Turns Out You and Your Best Friend Aren’t Two Peas in a Pod

Same pod. Different peas.

Source: Rachael Gorjestani / Unsplash

"Birds of a feather flock together" is one of those sayings repeated so often that we stop questioning it. We assume it, we build friendships expecting it, and we rarely ask whether it's really true, or whether it's the reason those friendships work in the first place.

A study published this month decided to check (Yang et al., 2026). Researchers recruited 371 real friend groups of exactly four people each, verified over video call, and had every participant rate their own personality and each friend's personality across the Big Five traits: openness, conscientiousness, extraversion, agreeableness, and emotional stability.

What the Data Showed

The overlap was there, but barely. Across four of the five traits, friends showed only a weak statistical similarity, small enough that the researchers themselves said it "should not be particularly expected." Extraversion showed no meaningful similarity at all. A wildly outgoing person was just as likely to be close friends with someone quiet as with another social butterfly.

Openness was the one trait where friends actually did line up most. That tracks: People drawn to new ideas and unusual experiences tend to end up in the same rooms, at the same events, pulled toward the same things.

The Twist: We Believe It Anyway

Here's where it gets interesting. Even though the real overlap was small, participants consistently rated their friends as far more similar to themselves than the data supported. This wasn't a minor effect. Across nearly every trait, perceived similarity outpaced actual similarity by a wide margin.

We don't just believe the saying. We rewrite our friends in our own image, then mistake the rewrite for accuracy.

This connects to something else we consistently get wrong about the people close to us: we misread their intentions too, and always in the same direction.

The Part That Should Change How You Think About Friendship

The researchers then tested something that hadn't been carefully separated before: Does matching personality, real or imagined, actually make a friendship better? It didn't. Neither actual similarity nor perceived similarity predicted how satisfied people were with the friendship.

What did predict satisfaction was simpler, and had nothing to do with matching. People were happier with friends who were generally agreeable, conscientious, and emotionally steady, regardless of whether those traits mirrored their own. A person's subjective sense of their friend, seeing them as warm, dependable, and open, mattered more than how that friend actually scored on any test.

What This Means for the Way You Pick and Keep Friends

If you've ever felt slightly guilty about being close to someone who doesn't seem much like you on paper, this is a decent reason to let that go. Sameness was never doing the work you assumed it was doing.

What the study points to instead is almost more useful: Look for people who are steady and kind, and pay attention to how you experience them day to day. That combination seems to matter more than any personality quiz could tell you.

The old saying isn't exactly wrong. Birds probably do flock somewhat together. It's just not why the flock stays close.