What Is Premium and Discount Confluence?
Premium and discount confluence is a filter: you take an Order Block or Fair Value Gap (FVG) entry only when it also sits in the correct half of the dealing range. Buy bullish arrays from the discount half, sell bearish arrays from the premium half. Location grades the array.
Every dealing range has an Equilibrium at its 50% mark. Below it is discount; above it is premium. A PD Array is any price-delivery zone — an order block, an FVG, a breaker — that price reacts from. The confluence rule says the array and the half must agree, or you pass.
Most traders treat an order block or an FVG as a standalone signal: see the zone, take the entry. That is exactly why so many of those entries fail. The zone tells you where price may react; the half tells you whether that reaction is worth trading.
Premium and discount confluence answers the second question, and it separates reacting to structure from trading with the value logic that moves institutional size.
Why Location Turns a B Array Into an A+ Array
Institutions buy cheap and sell expensive. That is the whole logic of a dealing range: value sits below equilibrium, premium sits above it. When you buy a bullish order block in the discount half, you are aligned with that value logic. When you buy the same-looking order block in premium, you are paying up — fighting the institutions who are distributing there.
This is why two visually identical arrays produce opposite results. The candle structure is the same; the location is not. An unmitigated bullish FVG at the 30% level is a discount entry toward the draw.
The same FVG at 70% is a premium entry against it, and it fails far more often because price is more likely to continue lower to reach genuine discount first.
- Correct half: the array points toward the Draw on Liquidity and offers a favorable entry — room to the target, tight structural risk.
- Wrong half: the array is countertrend to value; even if it holds briefly, reward-to-risk is poor and continuation is likely.
Location does not create the setup — the order block or FVG does. Location tells you which instances of that setup are worth risking money on. That single distinction removes most low-quality PD array signals from your chart.
There is a mechanical reason the wrong-half array fails so often. If a bullish order block prints at 70% of the range, price still has the entire discount half beneath it — unfilled inefficiencies, untested lows, resting sell-side liquidity.
The algorithm has cheap prices left to deliver to, so it tends to trade down into them before any genuine expansion toward the draw. Buying at 70% puts you in front of that delivery. The discount array, by contrast, sits where that downside work is already done, so the path of least resistance is up toward the draw.
The cleanest way to hold the discipline is a one-line rule: do not buy above equilibrium, do not sell below it. In a ranging leg, price above 50% is premium — not a place to be long — and price below 50% is discount — not a place to be short.
The gate is deliberately blunt. It will occasionally veto a trade that would have worked, but it screens out the far larger set of entries taken at bad prices. Over a sample, refusing to buy premium and refusing to sell discount is most of what separates a positive expectancy from a leaking one.
How to Trade Premium and Discount Confluence Step by Step
The method layers the location filter onto a normal PD array entry. Six steps, in order.
1. Define the dealing range and equilibrium
Anchor the range from a clear swing low to a clear swing high (or the reverse) on your working timeframe. The midpoint is equilibrium — the 50% level that splits premium from discount. Draw fibs from 0 to 1 so 0.5 is marked. Get the anchors right; a wrong range moves equilibrium and mislabels every array inside it.
2. Determine the draw and bias
Decide where price is being pulled. If liquidity rests above — equal highs, an old high, a Buy-Side Liquidity pool — the draw is up and your bias is long. If it rests below, the draw is down and bias is short. The draw defines which array direction you are hunting.
3. Find PD arrays toward the draw
Mark the order blocks and FVGs that face your bias: bullish arrays if the draw is up, bearish if it is down. These are your candidate entries — the raw pool before filtering.
4. Filter by half — this is the confluence
Keep only arrays that sit in the correct half. Bullish setups must be in discount (below equilibrium); bearish setups must be in premium (above it). Discard every candidate on the wrong side. Most of your original list disappears here, which is the point.
5. Add further confluence
Stack more agreement onto the survivors: a Liquidity Sweep just before the array formed, a Market Structure Shift confirming intent, and a kill-zone time window (London or New York AM). Each layer raises the grade.
6. Enter, stop, target
Enter at the array. Place the structural stop beyond the array's origin — below the low of a bullish order block, above the high of a bearish one. Target the draw: the liquidity pool your bias pointed at. Partial at equilibrium or an opposing array if you want to bank risk.
Stacking OTE Inside the Correct Half
The Optimal Trade Entry (OTE) zone — roughly the 62–79% retracement of the impulse leg — is itself deep discount on a long or deep premium on a short. So when a PD array lands inside the OTE, inside the correct half, you have stacked confluences pointing at one price.
Concretely: an impulse up sets your range. Price retraces into 62–79% (deep discount). Sitting inside that OTE band is a bullish FVG left by the original impulse. Now three things agree — the array (FVG), the half (discount), and the fib zone (OTE). That is a top-tier long.
The array gives the trigger; OTE and the discount half confirm the location is as favorable as it gets.
The reverse is a useful sanity check. If your array sits at a shallow retracement — say 38% on a long — it is technically in discount but only barely, and the reward-to-risk is thin because your stop still sits well below entry.
Waiting for price to reach the array inside OTE gives you the same directional idea at a deeper, cheaper price, which tightens risk and lengthens the runway to the draw. Confluence is not just presence or absence; deeper alignment inside the correct half is measurably better than shallow alignment.
A Simple Way to Score Setups
Grade each candidate by counting agreements. This keeps premium and discount confluence objective instead of a feel.
| Factor present | Adds | Grade guide |
|---|---|---|
| Valid PD array toward the draw (OB or FVG) | Base requirement | Below this, no trade |
| Array in the correct half (the confluence) | +1 | Array + half = tradeable |
| Liquidity sweep before the array formed | +1 | Array + half + sweep = strong |
| Kill-zone timing / MSS confirmation | +1 | All four = A+ |
An order block on its own is a B. Put it in the correct half and it becomes tradeable. Add a sweep into the array and kill-zone timing, and you have the A+ setup you actually wait for. The score is not magic — it just forces you to name why a trade qualifies.
Worked Example: Same Order Block, Two Outcomes
Take BTCUSDT. Price rallies from 60,000 to 68,000, then pulls back. Your dealing range is 60,000–68,000, so equilibrium is 64,000. The draw is up toward buy-side liquidity above 68,000.
The winning array: a bullish order block sits at 62,400 — below 64,000, in discount, and inside the OTE band. Price sweeps a minor low into it, prints a bullish reaction, and you go long with a stop under 62,000. Target 68,000+. The location was cheap, so risk was small and room was large.
The failing array: now imagine the identical-looking bullish order block had formed at 66,000 — above equilibrium, in premium. Buying there means paying up inside the range. Price is more likely to roll back toward true discount (below 64,000) before any real push to the draw, so the premium entry gets run over.
Same candles, opposite location, opposite result. The filter is the only thing that would have kept you out.
Notice what the winning version required beyond the array itself: the range anchored on the real swing (60,000–68,000), equilibrium correctly placed at 64,000, and the draw identified above the range. Get any of those wrong and the same 62,400 order block could be mislabeled.
If you had anchored the range from 62,000 to 68,000 instead, equilibrium jumps to 65,000, and 62,400 now reads as deep discount for a different reason — the label survives, but only because your anchors happened to stay consistent.
This is why anchoring is step one, not an afterthought: the entire premium and discount confluence rides on where equilibrium sits.
Invalidation and Common Mistakes
The setup is invalid the moment price closes decisively through the array's origin — through the low of your bullish order block or the high of your bearish one. A wrong-half array that price ignores is not a delayed winner; it is a signal you should have filtered out. Cut it.
- Taking every OB or FVG regardless of location. This is the core error the filter exists to fix. An array in the wrong half is fighting value logic — pass on it.
- Anchoring the dealing range wrong. The wrong swing points move equilibrium, which flips arrays between premium and discount. Re-check your anchors before trusting the labels.
- Buying premium / selling discount. The most expensive habit. If the array is a long above equilibrium or a short below it, the confluence is absent — no trade.
- Ignoring the draw. A discount array is only useful if the draw is up. Discount plus a downward draw is not a long.
A scanner can shortcut the tedium here: LiquidityScan can filter detected PD arrays by their premium/discount location within the range, so you only ever look at order blocks and FVGs already sitting in the correct half.
Frequently Asked Questions
Can an order block in premium ever be a valid long?
Rarely, and only when your dealing range is nested — an array in premium of a small internal range can still be in discount of the larger range you are trading. If you resolve everything to one clear range and the array is above that equilibrium, treat it as premium and skip the long.
What if price never reaches the discount array?
Then there is no trade. Price reaching deep discount is a condition, not a guarantee. If it pushes straight to the draw without offering a discount entry, you miss it — which is correct. Chasing a filled move at premium prices is the exact mistake the filter prevents.
Does premium and discount confluence work on all timeframes?
Yes, because dealing ranges are fractal. The same 50% gate applies whether the range is a 4H swing or a 5-minute leg. Best practice is to set bias from a higher-timeframe range and take the array entry inside a lower-timeframe range that agrees with it.
Is this different from just trading OTE?
OTE is one deep-discount or deep-premium band; premium and discount confluence is the broader rule that any array must sit in the correct half. OTE is a subset — the deepest, highest-confluence slice of discount or premium. Every OTE entry passes the half filter, but not every valid half entry is in OTE.
Related query paths
Build the foundation for this filter, then take it into live setups.
- PD Array ICT Explained: A Trader's Guide to Premium & Discount — the array types this filter is applied to.
- How to Draw Premium & Discount Zones (ICT Guide) — mark the halves correctly before you filter anything.
- Equilibrium ICT: The 50% Level Explained — the exact gate that decides buy vs sell zones.
- How to Draw a Dealing Range in ICT (Correctly) — fix the wrong-anchor mistake at its source.
- OTE Explained: The ICT Optimal Trade Entry Zone — the deepest slice of discount and premium for stacked confluence.
- FVG vs Order Block: Which to Trade? — choose the array you run through this filter.
- The A+ Setup: Order Block + FVG Confluence Strategy — a related angle on order block fvg confluence.