ORB Trading Futures

Every quarter the contract you have been trading stops being the one everybody else is trading. The change does not happen at expiry. It happens across several sessions before it, as volume and open interest migrate to the next month, and during those sessions the question of which chart you are looking at has an unusually unstable answer.

Volume Leaves Before Price Does

Migration is gradual and it is not symmetrical. Open interest tends to move first, as holders of longer dated positions roll, while day trading volume lingers in the front month a little longer because that is where the immediate liquidity still sits. For a stretch of a session or two the two contracts each carry a serious share of the activity, and neither one is a complete picture of what participants are doing.

The effect on an opening range is that the range you measure is measured on a subset. If a large share of the interested parties are transacting in the contract you are not watching, the edges of your range were set by fewer hands than usual, and an edge set by fewer hands is easier to push through.

The Seam in the Chart

Continuous charts stitch contracts together, and the stitch has to accommodate the fact that the two contracts trade at different prices. The gap between them is not noise. It reflects the cost of carry across the period between the two expiries, and it can be a meaningful distance.

Whatever method your platform uses to join them, back adjusting the older data or simply splicing at the roll date, the result is a price series that is partly synthetic on one side of the seam. Horizontal levels drawn from before the roll and carried forward may sit at prices that never traded in the contract you now hold. A level derived from last week's range can be off by the whole roll differential, and it will look perfectly ordinary on the chart while being wrong.

Ranges Built on a Thinning Contract

A range formed in the dying days of a front month carries two problems at once. Participation is lower, so the extremes carry less authority, and the spread widens intermittently, so the recorded high and low may include prints nobody could actually have transacted at size.

Neither problem announces itself as a visual defect. The candles look normal, the range looks like a range, and the levels look like levels. What has changed is what they mean, and the change is invisible unless you know where you are in the calendar.

Working Through the Week

The straightforward handling is to move when the volume moves rather than when the calendar says expiry. Compare the day's volume in both contracts before the session and trade the one where the activity is. That single check removes most of the problem, because a range measured on the active contract is a range measured on the crowd.

Levels should be recomputed rather than carried. Anything derived from the previous contract, the prior day high and low, weekly extremes, a range from three sessions ago, needs to be redrawn on the contract you now hold or converted by the roll differential. Redrawing is faster and considerably less error prone than converting, and it removes the possibility of applying the adjustment in the wrong direction.

It is also entirely reasonable to skip the one or two sessions where the split is worst. Rollover happens on a schedule published far in advance. A strategy that sits out a handful of sessions a year for a known structural reason is not being timid. It is declining to measure something while the measuring instrument is being changed, which is a different decision from being afraid of the market.

The last piece of housekeeping is the record. If your results are logged by date and the contract changed midweek, a review months later will compare ranges and stop distances across two instruments as though they were one. Noting the roll in the log costs nothing at the time and prevents a confusing set of numbers from looking like a change in the strategy's behaviour.

In a share account, position size is continuous. You can buy any number of shares and tune the risk to whatever figure you settled on. Futures do not work that way. Size arrives in whole contracts, and the money at stake per unit of price movement is fixed by the contract specification rather than chosen by you. That single fact reorders the entire sizing calculation.

From Points to Money

Two numbers define the conversion. The tick is the smallest increment the contract can move. The tick value is what that increment is worth on one contract. Multiply the number of ticks between entry and stop by the tick value and you have the money risked on a single contract, before costs.

Everything else follows from there. The stop distance you were going to use, expressed in points because that is how the chart is marked, becomes a dollar figure only after passing through the multiplier. That multiplier is a property of the contract rather than of the market, so it does not move when volatility does. Only the number of ticks moves.

The Micro and the Full Size Version

Most popular contracts now exist in more than one size, with the smaller version a fixed fraction of the larger. They track the same underlying and their charts are effectively identical, so it is easy to treat them as interchangeable. They are interchangeable in every respect except the one that matters for sizing.

The smaller contract is what makes risk tunable. A stop distance that would risk more than intended on one full sized contract, with no way to buy half of one, becomes adjustable when the unit is a fraction of that size. Traders with modest accounts frequently discover that their real difficulty was never strategy selection but granularity. The smallest position they could take was already larger than their plan allowed, so every trade was oversized before it began.

Stop Distance Sets the Contract Count

The order of the calculation is worth stating plainly, because reversing it is common. Decide the money you are willing to lose on the trade. Measure the stop distance the setup actually requires, which on an opening range is generally the distance to the far edge or to some structure near it. Divide the first by the money value of that distance on one contract. Round down.

Rounding down is not a detail. Rounding up to reach a tidier number turns a fixed risk into a variable one, and it always rounds in the same direction, so the effect accumulates rather than averaging out. A trader who rounds up whenever the answer is awkward is running a slightly larger strategy than the one they believe they are running, permanently.

Where the Arithmetic Goes Wrong

The frequent failure is a constant contract count. A trader settles on a number of contracts and that is what they trade, on a tight range and a wide one alike. Because the stop distance varies with the range and the contract count does not, the money risked varies by whatever factor the range varies by. A single unusually wide session then costs several times what a normal one does, from a strategy that felt consistently sized throughout.

The other failure is moving between products. The same nominal size looks entirely different across contracts because the multipliers differ. Habit carries the contract count across and the risk changes silently, which is precisely the kind of error that produces a loss nobody can explain afterwards, because nothing about the trade looked unusual at the time it was placed.

Costs Are Measured in the Same Unit

Commission and exchange fees are charged per contract, and on the smaller contracts they represent a much larger proportion of a tick. A strategy taking modest targets on micro contracts can find that costs consume a meaningful share of the edge, while the identical strategy on the full sized version barely registers them.

That is not an argument for trading larger than the account supports. It is an argument for including the round turn cost in the arithmetic from the beginning, so that the size you choose and the target you choose are evaluated together rather than one at a time. A target that clears the costs comfortably at one size may not clear them at another, and the point at which it stops clearing them is a property of the contract you selected.

Ask a futures trader when the session starts and you will get at least three defensible answers. The contract trades close to around the clock, the pauses are short, and the word open is being asked to do work it cannot do unaided. Before a breakout rule can be written down, it has to say which open it means, because the ranges produced by each candidate describe genuinely different things.

Three Candidates, Not One

The first candidate is the electronic reopen after the daily maintenance break, which is where the trading day formally begins. The second is the point at which European participation arrives in volume. The third is the cash equity open, which for index products is the moment the underlying market itself starts trading.

Each produces a measurable range in its first thirty minutes and each range is real. They are simply measuring different populations of participants doing different things, and a level derived from one carries no particular authority over the other.

What the Overnight Range Actually Measures

Overnight volume is thin and unevenly distributed across the hours. A range formed then is often the product of a small number of participants, some of whom are hedging rather than expressing a view. Price can travel a long way on very little activity, and it can also sit almost still for hours because nobody with size is present. Both outcomes produce a range you can draw lines around, and neither says much about what the day will do once larger participants arrive.

What the overnight range is good for is context. It marks where the instrument has already been while most of the domestic market was away. If the cash open arrives and price immediately leaves that territory, something changed at the open. If the first hour of the cash session stays inside the overnight range, the open added nothing new and the earlier extremes are still the levels doing the work.

The Cash Open Concentrates Everything

The cash equity open differs in kind, not merely in degree. Orders accumulated overnight execute, index products reprice against an underlying that is finally trading, and participation arrives in a burst rather than a trickle. A range formed in that window is formed by a genuine crowd, which is what makes its edges meaningful: they were reached and rejected by real size rather than drifted through in the quiet.

This is the practical reason most opening range work on index futures uses the cash open rather than the electronic one. It is not convention for its own sake. It is that the edges are defended levels rather than incidental extremes, and a breakout rule depends entirely on the edge meaning something to somebody.

Levels That Survive the Handover

The interesting sessions are the ones where the two ranges disagree. Price spends the overnight hours building a range, then the cash open pushes straight through one of its edges and builds a new, smaller range on the far side. You now hold two sets of lines pointing in different directions.

The overnight extreme that was broken should not be discarded. It has become a level with history behind it, and price returning to it later in the session is a different event from price wandering through an unremarkable price. Treating the cash range as the tradeable structure while keeping the overnight extremes on the chart as reference is usually more useful than picking one and deleting the other.

Picking One and Keeping It

Whichever you settle on, the cost of switching exceeds the benefit. A trader who uses the cash open on quiet days and the overnight range on days when the overnight looks more interesting does not have a rule, only a series of judgements made with part of the outcome already visible. The record that accumulates cannot be evaluated, because it is a record of several different strategies interleaved.

Write down which open you measure, and write down the exception if you want one, for instance that a holiday shortened cash session is skipped rather than measured differently. An exception stated in advance is part of the rule. An exception discovered at the moment it happens to be convenient is not.