Why do we guess convenience-store drink prices so accurately, yet freeze when someone asks the price of one ounce of gold? Our brains don't process prices as "precise numbers" — they process them as familiar patterns. When a price falls outside that pattern, intuition stops working.

Here are five of the most common biases that sabotage price estimation, each mapped to what you'll feel when playing PriceGuess.

1. Anchoring — the shadow of the first number

One of the most studied effects in experimental psychology. Show someone a meaningless number first (a birthdate, a random digit), then ask them the price of something unrelated. Their guess drifts toward the anchor. Even if the anchor has no real relationship to the answer.

In our Daily mode, the past reference price of $1,200 is a hint — but many players also let it become a ceiling. They guess close to $1,200 even when the asset has since doubled. Anchors help; they also trap.

2. Missing scale sense — the million/billion/trillion blur

Our brains separate 100 from 1,000 easily, but 1,000,000 from 1,000,000,000? Both sound like "a huge number." Yet the gap is a factor of 1,000.

That's why market-cap comparisons trip so many players: $10B and $100B feel close, even though one is ten times the other. Is NVIDIA larger than GE? More telling than the answer is the wobble in intuition about how much larger.

3. Recency bias

The price you saw most recently stays vivid. Someone who remembers Bitcoin crossing $100,000 in 2024 will guess a 2026 price "somewhere around there" — even if it has since corrected to the $40k range. Memory stores peaks, not averages.

In Higher or Lower, when a recently surging asset appears first, players instinctively mark everything after it as lower. The most recent frame swallows the rest.

4. The narrative trap — "stories that feel obvious"

Prices are numbers, but we remember them via stories: "gold is safe so it always goes up," "Tesla is the future so it's always expensive," "housing only ever goes up." Stories have a gravity that real data often disagrees with.

Gold traded flat for most of 2013–2018. Tesla lost over 60% in 2022. Plenty of housing markets have seen more than a decade of real-terms declines. Stories shape patterns; patterns do not reserve the future.

5. Mean / median / extreme confusion

Ask "what's a CEO's pay?" and people think tens of millions. But "CEO" spans a tiny-startup founder to a Fortune 500 chief. The extreme values the press covers dominate intuition — they quietly drag our mental average up.

Salary Quiz questions often trigger this: we reach for "the highest case I remember" rather than the middle of the distribution. Outliers run the show.

Playing against your own biases

You can't win cleanly, but you can lose less.

Think in ranges. Before committing to a single number, decide a plausible floor and ceiling. Staying inside your range avoids the biggest misses.

Pin down date and unit first. "$1,000 in 2015" and "$1,000 in 2026" are not the same unit. Establish time and currency before guessing.

Step away from the latest headline. Recent spikes usually aren't representative. Recall a 1-year and 5-year average to balance out the freshest data point.

Quiz yourself repeatedly. Intuition improves fastest with fast feedback. Doing one or two modes consistently beats binging all of them once.

Closing thought

The goal here isn't "get right answers more often." It's "know where your sense breaks." Once you can name your biases, you gain the small but real moment to double-check before a real decision. If this article acts as a small mirror for your price sense, that's enough.

※ This is an educational overview of behavioral biases. Nothing here is investment advice.