Confirmation Bias in Trading: Why You Only Ever Find Reasons to Hold

You’ve probably noticed this pattern in someone else’s portfolio before you noticed it in your own: a friend is down badly on a position, and somehow every article they share, every account they quote, every argument they bring up is bullish. The warnings, the red flags, the people who were right early — none of it seems to reach them. It’s not that they’re lying to you. It’s that it never reached them either.

That’s confirmation bias, and it’s one of the best-documented patterns in how people process information under uncertainty: once you hold a belief — including “this position is going to work out” — you unconsciously seek out, notice, and remember information that supports it, while filtering out, dismissing, or forgetting information that contradicts it. It doesn’t feel like filtering from the inside. It feels like the evidence just happens to keep pointing one way.

Why this is worse in trading and investing than almost anywhere else

Confirmation bias affects how people think about politics, relationships, and diets too. But a few things make it unusually dangerous specifically in a losing trading or investing position:

You get to pick your information sources, and the ecosystem rewards picking ones that agree with you. If you’re holding a token or a stock, following its most bullish advocates on social media doesn’t feel like bias — it feels like “staying informed” and “not listening to FUD.” But an information diet built entirely from accounts that already agree with your thesis isn’t information. It’s an echo of your own hope, dressed up as research.

The feedback loop is fast, public, and social. A crypto Discord or a stock subreddit built around a specific position doesn’t just fail to challenge your thesis — it actively punishes challenges to it. Post something bearish in a community of holders and you’ll typically get downvoted, called a “shill for the competition,” or told you don’t understand the fundamentals, regardless of whether your point was actually wrong. That social cost trains people to stop looking for disconfirming information long before any conscious decision to ignore it.

Every dip becomes retroactively reframed as a buying opportunity. Notice the asymmetry: a price drop after you bought is “a great entry the market hasn’t caught up to yet.” A price drop after you sold is “exactly why I got out in time.” Both explanations can’t be your honest, forward-looking view of the same asset — one of them is usually the bias, fitting the story to whatever position you’re already in.

Selective memory does the rest. People remember the time a bearish call turned out wrong far more vividly than the times it turned out right, if the bearish call was about something they held. This isn’t intentional dishonesty — it’s how confirmation bias operates at the level of memory, not just information intake. You’re not lying about your track record of listening to critics; you genuinely don’t remember most of it accurately.

How it shows up in a real losing position

Take a concrete case: you bought a token at $40. It’s now at $14, a 65% loss. Over the past few months, without deciding to, you’ve:

• Muted three accounts that kept posting bearish takes on the project, because “they clearly have an agenda”
• Joined a Discord server built entirely around holders of this specific token
• Started treating every price drop as a “shakeout” designed to scare out weak hands before the real move up
• Noticed and shared every piece of good news (a partnership announcement, a roadmap update) within hours, but couldn’t name the last piece of bad news about the project if asked
• Started describing people who sold as having “given up right before it turned around” — a claim made about every past exit, regardless of what actually happened to the price afterward

None of these individually looks irrational. Each one has a reasonable-sounding justification in isolation. Together, they describe someone who has built an information environment where it’s nearly impossible to encounter a fact that would change their mind — not because the facts don’t exist, but because the pipeline delivering information has been filtered, one small, defensible-seeming decision at a time.

The test: can you steelman the bear case?

Here’s a single, concrete diagnostic that cuts through most of the self-deception: write down the single strongest argument against your position, in one paragraph, as if you were making the case to someone else — with no rebuttal attached.

Not the weakest bear argument you can find. Not a strawman version that’s easy to dismiss (“bears just don’t understand the technology”). The actual strongest version, stated fairly, in the voice of someone who has genuinely thought about it.

Most people who’ve been quietly practicing confirmation bias for months find this exercise uncomfortable in a specific way: not because they can’t think of any bearish argument, but because they realize every bearish argument they can recall is a weak or bad-faith one. That’s not evidence the strong version doesn’t exist — it’s evidence of exactly what this article is describing. If your mental library of counterarguments is stocked entirely with strawmen, that’s not because the real objections don’t exist; it’s because you’ve been filtering for the strawmen and filtering out the substantive version.

If you genuinely can write a fair, strong bear case and still conclude the bull case wins on the merits — that’s a real, examined thesis, and a legitimate reason to hold. If you sit down to do this and find you can’t produce one without immediately rebutting it in the same breath, that’s the bias talking, not the analysis.

Four ways to fix the information pipeline, not just the moment

Steelmanning once helps with one decision. These change the pipeline that’s been feeding you distorted information in the first place.

1. Audit your sources by ratio, not by vibe

List every account, newsletter, and community you actually check for information about a position you hold. For each one, mark whether it has ever, even once, said something bearish about that specific position that turned out to be worth taking seriously. If the honest answer across your whole list is “none of them,” you don’t have an information diet — you have a fan club, and fan clubs don’t publish bad news about the thing they’re fans of.

2. Follow at least one credible critic on purpose

Not a troll, not a competitor with an obvious axe to grind — someone who has a track record of being right some of the time and is willing to say uncomfortable things. The goal isn’t to be talked out of a position by one account. It’s to make sure disconfirming information has at least one path to actually reaching you, instead of zero.

3. Run a pre-mortem before you add to a position

Before increasing a position size — especially averaging down — write one sentence answering: “if this is down another 50% in six months, what will I wish I had paid attention to today?” If you can’t answer that question at all, that’s itself informative: it usually means you haven’t been exposed to the case against the position clearly enough to even imagine what the warning sign would have looked like.

4. Track your own bearish-call record honestly, in writing

Keep a simple running note of specific bearish predictions you’ve encountered about your own holdings — who made them, when, and what happened. Don’t rely on memory to score this later; memory is exactly the part of this system that’s compromised. A written record is the only way to find out whether critics of your specific position have actually been wrong as often as it feels like they have.

What this doesn’t mean

None of this means you should seek out maximum negativity, treat every bear as automatically right, or sell the moment you encounter a critical argument. Confirmation bias isn’t fixed by swapping one filtered diet for its mirror image. The goal is a genuinely two-sided information environment — one where both the bull case and the bear case can reach you, get a fair hearing, and update your view when the evidence actually warrants it. If you’ve done that and the bull case still wins, that’s conviction. If you’ve never actually tested it, what you have is a filter bubble that happens to agree with your entry price — which is a related but separate trap covered in the sunk cost fallacy in investing, and worth checking against the buy-today test in Down 50%? Here’s the Actual Plan once you’ve actually heard the other side.

FAQ

Isn’t it normal to be more confident in something after you’ve researched it? Yes — genuine research that updates your view based on new evidence is not confirmation bias. The distinction is direction: research should sometimes make you less confident, not just reinforce what you already believed. If you can’t remember the last time something you read about a position you hold made you more cautious rather than more convinced, that’s worth noticing. A one-way information diet is the tell, not research itself.

How is this different from loss aversion or the sunk cost fallacy? They usually travel together but aren’t the same mechanism. Loss aversion is about how a loss feels worse than an equivalent gain feels good — an emotional weighting problem. The sunk cost fallacy is about letting money you’ve already spent influence a decision that should only depend on what happens next. Confirmation bias is different from both: it’s not about how you feel about the loss or what you’ve already spent, it’s about which information reaches you in the first place. You can be perfectly rational about sunk costs and loss aversion and still be working from a badly filtered picture of reality, because you’ve spent months only reading the bull case.

What if the bearish arguments I find really are weak or in bad faith? Sometimes they genuinely are — not every critic is worth taking seriously, and dismissing an obviously bad-faith argument isn’t confirmation bias. The test isn’t whether you engage with every bear case indiscriminately; it’s whether you’re capable of stating the single strongest version of the bear case in your own words, accurately, without a rebuttal attached. If every bearish argument you can recall is a straw man or a scammer or someone who’s “just jealous,” that’s a sign you’ve been filtering rather than evaluating.

Does unfollowing bearish people or muting a bad community actually help, or does it make this worse? It depends on why you’re doing it. Muting genuine noise, harassment, or accounts with no actual argument behind their negativity is reasonable curation, not bias. Muting or leaving a community specifically because it made you uncomfortable about a position you hold — while keeping every bullish source active — is the bias doing the curating for you. A useful check: if every account or community you’ve cut ties with in the last year happened to be bearish on something you were holding, that’s not a coincidence worth ignoring.