Dating Dealbreakers: Why Your List Is Scoring the Wrong Things

Your dating dealbreakers are almost certainly measuring the wrong things — and there is a large, unglamorous body of research explaining exactly why. The list in your head, or in your notes app, is a list of traits. Traits turn out to be poor predictors of whether you will actually be happy with someone.

The short version: studies consistently find that what people say they want in a partner does not predict who they are drawn to in person, and that a partner’s individual characteristics predict relationship satisfaction far less well than the dynamic that develops between two specific people. A dealbreaker list is a filter built for a screening stage. The thing it is screening for is not the thing that determines the outcome.

The list gets you to the table. It cannot tell you what happens at the table. Photo: Priscilla Du Preez / Unsplash.

Do people actually know what they want in a partner?

In 2008, psychologists Paul Eastwick and Eli Finkel ran a study that should be more famous than it is. Before a speed-dating event, participants stated how much they valued specific qualities in an ideal partner. Then they met a series of real people, face to face, and reported who they were actually interested in.

The stated preferences did not predict the in-person attraction. People’s carefully articulated ideals failed to forecast what actually inspired their desire once a human being was sitting across from them. Eastwick and Finkel’s conclusion was blunt: individuals appear to lack reliable introspective access to the preferences they will exhibit when they meet someone.

This is not a claim that preferences are meaningless. It is a claim about a specific and very common error — treating a list assembled in the abstract as a reliable instrument for evaluating a particular person in the concrete.

Why traits predict so little

The strongest evidence here comes from a 2020 study in the Proceedings of the National Academy of Sciences led by Samantha Joel, with Paul Eastwick and dozens of collaborators. The team applied machine learning to 43 longitudinal datasets covering more than 11,000 couples, asking a simple question: what actually predicts relationship quality?

The answer reorders most people’s priorities. The strongest predictors were relationship-specific: perceived partner commitment, appreciation, sexual satisfaction, perceived partner satisfaction, and how conflict went. Individual characteristics mattered too, but the ones that mattered were largely about the person doing the rating — their life satisfaction, negative affect, depression, attachment style — not a catalogue of their partner’s attributes.

Put plainly: how committed your partner seems to you, and whether you feel appreciated, tells you far more about your future satisfaction than their job title, their height, their politics about pets, or whether they were on time to the second date.

The difference between a dealbreaker and a preference

Most lists collapse two very different categories, and the collapse is what makes them useless.

A real dealbreaker is structural. It describes a condition under which the relationship cannot function no matter how well the two of you get along: incompatible positions on having children, incompatible geography with no willingness to move, active untreated addiction, a pattern of dishonesty, contempt for you or for people you love. These are not preferences. They are load-bearing, and no amount of chemistry renovates them.

A preference is everything else — and the honest test is whether you would still care about it if the person were otherwise extraordinary. Height, industry, taste in music, how they spell, whether they read fiction. These feel like standards. They function like noise, and each one you enforce absolutely narrows the pool without improving the odds.

The practical move is not to abandon standards. It is to demote roughly eighty percent of them from dealbreaker to preference, and promote a small number of behavioral things you were probably not tracking at all.

Most lists are built in solitude and tested in company. Photo: Matt W Newman / Unsplash.

What to screen for instead of dating dealbreakers: behaviors, not attributes

If traits predict poorly and relational dynamics predict well, the obvious implication is to stop evaluating the person and start evaluating what happens between you. That is observable much earlier than most people think — usually by date three or four, and often on date one.

How do they handle being slightly disappointed?

Not a crisis. A wrong order, a cancelled reservation, a train that does not come. You are watching for the reflex underneath the manners. Someone who is charming to you and curt to a server has simply shown you their conflict behavior in advance.

Can they be told something small and take it in?

Try one low-stakes piece of feedback and watch. Do they get curious, or do they explain? Perpetual explaining is a preview of every disagreement you will ever have, and it is a far better predictor than anything on a profile.

Do you feel like a more or less generous version of yourself around them?

This is the most underrated signal in dating and the hardest to fake. Some people make you subtly meaner, more anxious, more performative. Others make you funnier and more relaxed. That effect is relationship-specific, it shows up fast, and it is precisely the category of variable the Joel study found to be most predictive.

Do they follow up on things you said in passing?

Not grand gestures — retention. Remembering that you had a hard meeting on Thursday and asking about it. This is what perceived partner commitment and appreciation look like at the earliest possible stage, and those two variables sat at the very top of the predictor list.

Why chemistry keeps overriding your list anyway

There is a reason people abandon their criteria the moment they meet someone compelling, and it is not weakness. It is that the abstract list and the live encounter are processed by different systems, and the live one wins because it has access to information the list never had: timing, humor, pace, how a specific silence feels.

The failure mode is not that chemistry overrides the list. It is that people then use chemistry to override the structural dealbreakers too — the small number of items that genuinely were load-bearing. Keep three or four things absolutely non-negotiable, and let chemistry have jurisdiction over everything else. The problem with most lists is not that they are too strict. It is that they are strict in the wrong places and permissive in the ones that matter.

Frequently asked questions about dating dealbreakers

How many dealbreakers should I actually have?

Three to five, and each one should describe a structural incompatibility rather than a taste. If you cannot explain why a given item would sink an otherwise excellent relationship, it belongs on the preferences list. Long dealbreaker lists tend to function as anxiety management rather than as filters.

Is it a red flag if someone does not meet my list at all?

Not by itself. The research suggests your list was assembled with limited access to your own actual preferences, so a mismatch is at least as likely to be a flaw in the list as a flaw in the person. What matters more is whether the relational signals are good — do you feel appreciated, do they show up consistently, does conflict go somewhere.

How soon can I tell whether someone is right for me?

You can identify structural dealbreakers quickly, often within two or three conversations, and you should ask directly rather than hoping. The relational quality that actually predicts satisfaction takes longer — typically a few months of ordinary life, including at least one genuine disagreement, because how you repair is more informative than how you get along.

Should I compromise on my standards?

Reframe rather than compromise. Holding firm on a small number of structural things while becoming genuinely flexible about surface traits is not lowering your standards — it is aiming them at the variables that determine outcomes. That is a raise in standards, not a cut.

The one-line version

Stop auditioning people against a specification and start noticing what the two of you produce together. The list tells you who to meet. Only the dynamic tells you who to keep.

For more on reading early signals accurately, see our pieces on green flags in dating, attachment styles in dating, and questions to ask before getting engaged — or browse the full Dating and Engaged section.

This article offers general information, not clinical or legal advice. If a relationship involves coercion, intimidation, or fear for your safety, that is not a preference question — please reach out to a qualified professional or a local support service.

Key takeaways on dating dealbreakers

• Stated preferences predict attraction poorly once you meet someone in person.
• Useful dating dealbreakers describe behaviour — how someone handles conflict, repair, and commitment — not attributes like height or job title.
• Keep the list short: a handful of genuine dealbreakers beats a long scorecard.

For research on the behaviours that actually predict relationship stability, see The Gottman Institute.