Buy cheap, sell expensive — measured against the swing you are trading. The data has a catch in it, and the catch is more instructive than the headline.
Premium and discount split a completed swing in half at its midpoint, so a buy below the middle is cheap relative to the move price has just made, and a buy above it is expensive.
Take any completed swing — a move from a swing low up to a swing high — and draw a line across the middle of it. Everything above that midpoint is the premium half. Everything below is the discount half.
The idea is simple and appealing: if you want to buy, buy in the discount half, where price is cheap relative to the range it has just travelled. If you want to sell, sell in premium. The midpoint is sometimes called equilibrium.
You do not need a Fibonacci tool for this. The midpoint of the swing is the only line that matters, and drawing it takes five seconds.
A shirt normally sells between £20 and £40. At £38 it is expensive; at £24 it is cheap. Nothing about the shirt changed — only where the price sits inside its usual range.
Premium and discount say the same about a market: the price has not become good or bad, it has become expensive or cheap relative to the move it has just made. What that framing cannot tell you is whether the shop is about to go out of business and mark everything down further, which is the part the data below is about.
Every completed swing in five years of hourly candles, sorted by how deep the pullback that followed it went, and then followed forward to see whether the swing resumed.
| Instrument | Shallow pullbacks | Made a new high | Deep pullbacks | Made a new high |
|---|---|---|---|---|
| XAUUSD | 418 | 83% | 1,586 | 36.9% |
| GBPUSD | 419 | 80% | 1,572 | 32.3% |
| EURUSD | 385 | 77.1% | 1,605 | 31.3% |
| US30 | 403 | 70.5% | 962 | 31.5% |
| NAS100 | 416 | 76.4% | 978 | 32.1% |
What was counted: For every confirmed swing from a swing low to a swing high, the range is split in half at its midpoint. A pullback is shallow if it stayed in the upper (premium) half and deep if it reached the lower (discount) half. Resumed means price made a new high above the swing high within 48 hours, before closing below the swing low.
Read the caution below before drawing a conclusion from these two columns — they are not a like-for-like comparison of entry quality, and the reason matters.
Taken at face value this looks devastating for the concept. Pullbacks that stayed in the premium half went on to make a new high 83% of the time. Pullbacks that reached discount did so only 36.9% of the time. Waiting for the cheap price appears to more than halve your chances.
A deep pullback has further to travel to make a new high. That is arithmetic, not evidence about entry quality — some of the gap in the table is guaranteed by the way the two groups are defined, and any honest reading has to subtract it.
What the table does establish is worth having anyway: most swings never reach discount before they are finished. Deep pullbacks outnumber shallow ones roughly four to one, and only about a third of them go on to a new high. A plan that only buys in discount will spend most of its time waiting, and much of that waiting will be in swings that have already died.
It also does not measure the thing that makes discount entries attractive: a better price and a closer invalidation. A 37% strategy with a stop a third of the size can easily beat an 83% strategy with a huge one. This page cannot settle that — only testing entries, stops and targets together can, which is exactly what the drill below is for.
Ben watches gold rally from 4,380 to 4,440. The midpoint is 4,410.
The impatient version. Price pulls back to 4,432 and he buys, because the trend is up and he does not want to miss it. He is buying in premium, near the top of the range, and his invalidation is all the way down at 4,380 — sixty dollars away.
The patient version. He waits for 4,410 or lower. If price gets to 4,405 and there is a demand zone sitting there, he buys with a stop just below the zone — maybe eight dollars of risk instead of sixty.
The honest cost. Around one time in five, price never comes back that far and he misses the trade entirely. And of the times it does come back, only about a third go on to a new high — because a market that pulls back that deep is often a market that has finished going up.
Why he still waits. Not because the discount entry wins more often. It wins less often. He waits because when it works the reward is several times the risk, and when it fails it fails cheaply and quickly. The impatient version has the higher hit rate and the worse arithmetic — which is the trade-off this whole concept is really about.
A lower hit rate at a much better price versus a higher hit rate at a worse one — that comparison depends entirely on your stop and your target, and no article can answer it for you. Replay gold or GBPUSD, take fifty of each, and compare what they actually produced in money rather than in win rate.
Compare both entries in the backtester →Run the drill in the free backtester → Free, no sign-in to begin.
What are premium and discount in trading?
Take a completed swing from low to high and draw a line at its midpoint. Above the line is premium — expensive relative to that move. Below it is discount — cheap. The idea is to buy in discount and sell in premium.
Does buying in discount actually work better?
Not by hit rate. Pullbacks that reached the discount half went on to a new high 36.9% of the time, against 83% for shallow ones. But that comparison is not like-for-like: a deep pullback has further to travel to make a new high, so part of the gap is arithmetic. What it does show is that most swings never resume once they have pulled back that far.
So should I stop waiting for discount entries?
Not necessarily. A discount entry has a lower chance of the swing resuming but a much better price and a much closer invalidation. A 37% setup risking one to make five can easily beat an 83% setup risking five to make one. The hit rate alone cannot settle it — only testing entries, stops and targets together can.
Is this the same as the 0.5 Fibonacci level?
It is the same line, though 0.5 is not actually a Fibonacci number — it is simply the middle. You do not need the Fibonacci tool for this; the midpoint of the swing is the only level the concept requires.
How often does price reach discount at all?
Often — deep pullbacks outnumber shallow ones by roughly four to one in this sample. The problem is not that discount is rare, it is that many of the swings that reach it have already finished going up.
What should I combine this with?
Something structural at the same price. A demand zone or an order block sitting in the discount half is the setup this concept exists to support. Discount on its own is a price filter, not a reason.
A better price with a lower hit rate against a worse price with a higher one — that comparison only resolves in currency. Take fifty of each in the backtester and add up what they actually produced.
Open the free backtester →Beginner exploration
Open each answer for a plain-language way to read Premium and Discount, test it carefully and decide what to explore next.
This page focuses on “Premium and Discount”.Splitting a swing at its midpoint to buy cheap and sell expensive. Measured across 8,744 pullbacks: deep ones resumed 36.9% of the time against 83% for shallow ones — and why that comparison needs care.For “Premium and Discount”, a beginner should identify what the learning guide measures, assumes or teaches before acting on its conclusion.Treat this page's account of “Premium and Discount” as a learning reference rather than a prediction, signal or promise of future performance.
For “Premium and Discount”, translate the idea into a definition you could apply the same way on two different charts.While exploring “Premium and Discount”, work through one example slowly and record which inputs or observations determined the result.Keep your “Premium and Discount” record honest: list the limitation or counterexample before using the concept in a trading plan.Before leaving “Premium and Discount”, practise the definition on unseen history and review consistency before judging performance.
Turn one idea from “Premium and Discount” into a rule with explicit inputs, dates, costs and pass-or-fail conditions.Ask AI to expose missing assumptions in that “Premium and Discount” test, not to guess the next market move.Use the FXAbsolute AI Backtesting Lab to inspect calculations connected to “Premium and Discount” and the assumptions behind them.Reproduce any important “Premium and Discount” result and reserve unseen data before deciding that an apparent pattern is useful.
Continue your exploration of Premium and Discount with the beginner AI prompt guide, or inspect public calculations in the AI Backtesting Lab.