ICT PD Array Matrix: How to Rank Premium and Discount Arrays

By this point in the course you have met a long list of zones: fair value gaps, order blocks, breakers, rejection blocks and more. Each one can look like a valid entry on its own, and on a busy chart you might find five of them within a few dozen pips. ICT traders group these zones under one name, PD arrays (premium and discount arrays), and arrange them in a reference list called the PD array matrix. The matrix answers a practical question: which of these zones is worth acting on right now?

What it is

The PD array matrix combines two ideas you already know.

The first is premium and discount. Any dealing range has an equilibrium at its 50% level. The upper half is premium and the lower half is discount. For a long, you want to buy in discount; for a short, you want to sell in premium.

The second is the catalog of PD arrays. A commonly taught version lists them from the range extreme inward, like this:

The two columns mirror each other. The top rows sit at the very edge of the range, where liquidity rests, and the lower rows are arrays that more often form closer to equilibrium.

Be aware that this list is not standardized. Different educators include or drop items (volume imbalances and opening gaps are sometimes added), change the order, or present it as a ranking of strength rather than a menu. This lesson treats it as a checklist organized by location, which is the most practical way to use it, and leaves the question of strict ranking open.

Why the dealing range comes first

The key rule of the matrix is that direction and location matter more than the type of array. A bullish fair value gap is normally a support zone, but if it sits in the premium half of your dealing range, the matrix says it is not a place to buy. It is too expensive relative to the range. You would either ignore it or treat it as a level price may pass through on the way to a deeper array.

This is why the matrix depends on choosing the right range. Draw the range on your bias timeframe using the most recent swing high and swing low that bracket price, as explained in dealing ranges. Change the range and the same array can move from discount to premium.

How to identify it

1. Set the bias. Decide whether you are looking for longs or shorts from the higher timeframe.
2. Draw the dealing range. Mark the swing high, the swing low and equilibrium. Optionally add the 25% and 75% levels.
3. Filter by side. For longs, keep only arrays below equilibrium. For shorts, keep only arrays above it.
4. Catalog what is left. Label each remaining zone: fair value gap, order block, breaker, rejection block, old low and so on.
5. Look for overlap. Zones that overlap each other, or a higher-timeframe array, are usually given more weight.
6. Order your plan. Decide which array you will act on first and what close beyond it sends you to the next one.
7. Size from the stop. Each array implies a different stop distance, so recalculate position size for whichever one you use.

Worked example: a EUR/USD retracement on the 1-hour chart

All prices are illustrative. EUR/USD has rallied from a swing low at 1.0800 to a swing high at 1.0960. The daily bias is bullish, and price is now pulling back.

The range is 160 pips. Equilibrium = (1.0800 + 1.0960) ÷ 2 = 1.0880. The 25% level is 1.0800 + 0.25 × 0.0160 = 1.0840.

Filter. The bullish FVG at 1.0905–1.0912 is a bullish array, but it sits in premium. The matrix removes it from the list of long entries.

Catalog. Three discount arrays remain: the FVG at 1.0862–1.0870, the order block at 1.0835–1.0848, and the rejection block and old low near 1.0800.

Plan. The FVG is shallow, just 10 to 18 pips below equilibrium. The order block sits deeper and straddles the 25% level, so the trader decides to act on the order block and to treat the FVG as a level price may run through. If a 1-hour body closes below 1.0835, the order block has failed and the rejection block near the old low becomes the last array before the bullish idea is in doubt.

Trade plan. A buy limit at 1.0848, the top of the order block, with a stop-loss at 1.0828, below the block with a small buffer. That is 20 pips of risk. A first target at 1.0920 is 72 pips away (3.6 times the risk), and the swing high at 1.0960 is 112 pips away (5.6 times).

On a $10,000 account risking 1% ($100), with a pip worth about $10 per standard lot, size = $100 ÷ (20 × $10) = 0.5 lots. The Position Size Calculator does this for you. If the order block fails, the loss is the planned $100, and the trader does not move the stop to wait for the rejection block.

How traders use it

• As a filter. The matrix stops you from buying a bullish array in premium or selling a bearish array in discount.
• As a sequence. It gives you a fallback order: if one array fails, you already know the next one to watch, or that the idea is finished.
• For targets. Arrays on the opposite side of the range, especially the old high or old low, are natural profit targets.
• Across timeframes. A 15-minute order block inside a daily discount array combines two levels of the matrix.

Common mistakes

• Trading arrays without a range. Without equilibrium, you cannot say whether an array is premium or discount.
• Treating the list as fixed law. The order varies by source and price does not follow a checklist. Use it to organize, not to predict.
• Using the wrong side. A bullish gap is not automatically a buy if it sits in premium.
• Averaging down through arrays. Adding to a losing long at each deeper array multiplies risk. Plan one entry per idea.
• Ignoring stop distance. A deeper array may need a wider or narrower stop; resize every time.

Summary

The PD array matrix organizes ICT's entry and target zones into premium arrays and discount arrays of a chosen dealing range. Draw the range, mark equilibrium, keep only the arrays on the side that matches your bias, and then decide in advance which one you will use and what invalidates it. The exact list and its order differ between educators, so treat the matrix as a way to structure your analysis, and keep risk per trade small, because any array can fail.