Reading price: naming swing highs and swing lows, saying whether the market is rising, falling or hesitating, drawing the levels that matter, combining timeframes, allowing for volatility, then deciding — and seeing what the ICT method adds to all of it.
Step 27
Swing highs and swing lows
Module 2 had you see that price moves in steps. To read the trend (step 28) and the levels (step 29), those steps first need names.
Swing high and swing low
A swing high is the top of a wave: a candle whose high is above those of its two neighbours, the one before and the one after. A swing low is its bottom: a low below those of its two neighbours.
That is the three-candle rule, the one the Arena game uses. So a swing high is only confirmed when the next candle closes, since that candle can still go above it.
Compare with the last one
Each swing is compared with the previous one:
a higher high is a swing high above the previous one;
a lower high, a swing high below it;
a higher low, a swing low above it;
a lower low, a swing low below it.
Keep the clear swings
The three-candle rule also picks up mere hesitations. To read structure, keep the clear swings: higher (or lower) than the two candles on each side. Below, it finds ten; six are enough.
A clear swing on 5 minutes is often invisible on 1 hour (step 30). And it is just beyond these points that most traders put their stop: that is liquidity, explained in the glossary.
Bullish candleBearish candle
Nasdaq-100 (NQ), 1-hour candles, 5 and 6 August 2025. Six clear swings: swing highs at 23,404.25 (5 August, 8:00), 23,252 (13:00) and 23,227 (6 August, 4:00); swing lows at 23,113.5 (5 August, 11:00), 23,045.5 (18:00) and 23,107.25 (6 August, 6:00). The three-candle rule finds four more, smaller ones, which are not marked. The swing highs come down (two lower highs), but the last swing low is a higher low: the next step says how to read that disagreement.
Key pointA swing high is above both its neighbours, a swing low below both of its own, and neither is confirmed before the next candle closes. Structure is read on the clear swings, not on every hesitation.
A trend is read on four points: the last two clear swing highs and the last two clear swing lows, each compared with the one before.
Up, down, or neither
a higher high and a higher low: the trend is up;
a lower high and a lower low: it is down;
they disagree (a lower high and a higher low, or the reverse): range or hesitation.
The close that confirms
The trend goes on as long as price breaks its structure in the same direction. In a downtrend, a close below the last clear swing low is a break of structure (BOS): the sellers still have the upper hand.
A wick that pokes below without closing there does not count. And a close above the last clear swing high is the first sign the reading has to be done again.
Range and hesitation
In a range, price moves between a ceiling and a floor, and its swings go nowhere. As long as no close leaves those bounds, there is no trend to follow.
Hesitation is the in-between: no new swing high, a swing low holding above the previous one, candles overlapping. The previous reading is no longer confirmed; the prudent thing is to watch without trading.
Always say which timeframe you are reading. And accept “I do not know”: it is often the right answer.
Last swing high
Last swing low
Reading
What you look for
Higher high
Higher low
Uptrend
Buys
Lower high
Lower low
Downtrend
Sells
Lower high
Higher low
Range or hesitation
Nothing: wait for a break
Higher high
Lower low
Range or hesitation
Nothing: wait for a break
The rule of the Arena game “The trend”: two clear swing highs and two clear swing lows, then a close to confirm.
Gold (MGC), 4-hour candles, 30 June to 12 July 2022. The fall: swing high at 1,826.7 (30 June), swing low at 1,783.3 (1 July), lower high at 1,815.2 (3 July), lower low at 1,762.9 (5 July), lower high at 1,771.5, lower low at 1,730.9 (6 July). On 5 July, the 8:00 candle closes at 1,768.4, below the 1,783.3 swing low: the break confirms the fall. From 6 to 12 July, price stays between 1,720.5 and 1,751.5, and on 8 July a single candle makes a higher high and a lower low: no readable trend any more, it is a range.
Key pointTwo swing highs and two swing lows are enough: a higher high and a higher low, uptrend; a lower high and a lower low, downtrend; in disagreement, range or hesitation. A break is confirmed by a close, never by a wick.
Support is a level below price where it has already been stopped; resistance, a level above where it has stalled. In practice: a clear swing high or swing low that price has not gone through since.
Finding the level
Above price, the nearest one is the next resistance; below, the next support: that is the rule of the Arena game.
A level is a zone, not a line: price stalls a little short or overshoots by a wick. Draw it between those reactions. It counts for more:
if price has reacted there several times;
if everybody watches it: yesterday’s or the week’s high and low, round numbers.
The close decides
A wick that crosses and comes back leaves the level intact: it has been “tested”. A clear close beyond it breaks it, and it often changes sides: a resistance cleared becomes a floor, a broken support a ceiling.
Neither promise nor signal
A level says where price reacted, not that it will react again: they all give way in the end. In the middle of a fall, a support is a place to watch, not to buy: it is read with the trend.
Buyers at a support put their stop just below it; sellers under a resistance, just above. Hence those wicks that overshoot before price turns back: liquidity, at the heart of the ICT method.
Bullish candleBearish candle
6E (euro FX future), 4-hour candles, 8 to 21 October 2025. From 8 to 15 October, price stalls under a zone from 1.16705 to 1.1690: swing highs at 1.1689, 1.169, 1.1672, 1.16705 and 1.1671, without a single close above. On 15 October at 20:00, a candle closes at 1.17005: the resistance is broken. On 16 October, price comes back to the zone from above (low at 1.16815) and climbs to 1.17655: the old resistance acts as support. On 20 October at 20:00, a candle closes at 1.16695, below the zone: the support gives way in turn.
Key pointSupport or resistance is a swing high or a swing low price has not gone through: a zone, not a line. It holds as long as no candle closes beyond it, and once broken, it often changes sides.
Module 2 showed it: one market tells different stories depending on the timeframe. Timeframes are always put together from the top down: the higher timeframe gives the context, the lower one the detail.
Three timeframes, three questions
daily or 4-hour: which way is the market going, where are the levels that mattered?
1-hour: is price coming back towards one of those zones, or is it in the middle of nowhere?
15 or 5 minutes: does the structure turn on the zone, where does the stop go?
Never the reverse: a fall on 5 minutes says nothing about the daily trend, which in turn says what that fall is worth.
The useful case: disagreement
The daily rises, the 1-hour falls: not a contradiction, but a pullback in an uptrend. What remains is how far it goes: the answer lies on the higher timeframe’s zones.
Until it reaches a zone, nothing to do; when it gets there, the lower timeframe tells you whether price stops.
Start at the top
A daily swing low sums up days of trading, a 5-minute swing low a few minutes: higher-timeframe levels weigh more. From the lower timeframes, keep only what refines a zone already drawn.
Three timeframes that nest are enough. Stacking six means always finding a reason to do what you already wanted to do.
Timeframe
The question
What you keep
Daily or 4-hour
Which way? Which zones?
The direction, the levels
1-hour
Is price heading to a zone?
The scenario, or nothing
15 or 5 min
Does structure turn on the zone?
The timing, the stop
Three timeframes, three questions, always from the top down.
Daily1 hour
Key pointThe higher timeframe gives the direction and the zones, the lower one the timing. A disagreement between them is most often a pullback: wait for it to reach a higher-timeframe zone.
Volatility is the size of price moves, and it is never constant: the same market can move 200 points one day and 2,000 three weeks later. Both are normal, but not with the same stop.
Measuring it
a candle’s range: its high minus its low;
the ATR (average true range): the average range of the last 14 candles by default, counting the gap from the previous close when price opens higher or lower.
Every platform shows it. It says nothing about direction, only how much the market moves right now.
It comes in waves: calm days follow one another, wide days too, often set off by a big release (step 25). Check the ATR before the session.
Wide days, calm days
The stop goes where the idea becomes wrong (module 3). On a wide day that point is further away: a 20-point stop when price swings 100 points in a few minutes is inside the noise.
The risk in money does not change, so a wider stop forces a smaller size (step 23), sometimes zero. Some days are too wide for a small account: you watch them.
Conversely, a very calm day often does not offer enough movement for a sensible target. Trading nothing that day is not a failure: it is the correct reading.
Measure
Calculation
What it tells you
Range
High − low
The size of one candle
True range
The range, counting the gap from the previous close
The real size of the move
ATR 14
Average of the last 14 true ranges
The usual range right now
The ATR says nothing about direction: only how much the market is moving right now.
Nasdaq-100 (NQ), daily candles, 24 March to 25 April 2025. From 24 March to 1 April, the gap between the day’s high and low averages 413.5 points; on 25 March it is only 231.75 points. From 3 to 11 April the average rises to 1,442.4 points, and on 9 April the day runs from 16,735 to 19,386.75, 2,651.75 points. The same market, two weeks apart: the same stop cannot suit both periods.
Key pointVolatility changes from one week to the next, and comes in waves. The stop follows the current range, the size follows the stop, and a day too wide or too calm is watched, not traded.
Everything so far serves one decision, every session: enter, or do nothing. It is made with a list, in order, and the default answer is “no”.
Three questions about the reading
a readable trend on the timeframe that decides (in a range, nothing);
price sitting on a zone that matters, preferably from the higher timeframe;
structure confirming on the lower timeframe, with a close, not an impression.
Four questions about the trade
the stop goes where the idea becomes wrong: below the last clear swing low on a buy, above the last clear swing high on a sell;
the target is the next level in the direction of the trade, at least 1 R away: closer than the stop, the trade would risk more than it aims for;
the size, computed from the stop (step 23), is above zero;
no release (step 25) or unusual day (step 31) falls in the window.
If everything passes, the trade is taken as it is, stop and target attached as it is sent. At the first “no”, you stay out, and the reason goes in the journal.
Judge the decision, not the result
A well-built trade can lose, a badly built one can win. Over ten trades, chance has the last word; over a large number, what remains is the way you decide.
The Arena game “Deciding” scores that, with no win rate.
Question
If the answer is no
1. Is the trend readable?
Range or hesitation: nothing
2. Is price on a zone?
Wait
3. Does structure confirm, on a close?
Wait
4. Is the stop beyond the last swing high or swing low?
No trade without a stop
5. Is the target worth at least 1 R?
Do not enter
6. Is the size above zero?
Do not enter
7. No release, no unusual day?
Wait
Seven questions, in order. A single “no” is enough not to enter.
Key pointThe default answer is no. Trend, zone, confirmation, stop, a target of at least 1 R, size, calendar: a trade passes all seven questions, or is not taken.
This module belongs to classical technical analysis, and none of it disappears with the ICT method the rest of the site teaches: same charts, same notions, sometimes under other names.
What ICT adds
Classical analysis says a level made price react; ICT asks why price goes there.
Beyond an obvious swing high sit the sellers’ stops and the buy orders of those waiting for the break: it is that liquidity price goes to get.
ICT adds the zones left by fast moves (FVGs) and the hours when those moves happen most (killzones).
The gain is not being right more often, but knowing before entering where the idea becomes wrong: a stop “below support”, placed by eye, becomes a stop below the swing low whose sweep launched the move.
What ICT does not promise
A higher win rate. Fourteen quality criteria of a setup were measured here over thousands of trades (strength of the level, session, premium and discount, higher-timeframe bias…), and none separated good setups from bad ones.
With or without them, the result at 1 R stayed the same: so no page of this site shows a win rate. Be wary of the ones that do.
Question
Classical analysis
ICT reading
Which way?
Swing highs and swing lows
The same, and their breaks
Where will price react?
Support and resistance
Liquidity, FVGs
When?
Rarely specified
Sessions and killzones
Where am I wrong?
Below support, by eye
At the point that invalidates the idea
Win rate
None guaranteed
None guaranteed
The same questions; ICT adds a “why” and a “when”, not a guarantee.
Key pointICT keeps the trend, the swings and the levels, and adds a reason (liquidity), zones and hours. It helps you know where you are wrong; it does not make you win more often.
