Two Positions in the Same Sector Are One Position

Diversification is supposed to arrive automatically once there is more than one trade on. It usually does the opposite, because a morning scanner sorted by relative volume tends to return four names from whichever group is moving that day. The list published at orb trading stocks bizstone is grouped rather than ranked for that reason, since trading the opening range on several equities at once can quietly assemble a single large bet on one theme while the risk report still shows four small independent ones.
The Screen Is the Cause

Any filter that selects for unusual movement will select for whatever sector is having an unusual day. Chip names move together. Regional banks move together. Anything driven by an energy price moves together. The screen has no concept of the relationship, so it hands over a list that looks varied by ticker and is not varied at all by exposure.
Correlation Is Highest at the Open

The problem is worse in the first fifteen minutes than at any other point in the session. Early movement is dominated by broad flows, so names that behave independently by the afternoon break in the same direction at the market open. That is precisely the window an opening range breakout operates in, so the assumption of independence fails exactly where the strategy needs it.
Count Risk by Group, Not by Ticker
The workable fix is to write down a small number of groups and cap the total dollar risk allowed in each one, rather than capping risk per position. Two entries in the same group share that cap and each gets half the size. Three would get a third, which usually makes the third trade not worth taking, and that is the intended outcome rather than a side effect.
Include the Index Exposure
Highly weighted large caps carry a further overlap. Being long two of the largest constituents of the nasdaq is close to being long the index twice, and adding a position in qqq on top of that is a third helping of the same trade. Any grouping that ignores index membership will understate the concentration substantially.
Where It Shows Up
Grouped exposure is invisible on a winning day and unmistakable on a losing one. The trade journal entry that reads as four separate stop losses inside ten minutes is one decision taken four times, and it will keep repeating until the cap exists in writing. Reviewing losing days by sector rather than by position is how the pattern gets found before the drawdown does the teaching.