Trading Strategies
Breakout Trading: Entries, False Breaks and Sizing
A break through a level is a liquidity event before it is a signal. Where you enter decides your size, and your size decides what the failures cost.
Draw the line first. A level you notice after price has already cleared it is a memory, and the trade you build on it inherits none of the information that made the level worth watching. Breakout trading rests on one decision taken in advance: the figure at which supply has been proven to exist, marked on the chart before the market comes anywhere near it. Everything after that is execution and arithmetic.
The level is drawn before price arrives
A level earns the name by being tested. Two touches is a coincidence you can live with. Three or more, spread across weeks, with price turning at roughly the same figure each time, means real orders sat there and got filled, and the ones that did not get filled are still waiting.
Write the figure down with the date of each touch. A level that last mattered five months ago is weaker than one that stopped the stock nine sessions back, and by the time you are watching a live break your memory of which is which will be worthless.
Round numbers behave like levels for no structural reason at all. Whole dollars, and the tens above them, collect resting orders because people choose round figures when they decide where to buy and where to give up. Nothing on the chart explains it. The orders are there anyway, which is enough. How levels form and why they reverse roles is worked through in support and resistance.
Prefer horizontal levels to trendlines here. A trendline moves every session, so the breakout price moves every session, and a threshold that drifts is one you can always redraw until your entry looks early.
Contraction is the setup
What happens before the break carries most of the information.
A range that has been narrowing for weeks is a market where buyers and sellers have converged on a price and stopped arguing about it. Daily ranges shrink. Volume fades away. Average true range drifts toward the bottom of its own recent history, the volatility bands pinch in, and the chart becomes boring in a specific way that is worth learning to recognise.
Compression matters for a mechanical reason. When a stock has chopped between $62 and $68 for six weeks, every short sold near the top has a buy stop sitting a little above $68, every breakout trader has a buy stop in the same place, and the holders who bought the last failed rally are waiting to sell at breakeven just underneath. Orders pile into a narrow band of prices. That pile is the fuel for the move and the reason the move so often reverses, and you want to be thinking about both before you click anything.
Volume decides whether anybody showed up
Price clears $68. Now check who paid for it.
A break on volume in line with the last twenty sessions tells you some stop orders triggered and nothing else happened. A break on two or three times the recent average says size crossed, because there is no other way that much stock changes hands in a session. Relative volume is the measure worth watching: today’s total against the average for the same point in the day, since a half day figure compared with a full day average will mislead you every morning.
The confirmation is necessary and weak. Heavy volume says participation was real, which is worth knowing, and it says nothing at all about direction over the following week, because the identical print appears when a large seller is using the break to unload and when a large buyer is using it to accumulate, and the tape does not label which one you are looking at. Plenty of heavy volume breaks fail. Volume analysis covers climax volume and the exhaustion readings that look identical at the moment they print.
The stock screener finds range contraction and relative volume in one pass across a few thousand names. That is the only practical way to run this. Eyes do not scale.
The break, or the retest
Both entries work. They cost different things.
Take the $62 to $68 range. Price clears the top and you can buy at $68.20. The measured move adds the height of the range to the breakout point, $68 + $6 = $74, which is a first target with no promises attached, and your stop belongs back inside the range at $65, because price trading there says the break has been rejected and the range is alive again. Entering on the break, the distance is $68.20 - $65 = $3.20. Waiting for a pullback that comes back to $66.50, the distance is $66.50 - $65 = $1.50. With 1 percent of a $24,000 account at stake, that is $240 either way, and the two trades are nothing alike.
| Break entry | Retest entry | |
|---|---|---|
| Entry | $68.20 | $66.50 |
| Stop | $65.00 | $65.00 |
| Distance | $3.20 | $1.50 |
| Shares at $240 of risk | 75 | 160 |
| Position value | $5,115 | $10,640 |
| Gain at $74 | $435 | $1,200 |
| Result in R | 1.8R | 5R |
Same idea, same dollar risk, nearly three times the payoff. The retest looks obviously better on that table, and the table is missing the only thing that decides the question, which is how often the retest arrives at all.
Put placeholder numbers on it, and treat them as placeholders you will replace from your own record. Ten signals. Take every break: four run to $74 and six stop out, so 4 x $435 = $1,740 against 6 x $240 = $1,440, netting $300. Now assume six of the ten came back to the level, and that the four which never looked back contained three of your four winners. The retest rule fills you six times, one winner at $1,200 against five losers totalling $1,200, and ten signals have produced nothing at all.
Change that split and the answer flips. The mechanism is the part to hold on to. A break that refuses to come back refuses because demand is heavy, so the trades you screen out by waiting are drawn from the best ones you had, and your entry rule turns out to be a filter with a bias in it.
Where the idea is wrong
The stop defines the trade. It goes at the price that says the break failed, which means back inside the range with enough room that ordinary noise cannot reach it.
A stop a penny under $68 gets taken out by the first routine pullback, and taken out again on the second attempt, and you have paid twice for a breakout that eventually worked. The $65 stop in the example costs $3.20 a share and buys the right to be left alone. That trade off is settled in the share count, which shrinks to absorb it.
Two ways people get this backwards. They choose a position size that feels comfortable and then hunt for a stop producing a loss they can stomach, which leaves the exit sitting at a price the market has no reason to defend. Or they apply a flat percentage, an 8 percent stop on everything, which ignores the range completely and puts the exit at whatever distance the arithmetic of the share price happens to produce.
Size comes out of the stop
The formula is shares = dollar risk / stop distance. Dollar risk is fixed before you look at a chart and stop distance comes from the range, so the share count is the only thing left to calculate, and it is an output of the other two.
| Entry | Stop | Distance | Shares | Position value | Loss if stopped |
|---|---|---|---|---|---|
| $68.20 | $65.00 | $3.20 | 75 | $5,115 | $240 |
| $66.50 | $65.00 | $1.50 | 160 | $10,640 | $240 |
| $18.40 | $17.20 | $1.20 | 200 | $3,680 | $240 |
| $240.00 | $228.00 | $12.00 | 20 | $4,800 | $240 |
Every row loses $240. The expensive stock with a $12 stop gets twenty shares and the cheap one with a $1.20 stop gets two hundred, and the dollars at risk never move. Share price is irrelevant to sizing. It only decides how the same risk gets packaged.
One adjustment is specific to breakouts. Ranges are wide, breakout stops are therefore wide, and breakout positions are therefore small. Traders who size a breakout the way they size a pullback end up carrying more risk than they planned on exactly the setup most likely to reverse within two sessions. R multiples, portfolio heat, and the caps that stop four correlated breakouts from behaving as one position, are in risk management for traders.
False breaks are a mechanic
A buyer who needs 400,000 shares cannot find 400,000 shares anywhere near the current price. Above $68 sits a pool of resting buy stops, and a stop order is a market order waiting to happen, one that will pay whatever it has to. Push price through $68 and the pool becomes available. Sell into it. Price falls back inside the range within the hour.
From outside that looks like a pattern failing. From the order book it looks like a fill. The same logic runs the other way at the bottom of ranges, and it explains why false breaks cluster at the most obvious levels: obvious levels are where the orders are, and orders are what a large participant is hunting. Who is on the other side of your trade, and why your stop is worth something to them, is set out in market makers and liquidity.
You cannot engineer this away. Price it. Assume a real share of your breakouts will fail by construction, keep the stop at the price where the idea is wrong, and treat the failures as a running cost the way you already treat the spread. A trader who responds to three false breaks by widening the stop has decided to pay more for the same mistake.
Before you commit size to any version of this, run it against history with costs, a genuine out of sample window and an honest count of trades, which is the work in how to backtest a strategy without fooling yourself.
Frequently asked questions
What is a breakout in trading?
A move through a price that has repeatedly stopped the stock, taken as evidence that the supply sitting at that price has finally been absorbed. The level has to be identified before the move, from earlier touches. A breakout found after the fact is a description of something that already happened.
Should you buy the breakout or wait for the retest?
Entering on the break fills you on every signal and gives you a wide stop, so the position is small and the payoff per dollar risked is modest. Waiting for a pullback to the level gives a tight stop and a much larger reward per dollar risked, and it leaves you out of every break that never comes back. The breaks that never come back tend to be the strongest ones, so the choice carries a real cost either way.
Where do you put the stop on a breakout trade?
Back inside the range, at a price that says the break has been rejected. A stop just under the breakout level sits inside normal noise and gets taken out on routine pullbacks. Widening the stop makes the position smaller, which is the correct adjustment and the one most traders skip.
How much volume confirms a breakout?
Compare today's volume with the average at the same point in the day, since a half day figure measured against a full day average will mislead you every morning. A break on ordinary volume usually means a handful of stop orders triggered. Heavy volume says real size crossed, which is a necessary condition and a weak one, because plenty of high volume breaks fail.
Why do so many breakouts fail?
Because resting buy stops pile up above an obvious level, and a stop order is a market order waiting to happen. A large seller who needs liquidity can push price through the level, fill into those stops and let price fall back inside the range. The failure is a feature of how orders cluster, and no amount of chart reading removes it.