It ends one of two ways, and both of them feel like proof the story was right all along.
It usually ends one of two ways, and you've probably lived both.
The first: they get tired. The person who was reaching for you, planning around you, pouring energy in, slowly stops. The texts thin out. The effort dries up. And eventually they leave, or just quietly fade, worn down from trying to get close to someone who kept adjusting the distance every time they got near. You feel relief, mostly. And underneath the relief, a grim little confirmation clicks into place: see. They wanted too much. This is exactly why I'm better on my own.
The second: they don't leave, and that turns out to be its own kind of stuck. You're still together, months or years in, and on paper it's fine. But you're half in it. Present at dinner, absent underneath. You've built something functional and pleasant and faintly hollow, a life you technically share without ever fully arriving inside it. And you're lonely, which makes no sense on the face of it, because you are not alone. You are lonely inside your own relationship, which is a specific and disorienting flavor of the feeling.
Here's the mechanism that keeps you avoidant for the long haul. Whichever of those two endings you land on, your system reads it as evidence. If they leave worn out, that proves people want too much and independence is safer. If you stay half-in and quietly lonely, that proves relationships just don't deliver whatever everyone else seems to get. Both outcomes confirm the exact belief that produced them, and the pattern keeps renewing its own lease by handing you the wreckage as the receipt.
The first ending, and the part the verdict skips
The uncomfortable logic of the first ending is this. You kept the distance to stay safe. The distance wore them out. They left. And you concluded people want too much.
But look at the actual order of events. It wasn't that they wanted too much, it's that reaching for you eventually stopped being sustainable, which is a genuinely different thing. And the research is blunt about it. Dyadic studies, the kind that measure both partners, find that one person's avoidance lowers not only their own relationship satisfaction but their partner's as well, and meta-analyses have found avoidance is actually a stronger drag on satisfaction than anxiety is. A 2025 longitudinal study by Sierra Peters and colleagues, following 539 newlywed couples, found that attachment insecurity predicts divorce specifically through a slow decline in satisfaction over time. The distance drains the person across from you, quietly, on a schedule, until the tank is empty and they go.
I want to be careful here, because this is not "you drove everyone away and it's all your fault." Sometimes a partner genuinely does want more than you want to give, and that's a real, no-fault mismatch. But if it's partner after partner, worn down in the same particular way, the common denominator is worth an honest look. Not to convict yourself. Because a self-fulfilling prophecy is the only kind you can actually interrupt.
The second ending, and why it's lonelier than being alone
The half-in version has its own body of evidence. Avoidance predicts lower intimacy and satisfaction even inside relationships that stay perfectly intact. Ting Li and Darius Chan's meta-analysis found avoidance especially undermines closeness and the felt sense of connection, which happens to be the exact ingredient that separates a relationship from a room you share with someone.
So the hollowness is real and predictable, and it says less about the relationship being broken than about you never having fully joined it. You are getting precisely the amount of connection you allowed in, which turns out to be not quite enough to feel met. Being lonely next to someone who would meet you, if you let them, is a harder loneliness than an empty apartment, because the empty apartment isn't standing right there wondering why you won't let it in.
My honest take
"I'm better off alone" has a way of being true, but only because you keep quietly making it true.
You hold back. The connection starves. It either ends or goes hollow. And you conclude closeness doesn't work for you. But you never actually ran the experiment, because you were alone the entire time, even inside the relationships, so of course becoming officially alone doesn't register as much of a change. You're not really choosing solitude over connection so much as choosing a familiar solitude over an unfamiliar risk, and then reading the result back to yourself as proof. The verdict was written before the trial started.
What to do with it
The work here is catching the verdict as it forms and refusing to sign it on sight. Notice, name, and then test the thing you've always assumed.
Interrogate the "proof." When a relationship ends or goes hollow and "I'm better off alone" arrives, stop and ask a sharper question: is this proof that closeness doesn't work for me, or proof that holding back doesn't work? Those are wildly different conclusions drawn from the same event, and you've been reaching for the wrong one.
Look at the common denominator without flinching. If partners keep leaving worn out in the same way, that's information, not a character sentence. Get curious about your part in it instead of either ignoring it or using it to beat yourself up.
Name the loneliness while you're still in it. If you're in the half-in version, say it out loud, to yourself first and maybe then to them: "I'm here, but I've been keeping myself at a distance, and I think that's why this feels lonely." That single sentence runs exactly opposite to the pattern, and it's available any day you're willing to use it.
Run one real experiment before you conclude anything. Before you decide you're better off alone, actually show up, fully, one time, in a small survivable way, and watch what comes back. You cannot honestly rule out closeness on the strength of an experiment you never let happen.
You're not better off alone so much as better defended alone, and your nervous system has spent a lifetime treating those as the same thing. The two endings that keep convincing you otherwise were both authored by the same reflex, and a reflex is not a verdict on what you're capable of. You've proven, many times over, that holding back doesn't work. You've simply never let yourself run the other experiment.
Sources
Li, T., & Chan, D. K. S. (2012). How Anxious and Avoidant Attachment Affect Romantic Relationship Quality Differently: A Meta-Analytic Review. European Journal of Social Psychology, 42(4), 406–419.
Peters, S. D., Meltzer, A. L., & McNulty, J. K. (2025). Own and Partner Attachment Insecurity Interact to Predict Marital Satisfaction and Dissolution. Social Psychological and Personality Science.
Mikulincer, M., & Shaver, P. R. (2007). Attachment in Adulthood: Structure, Dynamics, and Change. Guilford Press.