What premium and discount actually mean, in ICT terms
In ICT methodology, premium and discount are the two halves of a dealing range split at its 50% midpoint, and ICT traders use them to decide which part of a range to trade in. The premium zone sits above the midpoint, where the methodology reads price as expensive and looks to sell.
The discount zone sits below the midpoint, where it reads price as cheap and looks to buy (The Inner Circle Trader, ICT methodology).
ICT stands for the Inner Circle Trader, the public name of trader Michael Huddleston, and premium with discount is a central pillar of his 2022 Mentorship Model. The terms are his, taught as method rather than derived from a textbook, and the recurring practitioner phrasing is that in premium you are a seller and in discount you are a buyer.
I keep the phrase "in ICT methodology" on the definition because the framing decides everything that follows. Premium and discount are labelled regions inside a community's method, not facts about where value sits on the tape, and whether trading the right half of a range reliably helps is an empirical question each time.
The dealing range and equilibrium explained
A dealing range is the price distance between a defined swing high and a defined swing low, and it is the container inside which premium and discount are measured. The range is simply the high minus the low, and ICT traders draw it between two structural reference points on the timeframe they are analysing (The Inner Circle Trader, ICT methodology).
Equilibrium is the 50% midpoint of that range, calculated as the swing high plus the swing low divided by two. It is mathematically identical to a 0.5 Fibonacci retracement of the swing, and it is the line that splits the range into the premium half above and the discount half below.
I read equilibrium as a reference level, not a tradeable signal, and the methodology agrees. It tells you which zone you are standing in, and the practitioner line that equilibrium is where buyers and sellers agree on price is the method's narrative explanation of the midpoint, not a claim about order matching you can verify.
The core rule: buy in discount, sell in premium
ICT compresses the framework into one directive. If your bias is bullish, you wait for price to enter the discount zone before looking for longs, and if your bias is bearish, you wait for price to enter the premium zone before looking for shorts (The Inner Circle Trader, ICT methodology).
I call this the WHERE layer because it answers where to look before any entry object appears. After a break of structure sets your bias, premium and discount narrow your search to the half of the range that trades with that bias.
The rule is an asymmetry preference, not a prohibition. Continuation traders and scalpers sometimes enter outside the strict zone, and the rule raises probability in the methodology's own terms rather than forbidding a trade.
Reading it as a law is how beginners skip good continuation entries or force bad ones.
What a PD array means, and why the term has two meanings
The term PD array is the most overloaded label in ICT, and the page that ignores either meaning leaves its readers confused. In its first sense, a PD array is the premium and discount framework itself, the grid of expensive and cheap areas stretched across the dealing range.
In its second sense, more common in the 2022 Mentorship Model, PD arrays are the specific price objects ICT traders look for inside those zones as entry triggers (The Inner Circle Trader, ICT methodology).
The acronym literally stands for premium and discount, so the framework sense is the original one, while the object-list sense grew as Huddleston grouped the entry tools together. Both uses are current, and Huddleston himself uses them interchangeably, which is the source of the confusion you will hit the moment you read a second ICT educator.
| PD array object | What ICT traders mean by it |
|---|---|
| Order block | The last opposite-colour candle before a structure-breaking move, already covered in its own guide. |
| Fair value gap | A three-candle price imbalance left by a fast move, also covered in its own guide. |
| Liquidity | Resting orders above old highs and below old lows that price is hypothesised to seek. |
| Breaker | A failed order block on the opposite side that flips to support or resistance after a structure shift. |
| Mitigation block | A zone price is said to return to in order to reduce a prior position before continuing. |
I read PD array as shorthand for the objects you find inside the premium or discount part of a range. An order block sitting inside the discount half is the combination the methodology rates highest, which is why the two ideas are taught as a stack rather than separately.
How ICT traders draw the zones
The drawing rule is mechanical. Identify a recent, significant swing high and swing low, stretch a Fibonacci retracement between them, and read the 0.5 level as equilibrium, with everything above it premium and everything below it discount (The Inner Circle Trader, ICT methodology).
The judgment lives in choosing the swings. ICT traders pick structurally meaningful highs and lows, the same ones a break of structure would mark, rather than every minor wiggle, because a range drawn on noise produces a meaningless grid.
You do not need the Fibonacci tool to find equilibrium, despite how it is taught. The midpoint is the high plus the low divided by two, and a horizontal line at that price does the same job.
I mark it once and read off the two halves rather than reaching for indicators, because a single calculation is all the method requires.
Picture a swing low at 100 and a swing high at 120 to see the whole grid at once. The dealing range is 20 points, equilibrium sits at 110, and the premium zone runs from 110 up to 120 while the discount zone runs from 100 up to 110.
With a bullish bias, I watch the discount half for an entry object, because that is where the methodology says the asymmetry favours longs. Flip the bias bearish and the same logic pulls me into the premium half to look for shorts.
The fractal principle: every timeframe has its own range
ICT doctrine holds that every timeframe carries its own dealing range. A range on the monthly chart contains ranges on the weekly, the daily, the four-hour and so on, nested inside one another, and a trader's higher-timeframe range sets the overall bias while the lower-timeframe range is where entries are sought (The Inner Circle Trader, ICT methodology).
The practical consequence is that a discount on the one-hour can sit inside a premium on the daily. A trader who buys the hourly discount while the daily range is in premium is trading against the higher-timeframe bias, which the methodology reads as the wrong side of the larger range.
I flag the nesting because it is central to the method and almost always under-explained. The fractal principle is directly analogous to Richard Wyckoff's idea that accumulation and distribution occur at multiple scales, and ICT's premium and discount framework is a more mechanical, Fibonacci-anchored restatement of that older range logic.
What is actually proven about ranges and the 50% level
The honest evidence sits one level up from ICT. The 50% retracement as a reference level has a documented lineage that predates Fibonacci-ratio analysis, running through Dow Theory's 50% principle and the work of William Delbert Gann on half-way retracements, so the midpoint ICT calls equilibrium is an old observation rather than a modern discovery.
Mean reversion toward a range midpoint is a real, documented phenomenon. Prices exhibit a temporary mean-reverting component alongside their random-walk component, which is the academic anchor for the observation that price tends to travel back toward the middle of its prior swing.
None of that validates ICT's specific institutional narrative, only the broader tendency.
I reach for the volume-profile value area when I want the data-driven version of premium and discount. The value area is the price range where roughly 70% of a session's volume traded, taught in exchange education such as CME Group's, and price above it is genuinely expensive relative to where volume settled while price below it is genuinely cheap.
That is what reading premium and discount looks like when you use traded volume instead of a Fibonacci line. Carol Osler's Federal Reserve Bank of New York research adds the mechanism, documenting how stop orders cluster near prior levels, which is why those zones appear to persist.
The win-rate claim nobody can source
You will read that trading in the discount zone gives a 70 percent win rate, or that premium and discount entries win eight times out of ten. I have not found a peer-reviewed or tier-1 study that tests ICT premium and discount rules and publishes a win rate, and neither has any page I have seen quote one with a citation.
Every specific percentage traces back to an unsourced claim repeated until it reads like data. The honest statement is that no verified backtest exists in the public literature, and anyone quoting a number should link the study or retract it.
I would rather tell you the evidence is missing than invent a reassuring statistic. The absence of a clean win-rate study is itself the finding, and it means the only honest way to size the edge is to test the method yourself on your own data and treat your result as a sample, not a certainty.
Where premium and discount fail
A dealing range is only valid until one of its boundaries breaks, and I want that limitation stated first. The premium and discount grid is built on a specific swing high and low, and when price closes beyond either, the range is broken and the zones drawn from it stop applying.
The zones also fail when traders treat equilibrium as a signal. The midpoint is a reference line that tells you which half you are in, and ICT doctrine itself says not to trade off it, yet beginner pages encourage entries at the 50% level as if it were a turning point.
The deeper failure is treating the framework as the trade. Premium and discount tell you where to look, but most zones do nothing without an order block or fair value gap inside them to trigger the entry, and the claim that banks are accumulating in the discount half is a story the methodology tells to explain the retracement rather than proof of it.
The swing-trading application of SMC is where the zone, the object and the structure combine into a setup you can actually execute.