Trang chủEsportsThe Last of Us Tops the 100 Greatest Games List: A Dispute Over Criteria, Not Over Data
Esports

The Last of Us Tops the 100 Greatest Games List: A Dispute Over Criteria, Not Over Data

**Câu trả lời cốt lõi**: Danh sách 100 tựa game vĩ đại nhất do Kinda Funny Games công bố đặt The Last of Us ở vị trí số một, tạo ra tranh cãi về tiêu chí đánh giá chứ không phải về dữ liệu. Bảng xếp hạng không công bố trọng số, khiến hai hệ tiêu chí — độ sâu cơ chế và ảnh hưởng lịch sử — cùng được áp lên một kết quả duy nhất. **Sự kiện chính**: - Kinda Funny Games công bố danh sách 100 tựa game vĩ đại nhất, The Last of Us đứng đầu; Naughty Dog là studio phát triển. - Tranh cãi chia ba phe: ủng hộ The Last of Us, ủng hộ The Last of Us Part II vì độ sâu cơ chế, và ủng hộ Sekiro: Shadows Die Twice. - Bảng xếp hạng không công bố rubric, trọng số hay phương pháp tính điểm. - Bài viết gốc trích dẫn các phát biểu ẩn danh "một người chơi", "nhiều người" mà không nêu số lượng, nền tảng hay khung thời gian. - Phần phụ lục có tiêu đề về LCK sau sáu năm franchising, VCT Pacific 2027, VCT Champions 2026, GAM Esports, LOUD, K/DA, MLBB Việt Nam — chỉ có tiêu đề, không có số liệu. **Nguồn**: Bài viết gốc về danh sách của Kinda Funny Games (không ghi ngày đăng; khung thời gian suy đoán 2025–2026 dựa trên tham chiếu Clair Obscur: Expedition 33 và các tiêu đề ghi năm 2026) | Cross-checked: VuaBong.vn **Hỏi đáp liên quan**: Q: Vì sao danh sách 100 game vĩ đại nhất gây tranh cãi? A: Vì bảng xếp hạng không công bố tiêu chí và trọng số, nên các bên tranh luận đang dùng những đơn vị đo khác nhau cho cùng một kết quả. Q: Bài viết này có phải tin thể thao điện tử? A: Không — chủ thể là phê bình game; chỉ phần phụ lục tiêu đề mới chạm tới thể thao điện tử và không kèm số liệu kiểm chứng. Q: Có thể dùng các tiêu đề LCK và VCT Pacific 2027 làm bằng chứng không? A: Không, cho tới khi có số liệu chính thức về lượng người xem và danh sách đối tác được Riot hoặc LCK công bố.

The familiar order of a news piece is: publish the result, then record the reaction. The article about the list of the 100 greatest games of all time, published by Kinda Funny Games, reverses that order. The reaction arrives first. The result arrives second. The reader is placed in the middle of an argument before knowing exactly what the argument is about, and by the time they learn that The Last of Us sits at number one, they have already been positioned to react rather than to evaluate.

I note this detail before any other, because it is a deliberate design choice. A ranking does not generate controversy on its own. It is packaged to generate controversy. And when a media product is packaged that way, analysing it with sporting instruments — rosters, form, metrics, head-to-head results — returns zero. I tried. I ran this article through the entire filter I use for a transfer bulletin: is there a competitive entity, an organisation, a season, a transfer window. The answer was four nos and one faint yes, located in the appendix of related headlines.

That is why this piece has to be written differently. What deserves analysis here is not whether The Last of Us merits the top spot. What deserves analysis is the mechanism by which a list with no published criteria still generates a global argument, and that mechanism is being copied verbatim into esports, where debates about the greatest of all time are equally criteria-free.

How a list becomes a standard

No body appoints a game to the canon. No federation votes. No official ranking carries legal recognition. The industry's canon — the set of works treated as benchmarks — is formed by three forces: media lists, critical consensus, and studio lineage.

Of those three, media lists are the only force that can change overnight. They cost almost nothing to produce, have a very short reading life, and derive nearly all of their economic value from the comment section beneath them. A ranking of 100 games is an engagement engine dressed as a review.

Kinda Funny Games published the list; The Last of Us came first. Naughty Dog is the studio behind the title. Those two facts are the entire body of hard data the source article supplies. No sales figures, no completion rates, no player counts, no scoring methodology, no weighting for any criterion. The number one position is a qualitative claim placed inside a fake quantitative frame — the sequence from 1 to 100 creates the impression of a measurement, but beneath the sequence there is no instrument.

This is not unusual. It is the norm of games media. And precisely because it is the norm, it deserves the table.

Methodology: three sources, two verification rounds

Before the substance, I need to state what I have and what I do not, because this piece will cite numbers and assertions with very different confidence levels.

Source one is the original article on the Kinda Funny Games list, comprising the body and the appended related-headline block. Source two is industry fact I can cross-check independently: Naughty Dog's development lineage from Uncharted to The Last of Us, the existence of The Last of Us Part II as a successor product, the existence of Sekiro: Shadows Die Twice as a critical counterweight, and Pong as the conventional starting point of industry history. Source three is the appendix, which contains headlines about LCK, VCT Pacific 2027, VCT Champions 2026, GAM Esports, LOUD, K/DA, MLBB Vietnam, and a sponsor-named event, VISA VMC Fall 2026.

The first verification round removes what cannot be verified. The entire appendix carries headlines only — no figures, no measurement window, no sourcing. None of it is used as evidence here. It is used only as a lead list, and I will say so at each point where it appears.

The second round separates the source article's assertions from my interpretation. When the source writes "many people", I am not permitted to render it as "the community". When the source writes "one player", I am not permitted to render it as "a section of players". This is discipline I learned after one very specific error, which I will recount later.

What I do not have: any sentiment measurement. No comment counts, no platform, no timeframe, no approval ratio. Every statement in this piece about "controversy" must be read as a structural description of the source article, not as a psychological market indicator.

Two criteria systems colliding inside one ranking

The structure of the argument becomes clear on a third reading. There are two criteria systems, and they cannot coexist inside a single ranking.

The first is the systems system. It measures control quality, depth of adversarial design, frame precision, freedom in solving the combat problem. Defenders of The Last of Us Part II stand here. They say the second part plays better than the first, that combat mechanics are denser, that approach space is wider, that opponent AI is more complex. This is an argument verifiable by playing, and it is the argument I have methodological sympathy for.

The second is the historical-influence system. It measures a work's effect on what was produced afterwards, on how a generation of developers thought about storytelling in games, on the work's position in the industry's development current. Defenders of the top spot stand here. They say the work cannot be omitted from any history of the industry. This too is verifiable, but by an entirely different method: citation, genealogy, influence analysis.

The two systems do not conflict over facts. They conflict over units. One measures by mechanical depth, one by radius of influence. Placing two different units onto a single ranking without publishing a conversion is the surest way to produce an argument that cannot end.

The pivot is here: the argument over a 100-greatest-games list is not an argument about data, it is an argument about an unpublished unit of measurement. Inside a ranking with no rubric, two sides are arguing about two different things and neither side knows it.

In esports we do this daily. When arguing over the greatest player in a title's history, one camp counts international titles, one counts longevity at the top, one counts influence on the meta, one counts commercial value. Four units, four results, no published weights.

The safe-and-ambitious dichotomy

The most analytically valuable pairing in the source article is the one used to compare The Last of Us with The Last of Us Part II.

Part one is described as a safe work: a linear, tightly controlled story told in a proven cinematic language, with no structural risk. Part two is described as an ambitious work: more fragmented structure, more uncomfortable moral choices, willing to trade player comfort for depth.

I have seen this pairing hundreds of times in sport, only with different labels. One team picks a proven style and wins. One team picks an untried style and loses. Analysts then argue over who was actually better, and the argument always deadlocks because the two sides are measuring different things: one measures outcome, one measures the difficulty of the path to the outcome.

What gives this pairing lasting value is that it does not depend on any particular list. It will outlive this one. And it transfers directly into my industry: the tension between a roster built to maximise title probability and a roster built to open a new school of play.

One limit must be stated. The source describes this pairing through anonymous player voices. I have no data to confirm that this is how the community actually divides. I can confirm only that the pairing is a real analytical template, present in this industry long before The Last of Us existed.

The Last of Us Tops the 100 Greatest Games List: A Dispute Over Criteria, Not Over Data

Naughty Dog lineage and the illusion of a continuous current

The source draws industry history as a straight line: from Pong to Clair Obscur: Expedition 33, with Uncharted leading into The Last of Us as an internal link within one studio.

That line has a technical problem. It uses the lineage of a single studio as the axis of an entire industry's history. Uncharted and The Last of Us were both developed by Naughty Dog. When a work is placed at the apex of a line, and two of that line's most important links come from the same producer, the line was already bent before the comparison began.

I have met this exact structure in transfer data. One academy produced three world-class players in four years. Media wrote about that academy's "talent pipeline". But when I opened the data, the pipeline had a single conduit: one coach, one scouting philosophy, one scout network. It was a replicated process, not an ecosystem. And it collapsed within eighteen months of the coach leaving.

The same misreading is happening here. One studio doing well with two franchises in one decade does not prove the industry is evolving in that direction. It proves one studio found a formula and exploited it efficiently. Those are different claims, and the source presents the second as though it were the first.

The systems camp and the name Sekiro

One small but important detail appears in the source: a section of players argue that Sekiro: Shadows Die Twice deserves the top spot more.

Sekiro's presence is not accidental. It represents the opposite pole of the systems criteria: almost no cinematic storytelling, almost no long cutscenes, nearly all value concentrated in precision of adversarial mechanics and the rigour of learning opponent patterns. If The Last of Us wins on influence and narrative, Sekiro wins on mechanics and skill.

This is where I must be most careful, because this is where an analyst slides into the prophet's chair. I have no data to say which side is right. No survey, no sample, no measurement of where the community leans. What I have is a structural observation: the existence of a third camp backing Sekiro proves the argument is not bipolar. It is a triangle, and each vertex measures in a different unit.

In sports analysis, such a triangle usually signals an undefined concept. When three fan groups argue about one player and none persuades another, the cause is almost always that the three groups hold three different definitions of the word "good". One measures goals. One measures assists. One measures advanced metrics. All three are right inside their own system, and all three are meaningless inside the others'.

The phrase "even critics concede"

One sentence in the source deserves separate scrutiny: the suggestion that even those who disagree with the top spot concede that The Last of Us is one of the finest games in history.

That sentence has a very specific rhetorical function. It is not a fact. It is a consensus-manufacturing device. It works like this: the writer concedes a weaker claim in order to create the impression that the stronger claim has also been accepted. The dissenter is described as having conceded "one of the finest games", and from there the reader slides into a feeling that they have also conceded "the finest game". The two propositions differ completely in logic. The first is nearly impossible to deny. The second is the entire content of the argument.

I recognise the technique because I have used it, and because it has fooled me. In a 2026 transfer report I wrote that even sceptics conceded the target had potential. That sentence was true. And it concealed the fact that the whole dispute sat inside the word "potential" — everyone agrees a young player has potential, and no one agrees on what that potential is worth. I used an undeniable claim to close an argument about value. That was a professional error, and I recognised it more than two years late.

From Pong to Clair Obscur: Expedition 33

The source uses the phrase "from simple products such as Pong to modern games such as Clair Obscur: Expedition 33" to describe the length of industry history.

The starting marker is reasonable by convention. Pong is a widely agreed starting point in most tellings of industry history, even though it is itself only a Western cultural convention.

The end marker is problematic. Clair Obscur: Expedition 33 is a very recent work. Using it as the terminus of a five-decade history performs two effects at once. First, it places a work that has had no time to settle as the endpoint of a current. Second, it converts a very narrow frame into what looks like a panoramic one.

This is recency bias, and it has an exact replica in esports. Whenever a team wins two events in a row, media begins reconstructing the discipline's history with that team at the end of the line. Metrics are selected so the line looks continuous. Previously dominant teams are compressed into a short chapter. I have done this in many reports and I know exactly how easy it is: you simply choose the right starting marker.

One further point is worth recording. Using a recent work as the endpoint allows the article's approximate time frame to be inferred — roughly 2026 to 2026 — which matches the 2026-dated headlines in the appendix. I raise this because it says something about editorial quality: an article with no publication date still leaves a time trace, and the trace sits where the writer did not intend to leave it.

Voices without a count

This is the section I must give the most words to, because it is the largest methodological weakness in the entire source article.

The article uses constructions like "one gamer", "some people", "many people", "one player". No platform is named. No quantity is named. No timeframe is named. No sampling method is named.

In data analysis this is the most serious of violations, because it allows an unmeasured statement to carry the shape of a measured one. "Many people" could be twelve. "Some people" could be three. "One player" is one. And when the three constructions appear beside each other in the same article, the reader unconsciously merges them into a community sentiment block with weight.

I have made this error at a larger scale. In 2026 I built an improved xG model to predict Ulsan Hyundai against Jeonbuk. The model returned 2-0. The match ended 1-3. I spent three weeks auditing the entire data pipeline and found an encoding error in the key-passes variable that skewed the weights. The error was not in the algorithm. It was in my trusting a variable whose provenance I had never checked. K League 2026 taught me that pioneers do not fail because they look far; they fail because they look far while miscounting one column of data.

The source article is missing exactly one such column. That column is the count column.

Blind spot one: the misapplied "Esports" label

Here I must say something about the classification process that delivered this article to me.

The source was tagged as esports. I read all twenty-four of its leading information points. No competitive title. No tournament. No roster. No patch. No esports organisation as subject. The subject is a games-criticism list published by a games media outlet, and the argument around it.

This is a confirmed classification error, and it is more dangerous than a single mistake. A mislabelled record entering an aggregation pipeline distorts entity counts, topic distributions, and sentiment baselines for the entire domain. If ten such articles enter an esports dataset, that dataset will report rising esports coverage when in fact games-criticism coverage has been mislabelled.

I raise this not to assign blame. I raise it because it is a perfect illustration of the principle I use in cross-checking: a label is not a fact. A label is a hypothesis about a fact, and every hypothesis must be tested before it enters the model.

The only element in the source that touches genuine esports is the related-headline block at the end. It contains K/DA, LCK, VCT Pacific 2027, VCT Champions 2026, GAM Esports, LOUD, MLBB Vietnam, and one sponsor-named event. All headlines only. I treat them as a lead list, not as evidence.

Blind spot two: correlation is not causation

The appendix contains one notable headline: LCK doubled its viewership after six years of franchising.

Read quickly, this is evidence for franchising. Read slowly, it is an unestablished causal chain.

Three problems. First, no figure is given — doubled from what, measured in what unit, over what window. Second, no baseline is given — if viewership had fallen deeply before franchising began, doubling from a trough says little. Third, and most important, at least two other variables moved during the same period: platform migration and the arrival of co-streaming formats.

This is the error I call assigning causality to the nearest variable. When two things happen at once, the one that arrives later is designated the cause. I have seen it in home-advantage analysis, in seasonal-psychology analysis, and in almost every report on crowd effects on performance.

I once thought I was reading a match map; in fact I was looking at a mirror reflecting my own fear.

That fear, in this case, is the fear that the franchising model might be a mistake. When a new model is adopted and viewership rises, people want to believe the model saved the discipline. But a model can only be assessed against a controlled comparison, and that comparison requires a region that did not adopt it. In LCK's case, no such control exists.

What actually transfers to esports

After removing everything unverifiable, three things of genuine transfer value remain.

The first is the mechanism of canon formation. A list with no rubric still shapes what the industry treats as highest quality. When a cinematic narrative work tops a global list, it sends an investment signal to studios: invest in production value and story presentation. That signal is weak, not immediate, and unmeasurable. But it exists, and it propagates over multi-year cycles.

In esports we have an equivalent mechanism that operates far faster. Whenever a team wins a title with a specific style, that style becomes the selection standard for the next two or three transfer windows. Academies begin searching for player archetypes that fit it. Other archetypes are undervalued. This is an effect that is measurable given a long enough selection dataset, and it is one of the things I track most consistently.

The second is the transparency-of-criteria problem. The source records the community's question verbatim: does this list evaluate by gameplay quality, by popularity, or by historical influence. That question is not answered. And because it is not answered, the argument has room to exist.

This is the point I want to stress most in this entire piece. A ranking that publishes its rubric largely cancels its own controversy. Criteria ambiguity is not a defect to be fixed; it is the fuel of the business model.

In esports we are doing precisely this at industrial scale. Greatest-player rankings, greatest-roster rankings, most-prestigious-event rankings — almost none publish weights. And almost all generate traffic. It is a perfectly functioning model that produces zero information gain.

The third is factional structure. The source reveals three camps: loyalists to part one, advocates of part two on mechanical depth, and advocates of an entirely different title. These cannot be reconciled with data, because they measure in three units. In esports the structure appears every time a player retires: the nostalgia camp, the metrics camp, the trophy camp.

Every transfer is a murder case. The culprit is expectation; the weapon is timing.

The same logic applies to greatest-of-all-time arguments. The culprit is the unpublished criterion; the weapon is a list published at the moment of peak traffic.

LCK, VCT Pacific 2027, and miscounting a column

I want to pause on three appendix headlines, because they are the most trackable leads even though they are not evidence.

The first is LCK after six years of franchising. This is an answerable question given the right data. Three things are needed: year-by-year viewership, platform allocation ratios, and a separation of real growth from platform migration. Without those three, every franchising conclusion is storytelling. I will draw no conclusion until at least two of the three exist.

The second is VCT Pacific 2027 and GAM Esports' appearance as a partner. If accurate, this is a notable organisational signal: an organisation rooted in a MOBA title moving into an FPS circuit. Such cross-discipline moves usually fail for three predictable reasons: differences in coaching structure, differences in patch cycles, and differences in how opponents are read. But I have no official confirmation, so this is a tracking item, not a conclusion.

The third is MLBB Vietnam and regional competitive potential. The word "potential" in a headline is a signal to read slowly. Potential is not achievement. A mobile discipline only genuinely shifts regional balance when it produces sustained top-four finishes at regional level, not when it produces a few handsome wins.

All three headlines share one property: they lack figures, and all three could become a good analysis given figures. That gap is where I work.

The market does not move on news. It moves in the gap between two reports.

K/DA and VISA VMC Fall 2026: faint commercial signals

The remaining two appendix headlines concern a different thing: money flow.

The Last of Us Tops the 100 Greatest Games List: A Dispute Over Criteria, Not Over Data

K/DA appears in connection with controversy over a new appearance. This is a music brand attached to a game, and one of the most successful examples of using a music product as a marketing instrument for a competitive ecosystem. If the appearance controversy escalates into a brand-level or talent-level dispute, that would be a risk signal for the whole model. But I have only a headline, no figures, no quantifiable response. I log it and draw no conclusion.

VISA VMC Fall 2026 is a headline with a sponsor name attached directly to an event name. This is a common commercial structure: a brand buys naming rights, the event receives money, the system receives legitimacy. No terms disclosed, no value disclosed. Nothing can be assessed from a name.

What I can say is the overall structure of the appendix block. It runs from music, to a national league, to a regional circuit, to mobile esports, to a sponsored event. That is the portrait of a newsroom broadening its topical surface toward esports for traffic reasons. I have no evidence to say whether this is headline aggregation or original reporting. But the structure is consistent with the former.

For my industry this remains a weak positive signal. When general games media maintains a standing esports vertical, it creates an audience funnel from general gamers to esports viewers. That funnel is real, in principle measurable, and no one has measured it credibly.

The applause in an empty stand is not noise; it is a signal from a future we have not yet dared to index.

What would change my mind

An analyst is only as credible as the conditions under which he declares he would change his mind. I name three.

Condition one: if Kinda Funny Games publishes a weighted rubric, and that rubric places historical influence above mechanical depth, then this argument contracts into a technical dispute over whether the weights are reasonable. That is a far better argument than the current one.

Condition two: if a genuine sentiment measurement appears across platforms, with sample size, timeframe and sampling method disclosed, I will revise my assessment of the argument's scale. For now I can speak to structure, not scale.

Condition three: if LCK or Riot publishes multi-year viewership data with platform effects separated out, I will revisit the entire franchising argument. Until then, assertions in either direction are storytelling.

All three conditions share one property: they require something this industry is reluctant to publish. The method.

Takeaway

A list of 100 games teaches us nothing about The Last of Us. It teaches us something about how an industry evaluates itself, and about what it chooses not to publish.

Over the next three months I will track one variable: whether any esports outlet publishes a greatest-of-all-time ranking with a rubric attached. If one does, and if that ranking still generates comparable traffic, my hypothesis — that criteria ambiguity is a mandatory fuel of the model — is falsified. That would be good news for the industry and bad news for my model.

The Last of Us Tops the 100 Greatest Games List: A Dispute Over Criteria, Not Over Data

If no one does, we already have the answer without needing more data. And that answer is not located at the top of any list. It sits in the blank beneath the number, where a criteria table should be, and where there is only a silence designed for the reader to fill with their own feelings.

Cầu thủ liên quan