The Last of Us Tops the List Again — and the Criteria Column Is Still Blank

One hundred games. One name at the top. And directly beneath it, the column that matters most in any ranking, left blank once again: by what measure?

Kinda Funny Games released a list of one hundred titles. Naughty Dog's 2026 game was placed first. The report itself notes that the 2026 sequel, The Last of Us Part II, is more ambitious but also more divisive. A third name surfaced in the argument — Sekiro: Shadows Die Twice — with its backers insisting it belonged higher.

That piece landed on my desk tagged Esports. After reading the thirty information points, the first number I found was not the ranking's. It was zero — zero teams, zero players, zero tournaments, zero patches, zero prize pools.

This is games-culture coverage, not esports reporting. And it entered my pipeline under the wrong tag — that is the most decision-relevant finding in the entire file.

The scope of the list, and the argument it never settles

One hundred titles, in an ordinal sequence, running from the Pong era to a 2026 release. Which means a 2026 abstract arcade game and a 2026 cinematic narrative title were placed on a single axis. That is not comparison. That is category mixing.

The case for The Last of Us, as recorded in the report, rests on story, empathy and cinematic craft — emotional authorship and industry-historical influence. The case against rests on two clear charges: limited combat depth, and a running time that stretches further than it earns. A third camp argues Part II is the bigger project on every count and should have been placed above the original.

The line that stings most is the one where the participants themselves cannot agree on what the evaluation axis even is — gameplay, personal preference, or historical influence. A ranking that does not disclose its weighting is not a claim, it is a taste; contesting it is an argument about preference, not about fact.

I have a comparison point. In 2026, inside Bengaluru FC's three-person analytics unit, I hand-tagged 3,800 shots across two ISL seasons to build the club's first xG model. That work taught me something that applies to every ranking ever published: three thousand eight hundred shots in, the model finally opened its eyes — because by then every shot had been logged with its own yardstick attached. A list without a yardstick is the model that never opens its eyes at all.

The four places it breaks

The first break is comparability. Putting Pong and The Last of Us in one sequence requires an axis that applies to both. No such axis is declared anywhere in the report.

The second break is the identity of the artifact. The report calls it a list published by Kinda Funny Games in one place and a poll result in another. Those are different objects. A list is an editorial decision; a poll is a measured majority. The first carries an editor's authority, the second carries a denominator. The report never resolves whether the outcome was editorial or popular — and that unresolved ambiguity is exactly where a false impression of consensus gets manufactured.

The third break is attribution. Both sides of the argument are sourced to anonymous collectives — many gamers, some people, one player, another opinion. No survey size, no vote count, no engagement metric. The controversy is asserted, not measured.

The fourth break is the studio. Across the entire report the creator appears only as a collective entity, Naughty Dog. No director, no writer, no designer. Creative-lead continuity is the single most predictive variable for a studio's next release, and here it registers at zero.

The one number that holds

The only measurable thing in the whole file is also the most modest-looking and the most durable: thirteen years after the 2026 release, the title is still generating argument. That is a longevity signal, not a quality proof.

One external fact belongs here, with its confidence label attached. The only historically esports-adjacent vector for this IP was its multiplayer component — Factions in the 2026 title, and the standalone multiplayer project announced later. That project was publicly cancelled. This is not in the source report; it is my external knowledge, high confidence on the cancellation having been announced, medium on current status. The IP therefore has no active competitive footprint today, which makes any attempt to read a meta off it a category error.

I will also say plainly that I am not running a model here. Football's xG framework does not port for free, because its entire mathematics rests on repeatable competitive events — a shot, a location, an outcome, and the same situation occurring again. There are no repeated events here. What translates is not the arithmetic but the discipline: publish the yardstick before the claim.

The contrarian read: the backlash is a format effect

The easy conclusion is that people turned on The Last of Us. I do not trust that conclusion.

Ordinal lists are engineered to produce disagreement, because disagreement is the monetisable output. Asking readers to pick one out of a hundred is itself an invitation to build their own. A large share of what reads as backlash is a product of the format, not an independent reaction to the verdict. I am writing this at medium confidence, because I have no engagement data — only the structural economics of the format.

My own error goes on the record too. The bad tag reached my dashboard and I let it sit there for thirty minutes. That is not a single-item mistake; it is a labelling-layer mistake. The misclassification never touched my models. It touched my dashboard. If the Stage-1 pipeline routinely labels general gaming-culture coverage as esports, the error rate is not isolated, and a sample audit is warranted. The audit cost is worth publishing as well: four hours, one outlet's video archive, and a subscription taken solely to locate the primary source, which I never found. No methodology page was published anywhere.

The correction still needs the strongest true number beside it, or calibration hardens into contempt: sustained argument thirteen years after release is genuine brand equity, and it is the most trustworthy element in the entire piece. The debate is not fake. It is simply an argument about criteria rather than evidence — and criteria disputes cannot be resolved by supplying more data. I trust the shot map over the highlight reel, and in this ranking the shot map is missing.

What to watch next

My call goes on record in advance, with a date attached. The next time an outlet publishes a GOAT list, I expect the mechanical-depth-versus-narrative-prestige frame to resurface within twenty-four hours, exactly as it did here under Sekiro's name. I expect the methodology to stay unpublished. And I expect the distinction between a list and a poll to blur again. When the results land, I will grade myself with the same severity I apply to everyone else.

A model is a promise you renew weekly, not a chart you publish once. Rankings should be held to the same standard. The question is not who tops the next list. The question is whether the next list arrives with its yardstick attached — and if it does not, my desk will file it under culture, not esports.