ICT Three-Candle Swings: When a Pivot Becomes Known

A pivot appears under the 8:05 candle on a finished chart. It is tempting to write that the low was known at 8:05. But the mark depends on what the next candle did. At the moment the middle candle closed, that next candle did not yet exist.

This small timing error can give a backtest entries that nobody could have identified live. A three-candle swing is a useful ICT building block precisely because its price and its confirmation time can be written down separately.

Definition and chart context: confirmed three-candle swing

Write five consecutive highs on cards: 1.09174, 1.09236, 1.09236, 1.09212 and 1.09261. Turn them over one at a time. Under a strict three-candle rule, neither equal high is higher than both immediate neighbours. An algorithm that selects the rightmost bar of a plateau can label a high, but it is running a different rule. The two methods should not quietly share the same test results.

For the ICT background, see Institutional Market Structure. The worksheets below use stated study rules for measurement, confirmation and simulated fills. Those choices let you reproduce the examples; they are not claims that a public candle identifies the owner, size or intent of an institutional order.

Why this distinction matters

Liquidity studies commonly use old short-term highs and lows to describe where a raid occurred or which structure displacement broke. A mark that required future bars cannot supply an earlier trading decision. This is especially important when an indicator draws its arrow back on the central candle. The display location answers where the extreme occurred, not when the trader learned it was a pivot.

How to identify it before the next candle

Record a tuple: middle-bar timestamp, extreme price, right-hand confirmation timestamp. On a five-minute chart with bars labelled by opening time, a middle bar opening at 08:05 has a right neighbour opening at 08:10. That right neighbour closes at 08:15. Confirm the platform’s convention before logging the event. At 08:14:59 a last-second print can still disqualify the pivot.

Repeat with highs of 1.09174, 1.09236 and 1.09212. The middle candle now satisfies the strict rule once candle three closes. Its timestamp identifies the pivot’s location, not its discovery. Before the third close an expanding third candle could still invalidate the observation. A historical platform that plots the marker back on candle two can obscure this delay.

Can you identify the pivot without looking right?

Consider three completed five-minute highs: 1.08674, 1.08718 and 1.08703. The middle bar qualifies. Replace the third high with 1.08718 and it becomes an equal-high case under this rule. Replace it with 1.08724 and the original candidate disappears. All three histories are identical up to the second bar’s close. No real-time decision made there can legitimately rely on which of the three endings follows. This exercise catches lookahead more reliably than arguing over the shape of a finished zigzag.

Test the cost of waiting fairly. Count the earlier entries that failed as well as those that would have won. If a simulated order benefits from strict confirmation but takes the price offered three minutes before it, the apparent advantage is simply two incompatible versions of the decision joined together.

How to trade the idea in a paper replay

Use a completed swing as the earlier reference for a later raid or displacement. Keep a separate field for the break rule: a traded price beyond the swing and a completed close beyond the swing are different triggers. If a hypothetical order needs both a confirmed swing and a later FVG, its earliest eligible time is the later of the two confirmation times.

The two replays use constructed five-minute candles and a controlled order model shared across this workbook series. Keeping the execution pattern comparable lets you isolate what the featured concept changes: the reference, the eligibility decision, or the management rule. These selected paths illustrate mechanics. They are not historical signals or a sample from which to estimate a winning percentage.

How a swing acquires a history

On a clean chart, put the letter L beside the candle where a level is located and K beside the candle where the level becomes knowable. If the drawing rule needs a later candle, those letters belong at different times. Add a third mark, A, when the trading instruction based on that information is active. One lesson of market structure is that a single horizontal line on the finished chart can conceal all three dates.

Take a five-minute bar labelled 08:15 by a platform that stamps opening times. At 08:17 you may know its running high and low so far, but the 08:20 close has not happened. A rule using completed closes could not act at 08:17 simply because the historical bar now has a closing value. A rule using a live break could act then, but it must also accept live breaks that later disappear. Those are separate policies, each with its own costs and false starts.

Next write the smallest observation that would actually change your structural label. A wick through the far end, a completed close beyond it and a second bar holding outside are not interchangeable. Requiring more evidence delays eligibility and can worsen the available price. Accepting less can introduce more failed breaks. An audit needs both the improved and the deteriorated cases before saying one rule is preferable.

Finally keep an old map when a new map replaces it. For each version record its activation time, the point it governs, the timeframe and the precise reason it stops controlling decisions. A later version can explain where the chart went without invalidating a decision that followed the map genuinely available earlier. This version history is especially useful when two traders agree on every price but disagree about when a structural change became actionable.

An outside candle can be both kinds of pivot

A middle candle can have a higher high and a lower low than both neighbours. With strict three-candle comparisons it qualifies as both a local high and a local low once the right neighbour closes. That is not a coding error. It is a consequence of using separate comparisons on an outside bar. The five-minute candle by itself does not tell you which internal extreme came first.

Decide whether your market-structure routine retains both labels, discards ambiguous outside centres, or reconstructs their internal order from a finer feed. These decisions affect the sequence of swings, especially when you use pivots to define a dealing leg. A rule that mechanically alternates high and low can create a plausible-looking zigzag while silently inventing the order inside a candle.

Try a second edge case: a plateau lasting four bars. Selecting the first equal high gives an earlier location than selecting the last. If your rule requires the plateau to end before confirmation, neither selection grants an entry during the plateau itself. Record the price, the chosen location and the eventual confirmation time. You can then compare plateau policies without pretending they saw different market prices.

The educational test is to identify which part of the answer changed. A different timestamp with the same extreme might leave a distant target untouched yet alter order eligibility by ten minutes. A different extreme may alter both the decision and stop distance. A pivot library that reports only a pretty marker loses both distinctions.

Worked replay A: A pivot confirmed before the replay begins

The earlier three lows on the setup sheet are 1.09079, 1.09058 and 1.09074. The centre exceeds both neighbours in the required direction. All three candles finished before the small chart began. That last sentence is as important as the numerical comparison: a plotted pivot without a confirmation time is insufficient for this exercise.

The old centre at 1.09058 is available as a liquidity reference before 07:15. The new turning point at 1.09040 needs its own right-hand bar. It becomes a confirmed local pivot only when the bar starting 07:20 ends. The displacement test through 1.09147 is yet another event. Recording all three prevents a marker on an older candle masquerading as a live alert.

Reconstruct the decision before reading the outcome

Cover the candles to the right of 07:15. On this EUR/USD bid chart the test candle opens at 1.09094, trades down to 1.09040 and ends at 1.09067. The close tells you that bids recovered part of the excursion. It does not yet tell you that anyone bought a particular number of lots, or that the prior short-term high will give way. Mark the low as an observation and leave the order field blank.

Now reveal the candle beginning 07:25. Its close at 1.09171 is above the earlier pivot at 1.09147. This is the completed break used by the exercise. Look two candles apart: the high of the bar beginning 07:20 is 1.09104 and the low of the bar beginning 07:30 is 1.09124. Their separation forms the bullish fair value gap in this model. One strong green candle alone would not establish those two outside prices.

A resting buy at their midpoint, 1.09114, can first be submitted at 07:35 under a next-bar convention. The 07:30 candle is named for its opening time, so its low cannot supply a fill for an order that depends on its close. This matters whenever the same screenshot appears to show both confirmation and a cheap entry. Write the order time before revealing the next low, not after finding a favourable one.

Assume a constant 0.7-pip spread throughout this small slice. On the return, bid reaches 1.09089 and the corresponding ask reaches 1.09096. The latter is at or below the requested buy price. We book 1.09114 as the fill, with no price improvement, no delay after activation and enough available size. Those are transparent simulation assumptions. A real candle, even from the correct side of the spread, is not a broker’s confirmation.

What the order would risk in cash

Give the practice account a USD balance of 1440. Its hypothetical cap of one half of one percent is $7.20 for this decision. For a USD-quoted pair with a 100,000-unit standard contract, one pip at one standard lot is $10. Adding 0.5 pip for assumed round-trip commission and other execution friction gives $96.00 of modelled downside at one lot. The spread is already present in the entry and exit quote rules and is not charged a second time.

Divide $7.20 by $96.00 and round down to the next permitted hundredth of a lot. That leaves 0.07 lots, risking about $6.72 under these assumptions. A gap or worse stop fill can exceed that number. Position sizing changes the money exposed; it cannot turn a missing confirmation into a valid entry. When the minimum tradable size exceeds the cap, the worksheet records a pass.

Change just one fact

Change the third earlier extreme to exactly 1.09058. Under a strict pivot rule the old centre no longer wins both comparisons. A platform that selects one member of an equal-price plateau may still display an arrow. The chart indicator and the research rule are now describing different populations. Do not let the arrow adjudicate the disagreement.

Worked replay B: A pivot confirmed before the replay begins

The earlier three highs on the setup sheet are 1.26982, 1.27007 and 1.26988. The centre exceeds both neighbours in the required direction. All three candles finished before the small chart began. That last sentence is as important as the numerical comparison: a plotted pivot without a confirmation time is insufficient for this exercise.

The old centre at 1.27007 is available as a liquidity reference before 07:20. The new turning point at 1.27034 needs its own right-hand bar. It becomes a confirmed local pivot only when the bar starting 07:25 ends. The displacement test through 1.26898 is yet another event. Recording all three prevents a marker on an older candle masquerading as a live alert.

Audit the short from the exit side of the quote

Read the second tape independently. The 07:20 bid high is 1.27034. By the end of the bar starting 07:30, bid has closed at 1.26871, below the pre-existing pivot at 1.26898. The workbook calls that downward displacement. It does not require a guess about which institution initiated the move. The mark is useful because a different reviewer can identify the same closing price on the same feed.

The low of the 07:25 candle is 1.26955 and the high of the 07:35 candle is 1.26925. The later high remains below the earlier low, leaving a bearish execution gap between them. Halving this interval gives 1.26940. A sell limit resting there becomes eligible only after 07:35 has finished. When bid subsequently reaches 1.26970 the paper order can fill at its stated price. No favourable slippage is credited.

It is tempting to score this short from the lowest red wick, because that is what the bid chart displays. The position actually has to buy back. With an assumed spread of 0.8 pip, an ask target at 1.26716 requires a bid at or below 1.26708. An ask stop at 1.27050 needs only a bid at or above 1.27042. Marking these two conversions before scoring prevents a one-spread error on both exits.

Keep the original order record even when the outcome is obvious from the last candle. A reviewer needs the decision price, the reference quote and the activation time, not just a screenshot captioned “short”. If the platform would have rejected a resting order this close to market, or the feed lacks quotes in the return interval, the correct response is to change the fill status. It is not to claim a market execution at the old limit price.

Score the path, then challenge the assumption

The deciding candle opens at 08:00. Its bid range is 1.26683 to 1.26827. With the assumed spread, the ask range is 1.26691 to 1.26835. The lower target is reached before the other exit without requiring a guess about two exits inside this bar. A later movement cannot reverse a booked exit. In particular, a later visit to the original target would not repair a prior stop.

Increase the estimated round-trip friction by another half pip as a sensitivity check, while leaving entry, stop and target fills fixed. Planned risk:reward becomes 1:1.72, obtained by dividing 21.1 pips of possible net reward by 12.3 pips of downside. This answers a cost question only. If widening the actual bid/ask spread changes whether an order is triggered, reconstruct the path instead of merely reducing the result by a fee.

Change just one fact

Change the third earlier extreme to exactly 1.27007. Under a strict pivot rule the old centre no longer wins both comparisons. A platform that selects one member of an equal-price plateau may still display an arrow. The chart indicator and the research rule are now describing different populations. Do not let the arrow adjudicate the disagreement.

What can go wrong in the review?

Some pivot indicators require two or more bars on each side. Their marks can be valuable, but their confirmation delay is longer. A test built on a five-bar fractal cannot claim fills at the confirmation time of a three-bar swing.

Reader exercise: Was the strict high confirmed?

Try the following classification without using profit or loss as a clue. Measure the smaller of the centre high’s margins over its two neighbours. For this worksheet, the value must be at least 1 fractional points. Also require this separate fact: the right-hand candle had closed by decision time. These are conditions for admitting a row to a study. They do not replace the order trigger, size calculation or exit plan. The six records are constructed cases, not a report of observed market frequencies.

Cover the answer column and decide which facts you would request before admitting each record. A passes the numerical test at 5 fractional points and has the other required evidence. B has the other evidence but measures 0 fractional points. Under the stated rule that is enough to exclude B. A clean-looking reaction after B cannot change the comparison smaller of the centre high’s margins over its two neighbours ≥ 1. If you want to study a wider definition, write it as a new version and rerun every observation.

C is the deliberate boundary case. Its value of 1 fractional points equals the threshold. This worksheet uses an inclusive comparison, so C passes. A version that demands strictly more would exclude it. Keep exact prices or timestamps in the original record before rounding for display; values that look equal on a chart may still differ at the feed’s actual precision. The most efficient place to look for a disagreement between two reviewers is often an equality case like this.

D has a smaller value than allowed, so it fails even though the separate prerequisite is unknown. E demonstrates the opposite error: the measured 4 fractional points is acceptable, but the prerequisite is known to be false. F lacks the numerical evidence and its remaining condition is satisfied. There is no basis to call it either a clean pass or a measured failure. Keeping “unresolved” distinct from “no” preserves what data collection could still settle.

A positive provisional difference can disappear on the last tick of the third candle. Two finished bars and one live bar cannot confirm this pattern.

What a second reader should be able to reconstruct

Give a second reader the unannotated candles ending at the decision point, the definition used on this page and the compact evidence packet specified below. Ask for a yes, a no or insufficient evidence before giving them the rest of the chart. If they disagree, compare the raw extrema first, then compare the timing and the inequality. A disagreement about what was knowable is not solved by pointing to the later target.

For this lesson the minimum useful packet is all three highs, their opening times and the third close time. Preserve the original precision and include the clock used. If another reader can reproduce the number but not the label, examine the equality rule and prerequisite. If they cannot reproduce the number, settle the input discrepancy first. These are different faults, and only the second can be repaired by retrieving better data.

Two of the five settled observations pass the written screen. If F ultimately passes, the count becomes three of six; if it fails, two of six. This bound is useful because it exposes the effect of missing evidence without inventing a future bar. It says nothing about predictive accuracy. To investigate that, lock the rule, select new sessions and only then examine what happened after each independently labelled structure.

A test you can run on unseen sessions

Run the same marked chart twice: once using pivot time and once using confirmation time. Count orders that become impossible in the second run. Those are timestamp errors, not trades lost to poor execution.

Make one dated record for each session you inspect, including those in which the required condition never appears. Keep a count of eligible signals, orders placed, orders filled and positions closed. A result per fill answers a different question from a result per session. Separate them before comparing this rule with another way of reading confirmed three-candle swing.

Frequently asked questions

To put this worksheet beside a live chart, use the guides to London Kill Zone timing and the Judas Swing to define the session question. Then compare market structure shifts for confirmation, fair value gaps for the return price, and order blocks for the larger area behind the move.