The Coincidences We Remember—and the Thousands We Don’t

Why do certain coincidences stay with us while thousands vanish? Explore selective memory, base rates, confirmation bias, and how to test remarkable patterns.

You’ve probably had this experience.

You think about someone you have not heard from in ages.

Twenty minutes later, they text.

You stare at the screen.

That’s impossible.

And you remember that incident for years.

What you probably do not remember are the 147 other times an old friend crossed your mind and nothing happened.

No message.

No phone call.

No knock at the door.

Those moments leave almost no psychological footprint.

There is nothing to tell.

Nothing to screenshot.

Nothing to mention over dinner.

The coincidence survives.

The non-coincidences vanish.

This simple asymmetry explains far more about extraordinary timing than most discussions acknowledge.

It does not prove that every meaningful coincidence is “just confirmation bias.” Nor does it mean intuition is worthless. It means that before deciding an event was astonishingly improbable, we need something our memory rarely provides automatically:

the denominator.

How many chances were there for the remarkable event to happen?

Once you start asking that question, coincidences become not less fascinating—but much easier to think about clearly.

The Phone Call That Became a Story

Imagine that over the course of one year you briefly think about 300 different people.

Former colleagues.

Old classmates.

Relatives.

Neighbors.

People whose names appear when a song, restaurant, photograph or random memory activates them.

Most thoughts disappear within seconds.

Then one afternoon you think:

I haven’t heard from Maya in forever.

Your phone buzzes.

Maya.

That thought becomes a story.

You may tell your partner.

You might message Maya:

“You won’t believe this—I was literally JUST thinking about you.”

Years later, you still remember it.

But suppose the 299 thoughts that produced no matching contact are gone from memory.

Your internal dataset now contains:

Thought of Maya → Maya contacted me.

It does not contain:

Thought of Daniel → nothing.
Thought of Priya → nothing.
Thought of Luke → nothing.
Thought of Aunt Claire → nothing.

Your memory has not necessarily fabricated anything.

The remembered event may be perfectly accurate.

The distortion appears because the comparison set has disappeared.

That is the central problem.

Why Do the “Hits” Stick?

Memory is not a surveillance camera recording every second with equal fidelity.

It is selective.

Events that are emotional, surprising, personally relevant or repeatedly retold often become easier to retrieve than ordinary background experience.

Research on judgment has long shown that people frequently estimate probability partly by how easily examples come to mind—what psychologists call the availability heuristic. Memorable or easily retrieved events can therefore feel more common or probable than less accessible ones.

Coincidences are almost engineered for availability.

They have:

surprise
emotion
narrative structure
personal relevance
a beginning and payoff

“Thought about someone and then they called” is a story.

“Thought about someone and they didn’t call” is not.

Which one do you think your mind files under important?

Selective Memory Does Not Mean “You Imagined It”

This is where discussions of cognitive bias often become unnecessarily insulting.

Someone shares a striking coincidence.

Another person responds:

“That’s just confirmation bias.”

Conversation over.

But identifying a bias is not the same thing as explaining a specific event.

Your friend really may have called moments after you thought of her.

You really may have dreamed about a red bicycle and encountered one the next morning.

You really may have opened a book to a sentence eerily relevant to the question bothering you.

Selective memory does not mean the event did not occur.

It means our assessment of how unusual the event is may be distorted if we remember only matching cases.

That is a much subtler claim.

And a much more useful one.

What This Is Not

Before we go deeper, four misconceptions are worth removing.

Selective memory is not proof that every coincidence is meaningless.

A psychological mechanism can explain why an event is remembered without deciding what significance you take from it.

Probability does not mean “anything can be explained away.”

Good probability analysis requires actual assumptions: how often events occur, how many opportunities existed, what counts as a match and whether hidden causes are present.

Simply saying “random chance” is not analysis.

Confirmation bias does not mean people are stupid.

Biases often arise from cognitive systems that are useful most of the time. We prioritize salient information because paying equal attention to everything would be impossible.

Being skeptical does not require becoming cynical.

You can investigate an extraordinary event precisely because you respect it enough to want to know whether it survives scrutiny.

The Missing Denominator

Suppose I tell you:

“I correctly predicted three things last month.”

Impressive?

You cannot know.

If I made three predictions and all three came true, that is interesting.

If I made 4,000 vague predictions and later selected three successes, it is not nearly as impressive.

The numerator is:

successful predictions

The denominator is:

all opportunities for prediction

Coincidence stories usually give us the numerator.

Our memories rarely preserve the denominator.

This is why base rates matter.

A base rate is the underlying frequency with which something occurs before we consider the striking details of one particular case.

Research on human reasoning shows that people can underweight base-rate information when vivid case-specific details capture attention, although how strongly this occurs varies substantially between individuals and situations.

In everyday coincidence reasoning, that might mean asking:

How often does this person usually contact me?

How many people did I think about this week?

How often do I check the clock?

How many dreams did I have this month?

How many songs did I hear?

How many strangers did I encounter?

How many possible details could have counted as a match?

Without those numbers, “What are the odds?” is often impossible to answer.

How Many Chances Did the Coincidence Have to Happen?

This is one of the strangest things about probability:

Events that sound unlikely become much less surprising when there are enough opportunities for them to occur.

The classic birthday problem demonstrates this beautifully.

Ask:

What is the chance that one particular person shares your birthday?

Fairly low.

Now ask:

What is the chance that any two people in a room of 23 share a birthday?

The probability is already above 50 percent.

Why?

Because you are no longer considering one pairing.

There are many possible pairs.

Statistical work on coincidences repeatedly emphasizes this issue of multiple opportunities. Diaconis and Mosteller described how large numbers of people, events and possible combinations can make seemingly extraordinary coincidences inevitable somewhere, sometime.

Life generates an absurd number of opportunities.

Think of a normal day.

You encounter:

faces
names
numbers
songs
headlines
places
phrases
memories
messages
dream fragments
conversations
images
dates
times

Each can potentially match something else.

Now multiply that by:

365 days.

Decades of life.

Billions of people.

Modern digital connectivity.

It would actually be strange if astonishing coincidences never happened.

“But Mine Was Too Specific”

Sometimes it was.

Specificity matters.

But we need to ask whether the specificity existed before the event or was added afterward.

Imagine you dream:

You’re somewhere near water.

Someone is wearing red.

There is confusion.

You hear a loud sound.

Two days later, you attend a family gathering near a swimming pool. A cousin wears a red shirt. Someone drops a tray.

Afterward the dream feels predictive.

But before the gathering, how many outcomes could have counted?

Ocean.

Rain.

Bath.

Swimming pool.

Blue shirt instead of red?

Would orange have been “close”?

Could “loud sound” have been thunder, a car horn, an argument or broken glass?

The more flexible the matching criteria, the easier it becomes to find correspondence afterward.

This is sometimes called the problem of multiple endpoints: a prediction, memory or event can potentially match many later outcomes. Diaconis and Mosteller identified this flexibility as one of the central reasons coincidences can appear far more improbable than they are.

Specific predictions made beforehand are therefore much more informative than vague memories interpreted afterward.

Compare:

“I dreamed something bad involving travel.”

with:

“At 8:05 p.m. Monday I wrote: I believe my Thursday flight will be cancelled because of mechanical failure.”

The second can actually be tested.

The first can match half a newspaper.

Why Does Retelling Make Coincidences Feel Stronger?

Stories become cleaner with repetition.

Not necessarily because anyone is lying.

Because humans compress.

Imagine the original event:

“I briefly thought about Anna Monday afternoon. Maybe around three or four. I can’t remember exactly. She sent me a message around six.”

Six months later:

“I thought about Anna that afternoon and then she messaged me.”

Two years later:

“I thought about Anna and she texted me right afterward.”

The core event survives.

The uncertainty around timing quietly disappears.

This is one reason prospective recording is so valuable.

Human memory involves reconstruction, and research shows that remembering is shaped by attention, emotion, context and later interpretation rather than functioning as a perfect playback mechanism.

Again, this does not mean the story becomes false.

It means precision can increase in the retelling even when the original experience was fuzzier.

That matters enormously when judging improbability.

“Within ten seconds” and “sometime that afternoon” are very different statistical claims.

Confirmation Bias Is More Subtle Than “Seeing What You Want to See”

Confirmation bias is often caricatured as stubborn people deliberately ignoring facts.

More often, it works quietly.

Once you suspect a pattern, confirming examples attract attention.

Contradictory examples feel less noteworthy.

Suppose you begin believing that seeing a white feather means something important is about to happen.

You see a feather Monday.

Tuesday brings unexpected good news.

Hit.

You photograph the next feather.

Nothing noticeable happens.

Forgettable.

Next week: feather.

That evening someone you miss contacts you.

Hit.

Your confidence grows.

By now you may begin scanning for feathers more frequently without realizing it.

The process feeds itself:

belief → attention → memorable matches → stronger belief → more attention

This does not require dishonesty.

It requires a human brain.

The Frequency Illusion: When Reality Suddenly Seems Full of the Same Thing

Buy a particular model of car.

Suddenly it is everywhere.

Learn a new phrase.

Everyone seems to be using it.

Become interested in repeating numbers.

11:11 appears constantly.

The external frequency may not have changed.

Your detection threshold did.

Attention works by filtering.

If every object in your visual field received equal processing, functioning would be impossible.

So when something becomes important, the brain promotes related information.

This can create the subjective impression:

“This has suddenly started happening to me.”

when a more accurate statement may be:

“I have suddenly started noticing it.”

The distinction sounds small.

It is enormous.

Coincidences Also Benefit From Flexible Categories

You think about your aunt.

Your cousin calls.

Does that count?

You dream about a black dog.

The next day you see a dark-gray wolf on television.

Count it?

You ask for “a sign involving birds.”

A feather appears.

Count it?

This is where humans can unintentionally move the target.

The original prediction expands just enough to accommodate the later event.

Researchers studying coincidences have long noted the importance of “near matches,” flexible endpoints and the vast number of possible correspondences available after an event occurs.

The cure is not becoming humorless.

It is defining the match beforehand when you genuinely want to test something.

When Is a Coincidence Actually Impressive?

Now we reach the fun part.

Suppose you want to distinguish an ordinary coincidence from something genuinely unusual.

Ask five questions.

1. Was it specified beforehand?

Predictions recorded before the outcome deserve more weight than memories reconstructed afterward.

2. Was the prediction specific?

“Something surprising will happen” is nearly impossible to falsify.

“My sister will call between 2 and 3 p.m. today” is testable.

3. Was the time window narrow?

The longer the window, the more opportunities for a match.

4. How common is the predicted event?

Predicting that your spouse will text today is not the same as predicting contact from someone silent for twenty years.

5. Were misses counted?

Ten correct predictions out of twelve attempts would be worth examining.

Ten correct predictions selected from 8,000 attempts would be something else entirely.

This approach does not automatically explain the result.

It simply makes the result interpretable.

The Honest, Harder Truth About Intuition

Here is where both believers and skeptics can become too certain.

Suppose you prospectively track your intuitions and discover they perform better than you expected.

Does that prove psychic ability?

No.

But neither should you automatically dismiss the result.

There are ordinary ways a person can make surprisingly good predictions without consciously knowing why.

Humans absorb subtle cues.

Tone of voice.

Behavioral routines.

Facial expressions.

Timing patterns.

Environmental changes.

An experienced person may detect patterns before being able to verbalize them.

This can create the subjective experience:

“I just knew.”

Sometimes you did know—without knowing how you knew.

That distinction is important.

Unconscious cue detection can produce genuine predictive skill.

So the purpose of tracking is not to debunk intuition.

It is to discover what your intuition is actually doing.

Try This: The 30-Day Coincidence Experiment

For one month, stop relying on memory.

Use a notebook or note-taking app.

Every time you experience a prediction, unusual intuition or apparent synchronicity, record it immediately.

Use this structure:

DATE/TIME

When did you record it?

THE PREDICTION OR EXPECTATION

Write exactly what you believe may happen.

Bad:

“Someone from my past will appear.”

Better:

“I have a strong feeling that my former coworker Aaron will contact me within the next 48 hours.”

TIME WINDOW

When would the prediction expire?

Without an expiration point, predictions can remain open indefinitely until something fits.

CONFIDENCE

Rate it from 1–10.

OUTCOME

Hit, miss or partial match.

Be strict.

POSSIBLE CUES

What information might have influenced you?

BASE RATE

How often does this event normally happen?

NOTES

What did the experience mean to you emotionally or spiritually?

At the end of 30 days, review the entire record.

Not just the best stories.

That final instruction is everything.

Why Prospective Recording Changes the Game

A diary created after the fact captures stories.

A diary created beforehand captures data.

Suppose you write down 25 strong intuitions.

After a month:

3 are striking hits.

5 are partial.

17 are misses.

Your memory might have kept only the three.

The journal preserves all 25.

Now suppose instead:

18 are precise hits.

That deserves much closer examination.

You might investigate whether you are unconsciously using environmental cues, whether your prediction categories are too broad, or whether something genuinely unusual is occurring.

Prospective recording does not tell you what explanation to believe.

It gives you something better:

a clearer event to explain.

The Coincidence Collector’s Paradox

There is an irony here.

The more interested you become in coincidences, the more coincidences you are likely to notice.

That can make it seem as though they are increasing.

But becoming attentive also gives you the opportunity to become a better observer.

The same curiosity that generates bias can generate discipline.

You can ask:

What exactly happened?

How surprised should I really be?

How many chances were there?

What didn’t happen?

What would change my mind?

What meaning does this have even if the cause is ordinary?

Those questions do not destroy wonder.

They refine it.

We Remember the Bell, Not the Silence

Imagine standing in a room containing one thousand invisible bells.

Every thought you have pulls a string.

Most bells remain silent.

Occasionally one rings.

You turn immediately.

Of course you do.

The sound is the event.

But if you forget the hundreds of silent strings you pulled beforehand, the ringing bell can appear almost miraculous.

That is how coincidence memory often works.

We remember:

the dream that resembled tomorrow.

the person who called.

the number that repeated.

the song that appeared.

the prediction that landed.

We rarely archive:

the dreams that went nowhere.

the people who didn’t call.

the numbers we ignored.

the songs that meant nothing.

the predictions that failed.

Understanding this does not reduce life to cold statistics.

It simply restores the missing half of the picture.

Skepticism Should Make the Mystery Better

There is a version of skepticism that says:

“Coincidences happen. Get over it.”

That is intellectually lazy.

And there is a version of spirituality that says:

“There are no coincidences. Everything is a message.”

That is also intellectually lazy.

Both stop asking questions too early.

The more interesting approach is:

That was remarkable. Let’s look carefully.

Maybe probability explains it.

Maybe hidden causes connect the events.

Maybe selective memory magnifies the pattern.

Maybe unconscious perception contributed.

Maybe the event becomes meaningful regardless of mechanism.

And occasionally, after examining all of that, you may still be left saying:

I genuinely don’t know.

That is not failure.

That is the beginning of honest mystery.

The goal is not to make your life less magical.

It is to stop calling every glittering thing magic—so that when something truly unusual happens, you are capable of recognizing just how unusual it actually is.

If you want to explore the experiential side of these events next, read Why Some Coincidences Feel Deeply Meaningful. For experiences that seem to involve knowing something before it happened, continue to Did You Sense Something Before It Happened? Why Premonitions Are So Difficult to Explain—and Even Harder to Test.

And if you want to investigate your own experiences rather than rely on memory alone, begin a Synchronicity Journal and give yourself 30 days of prospective evidence.

You may end up less certain than you were before.

But you will almost certainly become more curious.

And much harder to fool—including by yourself.

Author’s Note

Human beings are storytelling creatures living in a world large enough to generate astonishing coincidences. Those facts do not cancel each other out. Spiritual Awakening Hub’s approach is simple: keep the wonder, count the misses, and resist claiming more than the evidence can carry.