No, it is not weird, and the belief that it is has been measured repeatedly and found to be wrong in a consistent direction.
Recipients of an out-of-the-blue message appreciate it more than the sender predicts, and the effect is larger the more surprising the contact is. People also over-predict how awkward a deeper conversation will be and under-predict the connection it produces. The hesitation is a prediction error, not politeness.
The short version
- Recipients are more pleased than senders expect, and the longer the silence, the more the sender underestimates.
- People consistently underestimate how much others liked them — a gap that persists even among people who have lived alongside each other for months.
- Deep conversations feel less awkward than predicted, and the mis-prediction is largest for exactly the topics that matter most.
- The interventions that move loneliness are the ones that change what people believe about their social world, not the ones that simply schedule more contact.
- Include a specific detail. It gives the message a reason to exist and does the opening for you.
The bias, and how consistent it is
Four separate research groups have measured versions of this and every one leans the same way. Reaching out is appreciated more than expected. Gratitude lands better than expected. Deep conversation is less awkward than expected. Conversation partners liked you more than you thought.
The consistency is what makes it usable. This is not "sometimes it goes well". It is a systematic error running in a known direction, which means you can correct for it without knowing anything about the particular person.
One likely reason: you have access to your own hesitation and none to theirs. You experience the awkwardness of composing the message and imagine they will experience something similar reading it. They mostly experience being thought of.
Why this matters more than a scheduling fix
Across three meta-analyses of loneliness interventions, the approaches that did best were the ones addressing maladaptive social cognition — correcting distorted beliefs about how others perceive you — rather than the ones that simply increased opportunities for contact.
That is a striking result, and it points at the belief rather than the calendar. If the thing standing between you and the message is "they don’t want to hear from me", more free time does not help.
Stated carefully: that ranking rests on a small number of studies within those meta-analyses, so treat it as suggestive rather than settled. The prediction-error findings underneath it are much better replicated.
What to actually send
Do not open by accounting for the silence. "Sorry it’s been so long, I’m terrible" makes the gap the subject of the message and asks the other person to absolve you. It is the most common opening and the worst one.
Open with the specific thing. "I was thinking about your mum’s surgery — did it go alright?" needs no preamble, carries no apology, and gives an obvious thing to reply to. It also does the work that matters: it demonstrates you retained something, which is the signal that produces the feeling of being understood.
If you genuinely have nothing specific, the honest version still beats the apology. "You came into my head today and I realised I have no idea what you’re up to any more" is a real sentence and people answer it.
What we can’t tell you
Whether it will work with this person. All of the above describes averages across studies, mostly of one-off contacts in research settings. Some relationships have ended for reasons that have nothing to do with drift, and no amount of prediction-error correction changes that.
What the evidence supports is narrower and still useful: the specific fear that an unexpected message is unwelcome is, on average, wrong.
Where this comes from
Every claim above traces to one of these. Each has its own page with the method and what it doesn’t show.
People are happier to hear from you than you predict
Liu, Rim, Min & Min (2023) · Journal of Personality and Social Psychology
Deep conversations feel less awkward than people expect
Kardas, Kumar & Epley (2022) · Journal of Personality and Social Psychology
Senders worry about sounding awkward. Receivers just hear the warmth.
Kumar & Epley (2018) · Psychological Science
People underestimate how much others liked them
Boothby, Cooney, Sandstrom & Clark (2018) · Psychological Science
Feeling understood is the working definition of intimacy
Reis, Clark & Holmes (2004) · Perceived partner responsiveness
Read next
Friendships don’t end. They drift.
Almost nobody loses a friend to a falling-out. The shared context disappears, nobody initiates, and one day it has been two years.
Why you forget to text people back
You read it, you meant to answer properly, and then the moment passed. The problem is that replying properly is a task disguised as a message.
How to remember the details your friends tell you
Write it down within minutes, keep it short and specific, and make sure something brings it back before the next conversation.
Kept is what we built from this.
Nine seconds of talking after you see someone. Kept files it against the right person and hands it back the day before it matters. No streaks, no relationship scores, and it never writes your messages.
Or see how it compares to the alternatives, including where they beat it.