should match to the tick.
THE SAME ON EVERY CHART TIMEFRAME
Everything the script draws is read from the mapping period's own context, never rebuilt from chart bars. Put a 1W mapping on a 1D chart, then a 4H, then a 1H, then a 5m: the anchor, all five levels and every statistic are the same prices every time.
That is a deliberate design decision, not a detail. Rebuilding the anchor from chart bars cannot guarantee it, because bar alignment, session definitions, holidays and gaps all differ between timeframes — and a level that moves when you change timeframe is worse than no level at all.
REPAINTING
No level moves once it is drawn. Every level is fixed the moment its period opens and stays there until the next period begins.
Being precise about how that holds, since the script does use lookahead:
1. Every statistic — the means, the medians, the Reach percentages, the
sample counts — is computed from COMPLETED periods only. The forming
period's high, low and close never enter any of them. This is the part
that would leak the future, and it does not.
2. The five levels are built from those statistics plus the current period's
OPENING price, and drawn between its start and end timestamps. All three
of those are known the instant the period begins, so reading them ahead is
not future information. This is the standard, documented way to anchor a
higher-timeframe open.
3. One further value is read: the current period's running high and low. It
feeds exactly one thing — the Hit column, which reports whether the period
SO FAR has traded through a level — and that column is only ever drawn on
the last bar, where "so far" means right now. No level, no statistic and no
alert depends on it, and nothing about it is plotted historically.
Point 3 is worth stating plainly because it is the kind of thing that deserves scrutiny in an open-source script. It is read from the period's own context rather than rebuilt from chart bars for a concrete reason: on a live chart, a script is not guaranteed to calculate over the full period, so counting back through chart bars can silently measure only a recent slice of it — and do so differently in Bar Replay than in real time.
WHEN THERE IS NOT ENOUGH HISTORY
If the symbol does not have as many completed periods as your Lookback asks for — 90 weekly periods is nearly two years, and plenty of symbols do not have that — the script does not hide anything and does not error out.
It uses every period that does exist, shows the real count in the table footer as for example "n 47/90", and puts a notice at the bottom of the chart telling you the sample is smaller than you requested. The levels remain valid; they simply rest on fewer samples, and you can decide whether that is enough for you. Lowering Lookback to a number the symbol can actually supply removes the notice.
The statistics table never disappears because of missing data. If a value genuinely cannot be computed it reads n/a, so you can always see what the script is and is not able to do.
════════════════════════════════════════
SETTINGS
1 — MAPPING PERIODS
Two independent slots, each with its own on/off toggle. Defaults are 1D on, 1W off.
Daily anchor mode: True Daily Open, or NY Midnight Open (00:00 America/New_York).
Lookback: how many completed periods feed the statistics. Default 90. In Matched sampling this counts occurrences of the current slot.
Sample Selection: Matched or Rolling. Defaults to Matched. See the section above.
Calculation method: Mean, Median, or Both. Defaults to Both.
2 — "BOTH" MODE ZONES
Fill colours for the mean-to-median zones. Sits directly under Calculation method because it only has an effect when that is set to Both.
3 — LEVEL LINES, COLOURS & LABELS
Colour, line style and thickness for each of the five levels, listed in the same top-to-bottom order they appear on the chart. Level labels can be turned off.
4 — STATISTICS TABLE
On/off; which mapping period it describes (defaults to period 1); slot label time zone (defaults to New York; set it to match your chart); position (all eight edge and corner slots, including top and bottom centre; defaults to Middle Right); text size (defaults to Normal).
5 — ALERTS
Which levels can fire, and what counts as reaching one.
6 — WARNINGS
On/off for the on-chart notices described above. The warning panel automatically places itself away from the statistics table, so the two never overlap wherever you put the table.
════════════════════════════════════════
ALERTS
Six alert conditions are available:
+D reached
-M reached
Open crossed
+M reached
-D reached
Any enabled level
ALERTS COME FROM MAPPING PERIOD 1 ONLY. Worth stating plainly, because nothing on
screen will tell you otherwise: if you untick mapping period 1 and run only period
2, these six conditions still appear in TradingView's dropdown and you can still
create the alert — it simply never fires. If you want alerts on a particular
period, put that period in SLOT 1.
Settings section 5 controls two things. First, which levels are allowed to fire — untick a level and it will never trigger, even if you created an alert for it. Second, what counts as reaching a level:
Touches the level (wick) — fires as soon as any part of the bar reaches the
level. Earlier and more sensitive.
Closes beyond the level — fires only when a bar CLOSES past it. Later, and
fewer false triggers.
TO CREATE AN ALERT
1. Set the tickboxes in section 5 the way you want them, then press Ok.
2. Right-click the chart and choose Add alert (or press Alt+A).
3. In the Condition dropdown at the top, select "Stat-Map (Gigi)".
4. In the second dropdown, pick the level you want. Use "Any enabled level" if
you would rather have one alert covering all of them.
5. Set Trigger to "Once Per Bar Close" for confirmed signals, or "Once Per Bar"
for intrabar.
6. Press Create. Repeat for each level you want separately.
The levels jumping to new prices at the start of a period never by itself sends an alert. A genuine touch on that opening bar does, though — an opening bar that runs from the anchor open straight into a level is a real event, and both the Hit column and the alerts treat it as one.
════════════════════════════════════════
Open source under the Mozilla Public License 2.0. You are welcome to read, learn from and build on the code.
Feedback and bug reports are genuinely welcome — if you find a symbol or timeframe where something looks wrong, please say so.
— Gigi_Luigino
Açık kaynak kodlu komut dosyası
Gerçek TradingView ruhuyla, bu komut dosyasının mimarı, yatırımcıların işlevselliğini inceleyip doğrulayabilmesi için onu açık kaynaklı hale getirdi. Yazarı tebrik ederiz! Ücretsiz olarak kullanabilseniz de, kodu yeniden yayınlamanın Topluluk Kurallarımıza tabi olduğunu unutmayın.
Chart tools built to be read, tested and argued with. If something looks wrong, tell me.
Feragatname
Bilgiler ve yayınlar, TradingView tarafından sağlanan veya onaylanan finansal, yatırım, alım satım veya diğer türden tavsiye veya öneriler anlamına gelmez ve teşkil etmez. Kullanım Koşulları bölümünde daha fazlasını okuyun.
Açık kaynak kodlu komut dosyası
Gerçek TradingView ruhuyla, bu komut dosyasının mimarı, yatırımcıların işlevselliğini inceleyip doğrulayabilmesi için onu açık kaynaklı hale getirdi. Yazarı tebrik ederiz! Ücretsiz olarak kullanabilseniz de, kodu yeniden yayınlamanın Topluluk Kurallarımıza tabi olduğunu unutmayın.
Chart tools built to be read, tested and argued with. If something looks wrong, tell me.
Feragatname
Bilgiler ve yayınlar, TradingView tarafından sağlanan veya onaylanan finansal, yatırım, alım satım veya diğer türden tavsiye veya öneriler anlamına gelmez ve teşkil etmez. Kullanım Koşulları bölümünde daha fazlasını okuyun.