Reaction vs Prediction
Markets don't always react the way plain logic says they should — good news can sink a stock and bad news can lift an index. That gap is exactly why measuring reaction beats guessing.
The obvious reaction isn't always the real one
It seems intuitive that good news should push a stock up and bad news should push it down. Markets don't run on that simple a rulebook. What actually moves a price is the gap between what happened and what was already expected — and expectations are invisible unless you go looking for them. That's why a naive read of a headline and an actual market reaction can point in opposite directions.
Strong data, falling stocks
A hotter-than-expected jobs report looks like unambiguous good news — more hiring, a healthier economy. But if the market has been pricing in interest rate cuts, strong employment numbers can argue the central bank has less reason to cut soon. Bond yields rise, future corporate earnings get discounted more harshly, and stocks fall on what was, by any plain reading, good economic news.
Weak earnings, rising stock
The mirror image happens just as often. A company misses its revenue estimate but raises full-year guidance, and the stock jumps because investors were bracing for something worse. Whisper numbers — the unofficial expectations that circulate ahead of a print — often diverge from the published consensus, so a headline "miss" can still beat what the market had actually priced in.
Why the reaction is the real signal
Guessing at a story's tone from the headline alone would get both of these backwards. That's the core reason AIOVEL classifies stories by how the market actually responded rather than by whether the news sounds positive or negative on its face. For deeper walkthroughs of each pattern, see Why Markets Rise on Bad News and Why Good News Can Crash Stocks — both are specific cases of this same reaction-versus-expectation dynamic.
Browse today's sentiment tags and see reaction versus expectation in action on the live dashboard →
Quick answers
Why do stocks sometimes fall on good news?
Because the news was still weaker than what was already priced in, or because it reduces the odds of something the market wanted, like a rate cut.
Why does AIOVEL classify by reaction instead of headline tone?
Because headline tone and actual market reaction frequently diverge — expectations, not just facts, drive price moves.
Is reaction more useful than a prediction?
For understanding markets, yes — a verified reaction is a fact you can build intuition from; a prediction is an unverified guess until it either happens or doesn't.