Why You Keep Dating the Wrong Person: The Pattern Constant in Your Last 5 Relationships

If your last five relationships ended in approximately the same way, the constant in the equation isn't them. It's you. Specifically: it's the version of you each of them met, and the failure point you reliably produce around the same week of every relationship. Five women producing the same ending aren't five flips of a coin. They're five iterations of the same coin. That coin is yours.

Most dating content tells you the market is broken. This piece tells you the market is fine, and your last five relationships are the best dataset you currently have about your own selection process.

The diagnosis nobody wants

A man at the start of his fourth relationship in twenty-three months. Week three: he sees the friction. Week six: he has the conversation. Week ten: he calls it a phase. Month four: he calls it irreconcilable differences. Month seven: he calls it timing. Month eleven: he's two drinks in at the same bar, with the same friend, explaining the same theory.

The friend nods. The friend has been nodding for three years. The theories change. The endings don't.

A man who keeps producing the same ending across different women isn't a man with bad luck. He's a man whose selection process has a leak. The leak isn't visible to him because he's been auditing the women for the cause, not auditing the process for the leak. He's looking in the wrong place.

You think they were different women. They were. They also met the same man — the one you've been selecting from inside yourself for the past five years. Same opening warmth. Same week-six caution. Same month-four defensiveness. Same exit posture in the final week.

The Pattern Constant is the diagnostic move that puts the variable back where it belongs. It costs you the comforting external explanation. That's why it works.

What "the same way" actually means

The first move most men make when asked to find the pattern: look at who the women were.

They were all anxious. They were all career-driven. They were all from the same city. They were all the youngest sibling. They all had a difficult relationship with their father.

Sometimes the surface trait is real. It is almost always a distraction.

The Pattern Constant doesn't live in the identity of the women. It lives in the structure of how the relationship broke. The shape of the ending is the data. Who the women were is the noise.

A man can date five completely different women — different industries, different families, different cities, different ages — and still produce the same ending across all five. Because the ending isn't a feature of who she is. It's a feature of how the process he ran metabolized who she was. Different inputs, same output.

The thing in common is the system, not the input.

When you look at the type, you find a story you can tell yourself. When you look at the shape, you find the lever.

Where to look

The shape of the ending is made of specific elements. None of them is about her identity. All of them are about the structure of the failure.

• Which week did the relationship break. Was it week three, week six, week twelve, month four, month six, month eighteen? Different weeks reveal different failure modes. Week three failures are reactivity failures. Week six failures are assumption failures. Week twelve failures are decision-defense failures. By month four, you're producing the conflict you already saw in week six.

• Who pulled back first. Did you cool off? Did she? Was it mutual? Notice how often the answer is she did, but only after I had quietly started auditing for the exit two weeks earlier.

• What the stated reason was. Timing. Distance. Compatibility. Different stages of life.

• What the unstated reason was. The thing you actually knew by week eight but spent twelve more weeks not admitting to yourself.

• What your behavior was in the four weeks before the break. Were you over-investing? Withdrawing? Picking small fights? Becoming unreachable? The four weeks before the break are the most diagnostic period in the entire relationship.

• What conversation you had at the end. The closing conversation is often word-for-word similar across relationships, because it's not really a conversation. It's a script your nervous system has rehearsed and deploys when the threshold is hit.

If those six elements look approximately the same across your last three or five connections, you have a Pattern Constant. The Constant is the lever.

The audit

This works as an actual exercise. It takes about an hour if you're honest. Most men will not give it the hour. The men who do are the men who change the next relationship.

Step 1. List your last three to five serious connections. By serious, I mean anything that lasted longer than ten weeks or produced a real ending you remember the texture of.

Step 2. For each one, write down — in plain text, no editorial — the following:

• The week it ended
• Who pulled back first
• The stated reason
• The unstated reason (the one you knew but didn't say)
• Your behavior in the four weeks before the break (specifics: what changed, what you said, what you stopped saying)
• The closing conversation, as closely as you can reconstruct it

Step 3. Strip the names. Strip the cities. Strip the jobs. Replace them with W1, W2, W3 if you need to. The Pattern Constant becomes visible exactly when identity stops being available as a hiding place.

Step 4. Look at the structural similarities across the rows. The shape of the ending. The week. The behavior pattern in the four weeks before. The closing conversation.

Step 5. Identify the constant. There usually is one. Sometimes two. Almost never more than that.

The constant is not a vibe or a feeling. It's a specific structural element that recurs across rows. I pull back at week eight. The conversation I have at the end is always some version of "I think we want different things." I become unreachable in the four weeks before the break. I always pick someone slightly less invested than me and then become hurt when the asymmetry shows up.

That's the lever.

The four predictable failure points

Most Pattern Constants for high-output men cluster around four points in the timeline. Knowing which one is yours tells you what to install.

Week 3 — Reactivity. You committed on a sample size of three messages, two photos, and a vibe. You "knew" by date two. The surgeon does this between cases. The partner-track associate does it after the second dinner. The founder does it after the second pitch meeting. You were deciding from the speed of your own response to her, not from data. The leak here is reactivity. The fix is upstream patience — and a system that makes patience structural rather than effortful.

Week 6 — Assumption. You filled the gaps with stories you found pleasing. She hasn't told you she's loyal but she said "no drama" in her bio, so you assumed. She hasn't told you she's looking for something serious but she replied at midnight with depth, so you assumed. Every assumption is a story you'll later be punished for believing. The fix is upgrading from interview questions (which she can curate) to the Either/Or System (which she can't).

Week 12 — Decision defense. The decision was made on week one's data. By week twelve you're defending the decision instead of reassessing it. The friction that was visible in week six is now producing the conflict you're blaming on her. The fix is treating the early decision as provisional and the data as ongoing.

Month 4–6 — Stored friction surfacing. You called the failure mode in week six and ignored it. Now it's month five and it's producing the conflict you knew was coming. This is the cleanest signal that the leak is upstream — you saw it, and you stayed anyway. The fix is exit discipline at the moment of recognition, not five months later.

Most men have one of those four as their dominant failure point. A few have two. Identifying yours is most of the work.

The line Robbins says under his breath

The Tony Robbins clip everyone shares for the "95% is selection" line has a sentence underneath it that almost nobody quotes:

You got to stop finding what you're looking for in order to justify protecting yourself.

The full read on that line is here.

It belongs in this piece because it's the warning that determines whether the audit produces a Pattern Constant or just produces more of the existing story.

Most men, running the audit, are not actually auditing. They're hunting for evidence that confirms what they've already decided. The pattern was that they were all too anxious. The pattern was that the city is broken. The pattern was that I keep meeting women who aren't ready. The audit, run in that mode, surfaces the same story it started with — better articulated, more confidently held.

That's the failure mode Robbins is naming. You're finding what you're looking for in order to justify protecting yourself from the harder finding: that the constant is upstream of the women.

The audit only works if you're willing to discover something you don't already believe. If you finish it with the same theory you started with, you didn't run the audit. You ran a defense.

"But I really did have bad luck"

This is the most common objection, and the easiest to dispatch.

Bad luck doesn't produce identical outcomes. Bad luck produces variance. Five different women producing five different endings is bad luck. Five different women producing approximately the same ending is not bad luck. It is a process running through different inputs and producing the same output, which is the definition of a process, not luck.

If you flipped a coin five times and got heads five times, you might call it luck. If you flipped it five times and the same specific sequence — heads, tails, heads, heads, heads — came up in roughly the same order on each independent flip, you would correctly suspect the coin. Different women producing the same week-eight conversation are not five flips. They're five iterations of the same coin.

That coin is yours.

What the constant usually is

Pattern Constants for high-output men cluster around a small number of structural categories. If you've done the audit and can't quite name what you found, see if it fits one of these.

• Reactivity timing. You commit too fast, then defend the commitment.
• Friction intolerance. You exit at the first sign of asymmetry rather than navigating it.
• Friction over-tolerance. You stay through obvious red flags because you've decided the woman is good and the decision is yours to defend.
• Layer 2 default-self. You keep bringing the version of you that wins at work to a relationship where it doesn't fit. Covered in detail here.
• Output-attachment vs engine-attachment. You keep selecting women who love your trajectory, not your underlying drive. Covered in detail here.
• Evidence asymmetry. You read attention as evidence of fit and ignore behavior that contradicts the read.
• Stored exit. You start auditing for the exit early, and the audit becomes a self-fulfilling prophecy.

Most men's Constant is one of those. A few are combinations.

What to do once you've found it

The Constant is the lever. Once you've named it, the system has somewhere to apply pressure.

Hard Filters take an upstream class of decisions out of the part of you that gets compromised in week three. If your Constant is reactivity, Hard Filters do the work your week-three judgment can't. More on Hard Filters →

The Either/Or System surfaces revealed preference where direct questions get curated answers. If your Constant is assumption, the Either/Or extracts the information your assumptions have been filling in.

SAI+ scores what's actually there across structural dimensions, so you stop deciding on vibes. If your Constant is evidence asymmetry, SAI+ replaces the part of your brain that's been running on attention rather than behavior.

Midchalance is the internal posture that holds Layer 2 steady when behavior would otherwise pull you back into your default self. If your Constant is Layer 2, Midchalance is the practice.

The full operating system — and the chapters that connect these tools — is the book.

Why most men won't run this

This piece is direct because it has to be.

The audit costs the most comforting story available to a high-output man: that the market is broken and he is the victim of it. Without that story, he has to look at the version of himself each woman met. He has to look at the four weeks before each break. He has to look at the conversation he keeps having and notice it's a script.

Most men will close this tab and tell themselves they got the gist. The gist isn't the work. The work is the hour with a piece of paper and five names on it. The hour is what changes the next relationship.

You can keep reading the next woman as a fresh dataset, or you can read the last five as a dataset that already contains the answer.

The honest sentence

Outcomes don't repeat by accident. They repeat by process. If yours have been repeating, the constant in the equation is you — and the best news available is that the constant is the one variable you can actually move.

Run the hour. Find the Constant. Then read the rest of the system, knowing what it's solving for.

---

Related:

• How to Choose the Right Woman: The Selection Problem Smart Men Keep Getting Wrong — the pillar piece.
• The 95% Rule: Why Most High-Output Men Misread Tony Robbins on Dating — Layer 2 and the version of yourself you've been bringing.
• Why Your Partner Resents Your Ambition (And the Selection Mistake Underneath It) — when your Constant is output-attachment over engine-attachment.
• Hard Filters vs Preferences — how to install the upstream fix once you've found the Constant.

The system end-to-end: The Selection Standard — a 30-chapter decision framework for choosing the right woman without burning years on the wrong ones. Hard Filters, Either/Or System, SAI+, Midchalance. Read more →