The Dispatch

Continued from Page One.

MOOD's answer is radical in its simplicity: remove the opaque matching algorithm entirely. There is no black-box system deciding who you get to see based on a hidden engagement objective. Instead, you declare your intent — your MOOD — and you see the people near you who share it. The "matching" is transparent and user-controlled: it's your stated intent plus proximity, not a secret model optimizing metrics you can't see against interests that aren't yours. This is, admittedly, less magical-sounding than "our AI finds your soulmate." But it's honest, and it can't be secretly turned against you, because there's no hidden objective for it to serve. The most trustworthy algorithm, it turns out, is the one you can fully see — or better yet, the one you don't need, because you told the system what you wanted directly.

LAFAYETTE — Many dating apps present a compelling number: a compatibility percentage, a match score, a data-driven prediction of how well you and a stranger will get along. It looks scientific. It feels reassuring. And the uncomfortable truth, which this bureau has chased through the actual research, is that these numbers have very little scientific basis. The compatibility score is, for the most part, decorative — a reassuring figure dressed in the costume of prediction.

The research here is fairly damning and worth knowing. Studies examining whether algorithms can predict romantic compatibility from the kind of data profiles contain — stated preferences, questionnaire answers, demographic traits — have repeatedly found that they can't, at least not to any meaningful degree. The most cited work on the question concluded that romantic chemistry between two specific people is essentially not predictable from the information available before they actually meet. The spark lives in the interaction itself — in chemistry, timing, and countless subtle in-person signals — none of which fits in a profile, and none of which an algorithm can see. A "94% match" is not a measurement. It's a number generated to make the app feel smart and to make you feel hopeful enough to keep swiping.

Why does the fiction persist? Because it works as marketing. A confident percentage makes the app seem sophisticated, gives users a reason to trust the system, and manufactures the little hit of hope that drives the next action. It's a magic trick, and like all good magic tricks, it works precisely because the audience wants to believe there's a science that can spare them the uncertainty of actually meeting someone. But there isn't. There's just two people, and the only real test.

MOOD declines to run the magic trick. There is no compatibility percentage, because this paper's parent company isn't willing to sell a number it knows is decorative. What MOOD offers instead is honest about its own limits: shared intent (you both want the same kind of thing, which is real and knowable) plus a fast path to the real, in-person meeting (which is the only actual compatibility test there is). It's less flattering than "our algorithm found your 94% match" — but it doesn't lie to you about a science that doesn't exist. The compatibility score promises to predict the spark. MOOD admits the spark can only be discovered, in person, by you — and then gets you there as fast as it can. That's not a downgrade from the magic number. It's the truth the magic number was hiding.

LOS ANGELES — Modern dating apps offer an ever-expanding menu of filters — height, age to the exact year, distance to the mile, body type, education, a dozen dealbreakers you can toggle — and they're marketed as tools for enforcing your standards. Your correspondent wants to raise an uncomfortable possibility: that somewhere in that filter menu, you may have quietly, permanently, and unknowingly deleted the person you'd have been happiest with. The filter that felt like discernment was, sometimes, just a wall you built between yourself and your future.

Consider the height filter, the most notorious example. A person sets "6'0" and up" and, with one toggle, makes invisible every wonderful, compatible, could-have-been-the-one human who happens to measure 5'11". Not because height genuinely mattered to their happiness — almost no one, asked honestly, believes an inch determines a life — but because the filter let them encode a lazy preference as a hard rule, and the app dutifully enforced it, deleting hundreds of people sight unseen. Multiply that across every filter — the age cutoff that erased someone perfect a year too old, the distance limit that hid someone across an arbitrary line — and you get a search space quietly gutted of its best possibilities by criteria that felt reasonable one toggle at a time and were, collectively, a disaster.

Here's the deeper problem: filters operate on the legible traits, and compatibility lives mostly in the illegible ones. You can filter for height; you cannot filter for the way someone makes you laugh, their kindness, the particular ease you feel around them — the things that actually determine whether a relationship works. So the filter menu optimizes hard for the traits that don't matter much while being structurally blind to the traits that matter most, and in doing so it deletes people who'd have scored low on the legible criteria and infinitely high on the illegible ones. It's precision applied to exactly the wrong variables.

MOOD deliberately de-emphasizes the endless filter menu, and this is a design choice with a clear rationale, not an oversight. The organizing principles are shared intent (which genuinely matters) and proximity (which is practical), not a sprawling toggle-wall of superficial dealbreakers that delete people before they ever get a real chance. The bet is that you're far more likely to find happiness by meeting a broader range of intent-aligned people near you — and discovering the illegible compatibility in person — than by pre-filtering the pool down to a narrow spec sheet that screens for the wrong things. Your standards are real and worth honoring. But a filter is a blunt instrument, and the bluntest ones have a habit of cutting out exactly the person you were looking for. Sometimes the most romantic thing an app can do is show you someone you'd have filtered out.

PORTUGAL — Having spent this edition dismantling what the algorithms actually do, I want to close with the constructive argument, because tearing down is easy and building is the job. The anti-algorithm case is not anti-technology — I am, after all, an engineer who builds the stuff. It is an argument about where the machine belongs: the claim that dating technology works best when the machine handles the logistics and then gets out of the way, leaving the actual romantic decisions to the humans, where they belong.

Consider the two things a machine can do in dating, which are very different. It can handle logistics — helping you find people near you, organizing plans, reducing the friction of coordination, surfacing who shares your intent. And it can attempt judgment — deciding who you should want, scoring compatibility, optimizing your love life against some objective. The machine is genuinely excellent at the first and genuinely terrible at the second, and the entire dysfunction of the dating-app era comes from letting it do the second while pretending it was doing you a favor. We handed the machine the one job it can't do — judging human compatibility — and let it optimize that job against its own interests instead of ours.

The anti-algorithm position simply puts the machine back in its lane. Let it do the logistics it's good at: connect you to intent-aligned people nearby, make the plan easy, remove the friction. And then let it stop — return the judgment to you, the human, who is the only entity actually equipped to know whether you like someone. The machine as a tireless, transparent assistant handling coordination; the human as the sole authority on the romance itself. Less machine, more human — not by rejecting the tool, but by confining it to what tools are for.

This is, precisely, the philosophy MOOD is built on. The structure is transparent — declared intent plus proximity, no black box you can't see into. There is no opaque matchmaker deciding who's worthy of your attention against hidden metrics; the human judgment is restored to the center, where it was always supposed to live. The machine handles what machines are good at — finding intent-aligned people near you, making the Offer easy, getting you to the meeting — and then it gets out of the way and lets you be the one who decides. That's the whole anti-algorithm case: not a Luddite retreat from technology, but a correction of its role. The machine was never supposed to be the matchmaker. It was supposed to be the tool that hands the matchmaking back to you. Build it that way, and technology stops being the thing that came between people and goes back to being the thing that brings them together — which was, after all, the entire point.