Confirmation Bias Hits Harder When Everything Is Going Up
Picture a portfolio that's up 30%25 over eighteen months. Every decision looks smart in hindsight. The stock you held through a dip — vindicated. The sector you overweighted — brilliant call. The warning signs you dismissed along the way — irrelevant noise, apparently. This is the environment where confirmation bias doesn't just exist; it compounds. And the insidious part is that a rising market actively disguises it as skill.
Confirmation bias — seeking out information that supports what you already believe, and unconsciously discounting what doesn't — is a cognitive fixture in all market conditions. But in bear markets, reality provides corrective feedback fast. A losing position tells you something. Drawdowns force reassessment. The feedback loop is tight and often brutal. Bull markets break that mechanism. Gains can accumulate for months while your thesis is fundamentally flawed, because price momentum masks the weakness in your reasoning. You're right for the wrong reasons, and the market isn't correcting you yet.
When the Evidence Base Gets Selective
The deeper problem is what happens to information-gathering behavior during sustained rallies. Investors in a bull market tend to read more bullish commentary, follow more bullish analysts, and spend less time engaging with bearish counterarguments — not because the bearish case is weaker, but because optimism feels validated and pessimism feels contrarian to the point of being socially awkward. Research in behavioural finance consistently shows that conviction reinforces selective attention, and rising prices reinforce conviction. By the time a correction arrives, many investors have effectively information-siloed themselves, holding concentrated positions they've never seriously stress-tested against adverse scenarios.
A bull market doesn't reward good analysis — it rewards exposure. Confusing the two is where confirmation bias does its worst damage.
This is partly why late-cycle bull market positioning tends to look so aggressive in retrospect. It isn't just greed; it's the accumulated effect of months of positive feedback making the existing thesis feel unassailable. Each confirming data point gets weighted heavily; each disconfirming one gets rationalized away. The position sizing that felt prudent at the start of the rally starts to look reckless once you examine what evidence was actually being processed — and what was being quietly ignored.
Systematic Scoring as a Check on Selective Reasoning
One structural advantage of machine learning-based analysis is that it doesn't have a narrative. It doesn't find a stock compelling because of a well-written thesis or a charismatic CEO presentation. It processes technical, fundamental, and hybrid signals across thousands of equities and scores them consistently — the same methodology on the way up as on the way down. This matters enormously in a bull market context, because the model doesn't get incrementally more bullish just because prices have been rising. It evaluates current signal quality, not momentum of belief. Platforms like AIStockAnalytics were built precisely to deliver this kind of dispassionate, institutional-grade assessment to investors who otherwise have no systematic check on their own reasoning.
None of this means bull markets should be greeted with suspicion rather than participation. The point isn't reflexive contrarianism. The point is that the most dangerous portfolio mistakes tend to get made when everything feels most certain — and the discipline to ask 'what would have to be true for this position to be wrong?' is exactly what rising prices quietly erode.