Today's market bias

Is the market bullish or bearish today?

There is an honest answer to that question and a dishonest one. The dishonest one is a single word delivered with certainty. The honest one is a direction, a confidence level, and the reasoning that produced both, so that you are able to disagree with it.

Our engine publishes exactly that, once, at the 9:30 AM ET open. It reads the overnight session, the morning's news and the economic calendar, weighs them against each other, and states whether US index futures are leaning bullish or bearish into the cash open. On mornings where those inputs contradict each other it says so instead of inventing conviction, and the confidence number falls to match.

How today's bias is worked out

The engine reads four things and weighs them against one another.

• Macro and the Fed. Inflation prints, rate expectations, what the bond market did overnight, what a Fed speaker said after the close.
• Earnings and guidance. Who reported, what they guided to, and whether the reaction is big enough to move an index or contained to one name.
• Global risk. Oil, geopolitics, the overseas sessions, anything that lands before New York wakes up.
• Market internals. Breadth, volatility, and how the last few sessions have actually been trading.

Each is scored positive or negative and the weighted result is the call. The confidence figure is not decoration: it is how far apart those four ended up. Four factors pointing the same way produces a high number. Two against two produces a low one, and a low-confidence bullish call is a very different instruction from a high-confidence one.

Nothing in it is a chart pattern, and that is deliberate. Most retail bias methods read price, which means they tend to break in the same way on the same day. A read built from news and macro is wrong on different days than your chart is, and two methods that fail independently are worth considerably more than two that fail together.

Using a daily bias without trading it blind

A bias is a filter, not an entry. It narrows which side of the market you are willing to take and it does nothing else. It will not tell you where to get in, where the stop belongs, or when to leave.

Used properly it does three jobs. It stops you hunting a short in a session that grinds up all day. It makes you size down when your own read and an independent one disagree, which is information rather than noise. And it gives you something falsifiable to write in your journal, so that later you can tell a bad day apart from a bad process.

The failure mode is treating it as a signal. A bullish bias is not an instruction to buy the open. There is a fuller treatment of this, along with four other ways traders form a view, in our guide on how to find your daily market bias.

Why we publish the misses

Most people selling a directional read show you the sessions it worked. We publish the record instead, in full, on this page, on the days it flatters us and the days it does not.

You will not find a headline accuracy figure here either, and that is arithmetic rather than modesty: a run this short cannot separate a good engine from a lucky one, so a percentage would imply a precision that is not there. The last ten sessions are above with their marks on them. Judge it by looking.

Flat days are the ones people ask about. When the indices finish within about a tenth of a percent of where they started, the market did not really go anywhere, and grading a directional call against noise would move the record up or down at random. Those sessions are marked with a dash and counted as neither.

Common questions