Why do I keep ending up with people who pull away?

Usually it is not bad luck. Attachment research suggests a loop: people who worry a lot about closeness and people who back away from it tend to end up together, and the harder one side reaches, the more the other side retreats.

That does not make you the problem. It does mean the pattern has a shape, and shapes can be noticed.

What does “pulling away” actually look like?

It rarely looks like a breakup. It looks like a reply that takes six hours after a weekend that felt like the start of something. It looks like plans that get vaguer as you get closer. Sometimes it is a person who was warm and cold from the very first week, so you never got a steady baseline at all.

Psychologist R. Chris Fraley, who studies adult attachment at the University of Illinois, describes the research as measuring two separate things. One is attachment anxiety: how much you worry about whether a partner is really there for you and whether they might leave. The other is attachment avoidance: how uncomfortable someone is with opening up and depending on another person. Everyone sits somewhere on both. People who run high on avoidance tend to step back when things get close. People who run high on anxiety tend to step forward when things feel uncertain.

Put those two tendencies in the same relationship and you get a very familiar dance.

Do anxious and avoidant people really find each other?

There is evidence that they do, at least in some pairings. In a 1994 study of 354 dating couples, Lee Kirkpatrick and Keith Davis found that attachment styles were not paired at random. They found no couples in which both partners were anxious and none in which both were avoidant. Relationships between avoidant men and anxious women were also surprisingly stable over three years, even though both partners had rated those relationships relatively poorly at the start.

That last detail is the useful one. A pairing can last without feeling good. If you tend toward anxiety, a partner who keeps a little distance can feel like a puzzle worth solving, and the effort of solving it can feel like intimacy. If you tend toward avoidance, a partner who brings the closeness means you never have to initiate it yourself. Each person fits the other’s expectations, and familiar can be mistaken for right.

Why does chasing make it worse?

Because pursuit and withdrawal feed each other. Communication researchers call this the demand-withdraw pattern: one partner pushes with requests, questions or complaints, and the other responds by going quiet or avoiding the topic. A 2014 meta-analysis led by Paul Schrodt, pooling 74 studies and more than 14,000 participants, linked the pattern with lower relationship satisfaction, less intimacy and poorer communication. As University of Florida extension educators summarize it, the person pushing feels shut out, and the person withdrawing feels nagged.

Here is how it plays out. Sam texts twice without an answer, then sends a longer message asking if everything is okay. To Sam, that is reaching for reassurance. To the person receiving it, it can read as pressure, which makes stepping back feel even more appealing. Then the silence makes Sam reach harder.

An older observational study shows the avoidant side of this up close. In 1992, Jeffry Simpson, Steven Rholes and a colleague secretly videotaped 83 dating couples while the woman waited for an activity she expected to be stressful. Partners who were more avoidant behaved differently from more secure partners in how much they touched, comforted and sought support. Under stress, the more avoidant partners did less reaching for comfort and less offering it. Distress did not pull them in.

None of that tells you what a specific person is thinking. It does suggest that sending more, louder, faster is unlikely to close the gap.

Are some people already halfway out the door?

Some are, and research suggests it can show early. In a set of eight studies published in 2011, Nathan DeWall and colleagues found that people higher in avoidance paid more attention to attractive alternatives, showed more day-to-day interest in meeting other people, and reported more infidelity over time. Lower commitment explained the link.

So when someone keeps the apps on their phone “just because” or talks about their ex like an open question, that may be information about where they are, not a challenge for you to overcome.

Why would you keep choosing that?

Partly because uncertainty grabs attention, and partly because of what you are afraid of. In 2013 research from the University of Toronto, Stephanie Spielmann, Geoff MacDonald and colleagues found that people with a stronger fear of being single were less selective about who they dated and more willing to stay dependent on relationships that were less satisfying. That held for men and for women.

If being alone feels like the worst outcome, someone who is only half in can look like a reasonable deal. And winning over a hesitant person can feel bigger than being easily liked by an available one.

What to do this week

• Name your version. Use the picker above. Knowing whether you are reacting to retreat after closeness or to inconsistency from the start changes what you watch for.
• Track actions, not feelings. For one week, write down what the person actually did: plans made, plans kept, replies. Read it back on Sunday.
• Make one clear bid, then stop. A single direct message (“I liked this weekend. Want to get dinner Thursday?”) gives them something easy to answer. A fourth follow-up gives them something to avoid.
• Notice who feels boring. If steady, available people leave you flat, that flatness may be the absence of anxiety rather than the absence of chemistry. Try a second date anyway.
• Ask the halfway question early. “Are you looking for something real right now?” is a fine thing to ask by date three.

What doesn’t the research show?

• These are averages, not verdicts. Attachment scores describe tendencies. They do not tell you why one particular person went quiet.
• The 1994 couples study looked at heterosexual dating couples decades ago. It shows a nonrandom pairing, not that every anxious person picks avoidant partners.
• The demand-withdraw findings are mostly correlational and come largely from established couples, so the cause and effect can run both ways.
• The infidelity and fear-of-being-single studies measured tendencies in groups. They cannot tell you whether a specific date is keeping options open.
• Attachment patterns can shift over time and with different partners. Nothing here is a permanent label.

Questions to ask yourself

• When someone is easy to reach, do I feel relieved or a little bored?
• What did I notice in the first two weeks that I talked myself out of?
• When they pull back, what do I do in the next hour?
• Am I trying to be chosen, or trying to find out if I want them?
• If they replied quickly and steadily, would I still be this interested?