Faker and Oner Are Not Declining Together — Three Holes in the Numbers That Are Telling T1's Story Wrong

I sat in my studio in Seoul at two in the morning, eyes fixed on a playoff replay, and the first thing that struck me was not a misplayed mechanic. It was a gap. Oner stood in a bush above mid lane, waiting for a gank that never came. Fifteen seconds. Twenty seconds. At twenty-three seconds, he turned and walked away. One instance proves nothing. But when I pulled the full game back, I counted four similar moments. Four times T1's jungler waited, no one came, and he withdrew like a man standing outside a meeting he should have been chairing.

The post-game numbers confirm this in their own language: Oner was involved in roughly five-sixths of combat metrics, near the bottom of an eight-team pool in kill participation, damage contribution and gold difference. Over the same window, Faker carried similar rankings across several columns — in some, near last. Two veterans at once. And the moment those two names appear in the lower half of a table, a default narrative fires: T1 is in crisis before Worlds.

That is the conclusion the forums are running with. And it is a conclusion I believe to be wrong — or at least built on ground that can collapse at any moment. Not because I blindly trust T1. Because the structure of evidence being used to reach this conclusion has three holes no one wants to name. Stacked together, those three holes turn an entirely normal dip for a veteran team into a manufactured tragedy — what the media needs rather than what analysis needs.

This piece is not here to defend T1. It is here to dissect how we read numbers.

Context: what we are actually talking about when we talk about a dip

Before the holes, let me draw the context correctly. This is the 2026 season. Patches have changed gameplay in many ways — anyone following professional League of Legends knows this. But one thing must be said plainly: when a commentary piece says "the game changed after patches" without naming a single patch, a single champion, a single item, a single win-rate figure, that sentence is not analysis. It is a frame. A frame used to hang any phenomenon that needs explaining.

I have followed the LCK long enough to know the jungle role remains central to map control in the current window. The jungler coordinates with support and mid to pressure side lanes. If that is true — and by my own viewing experience it is true at a general level — then Oner sits directly on the critical path of the map. A jungler with low metrics is not a weak player. He is a system hole.

But here is the point no one stops for: if the current meta truly favors jungler-driven tempo, Oner's low metrics are more damaging than they would be in a passive-farm meta. Meaning the meta structure itself is amplifying his number. A jungler in a passive-farm meta with low fight participation is fine. A jungler in a tempo meta with low fight participation is a disaster. Same number, two meanings. And the commentary I am reading blended those two meanings together.

This is not academic. It has direct consequences for how fans treat players. When you read a number without knowing which frame it belongs to, you are not reading data. You are reading your own emotion repackaged as numbers.

The core: three holes in the numbers

Hole one — the sample is too small to be a trend

This is the most serious error and the least discussed. The metrics being cited come from a six-team playoff, later expanded to eight teams in the statistical sample. Read that again. Six teams. Eight teams. In a tournament where each team plays only a handful of playoff games, what does it mean for a player to rank fifth of six? It means he is below four and above one. The gap between third and sixth in a six-team sample is usually within the band of one good or one bad series.

I once wrote a piece on an asymmetric left-side defense when I was nineteen, and I remember the desperation for evidence when I had only one half of football to lean on. I was right, but I was right by luck more than by system. If that match had unfolded differently by ten minutes, I would have been a fraud. That is the nature of small samples. They do not give you truth. They give you a chance.

When an article uses a six-to-eight-team sample to conclude two veteran players are "declining," it is doing exactly what I did at nineteen but without the excuse. It is turning variance into a trend. Variance is always present; a trend must be proven.

The damage contribution figure is the clearest example. In a small sample, damage output is dominated by a single factor: how many teamfights your team fought that game. A fast loss keeps everyone's damage low. A dominant win keeps everyone's damage high. That is not a signal about individual skill. It is a signal about match outcome split across five people.

If you want to judge a jungler's damage, you must normalize it by game duration and total fight count. No one does that in the narrative currently spreading. They just take the raw number and compare.

Hole two — same position but different tactical role

The interesting part is that the commentary I read claims it compares against "same-position players." That is a better method than comparing a jungler to a mid laner. I acknowledge it. But same position does not guarantee same role.

Two junglers can play two entirely different styles. One is an objective-control jungler, taking dragons and heralds, ganking little but holding tempo. Another is a pressure jungler, ganking constantly, accepting farm loss to create breakouts. These two share a position but their kill participation numbers will diverge widely. Put them side by side on one table and you are comparing apples to oranges and calling it science.

In T1's case, Oner's role is clearly defined: he is the link between map control and Faker's mid-lane strategy. If the team plays slow, he fights less. If the team plays fast, he fights more. His number depends on the team's tempo far more than on himself.

And here is what rankings never display: team tempo. A jungler on a slow team will always look worse than a jungler on a fast team, even at equal skill. That is a systemic flaw of the comparison method, not a player flaw.

With Faker, the error is subtler. Mid lane is the position most dependent on the jungler in the early game. If the jungler creates no pressure, mid is squeezed and the mid laner's numbers follow downward. In other words, these two numbers are not independent. They may be measuring the same thing: the quality of mid-jungle coordination in the early game.

When you read that Faker and Oner declined together, you may be reading a single phenomenon counted twice. That is a basic counting error and it changes the conclusion entirely. If it is two independent declines, the problem is individual. If it is one coordination problem double-counted, the problem is systemic — and systems can be fixed by practice, not by replacing people.

Hole three — gold difference does not measure skill, it measures map flow

This is the hole I see least discussed and the most important for the jungler role. Gold difference sounds like a measure of a player's earning ability. It is not. Gold difference measures how many resources a player received from the team's map system.

A jungler only out-earns his opponent when his team controls resource-rich areas. If his team loses control of both side lanes, he is pushed into low-resource zones and his gold difference goes negative. That is not a sign he played badly. It is a sign the map did not belong to his team.

In a context where the jungle role remains central to map control, as I argued earlier, a jungler's gold difference becomes a derived metric rather than a root metric. It does not tell you what Oner did. It tells you what percentage of the map T1 controlled in that window.

And when you turn a derived metric into a root metric, you invert cause and effect. You say the jungler is low-gold because he played badly, when the truth may be the reverse: the jungler played on a low-resource map because his team was losing control.

The contrarian angle: where I might be wrong

Now the part where I interrogate myself, because it is the part sports writers skip most.

First, I might be wrong: perhaps the meta truly shifted so that the jungler matters less, and Oner's low metrics are a real signal that he failed to adapt. I have no pick/ban or game-duration data for the current patch. If someone handed me a dataset showing jungle priority dropped across the league, I would revisit my entire argument. I say this seriously, not to hedge.

Second, I might be wrong: perhaps these two veterans are genuinely fatigued. Career length in esports is harsher than in most sports. A player with ten years at the top has spent a portion of body and mind that no salary can repay. I have no wrist-injury data, no scrim-schedule data, no sleep data. If someone told me both are running dry, I would have no basis to rebut. And that is slightly more worrying than a low metric in a six-team sample.

Third, I might be wrong: perhaps I am so focused on methodology errors that I forget sometimes a methodology error and a real trend coincide. An addict of correction must remember not every bad number is a wrong number. Sometimes a bad number is a right number badly presented. If that is the case — if T1 truly is weakening in the early game in a way measurable on a larger sample — then my defense of poor data becomes exactly what I hate most in others.

That is why I do not conclude T1 is fine. I only conclude the current data is insufficient to conclude T1 is not fine.

The data section: what can actually be measured

If I had to give a set of criteria to distinguish a normal dip from a real decline, here is what I would track.

First, kill participation normalized by game duration. If Oner's number is low even in long games, where tempo has time to settle, that is a real signal. If it is low only in short games, where all numbers compress, that is noise.

Second, fight participation split by phase. A jungler may have low numbers in the first ten minutes but high in the middle ten. If that pattern repeats across games, it says he changes his approach over the course of a match rather than declining. That is a completely different distinction.

Third, gold difference split by lane. If Oner is low-gold because T1's side lanes are low-gold, the problem sits in the side lanes. If he is low-gold while the side lanes still control their zones, the problem sits in his pathing.

Fourth, and most important, a larger sample. Not six teams. Not eight. I want the whole regular season, where each player has played dozens of games. That is the minimum for a trend to be called a trend.

I say this because I have been the person misjudged by a small sample. At twenty, I mispronounced a midfielder's name three times in a semifinal. Three times. Listeners called in to complain. If someone judged my career on those thirty minutes, they would conclude I should not be in a studio. But I spent the next thirty days rewatching every match, recording my pronunciation of every name, and learning pacing to give my commentary rhythm. By the final, I read one hundred percent correctly and was praised by the very people who had complained.

The lesson is not "give people a second chance." The lesson is: a small sample speaks to a moment, not to a person. And an article using a small sample to speak about a person is doing something other than journalism.

The broader context: why this story is so seductive

There is a structural reason the "T1 in crisis before Worlds" story sells. It matches a real historical motif: T1 has repeatedly underperformed domestically and then exploded at Worlds. It happened. It is not a myth.

But a real motif can still be abused. When a team has a history of Worlds rebounds, every mid-season dip can be assigned to the same script. The result is a self-protecting structure: if T1 wins at Worlds, the story is confirmed; if T1 loses, the story still holds because they underperformed all along. No outcome can falsify the hypothesis. And a hypothesis that cannot be falsified is not analysis. It is belief.

I am not saying T1 will not rebound. I am saying their rebound is being presented as historical fact when it is actually an unverified hypothesis for this specific season.

There is a second layer. The story of a Korean national team in trouble, of a legend aging, of a jungler criticized — these stories carry far more emotional pull than a metrics-only analysis. They touch the fear of loss. And fear of loss is the strongest sharing driver on social media.

I understand that because I do storytelling. At twenty-two, when global sports paused, I made a mini podcast called View From the Empty Seat, where I invited fans to tell their most memorable stadium memory. I interviewed forty-seven people over three months. From a seventy-eight-year-old woman in Busan who had not missed a home match in forty years, to a young man who once walked two hundred kilometers to see a cup final. The most-shared episode drew over fifty thousand listens in its first week.

What I learned from those forty-seven people is: fan emotion does not need numbers to exist. But analysis does.

On using numbers to tell stories

Here is what I want to say directly to those sharing playoff ranking tables without context.

A number without context is not information. It is a puzzle piece pulled from the picture and framed on its own. The picture may be a team testing a new lineup. The picture may be a team hiding cards for a bigger event. The picture may be a team playing through a patch it has not yet adapted to — something any team can face, not just T1.

I have been on the other side. At twenty-one, after a stoppage-time draw that all but ended a qualification dream, a wave of criticism crashed onto the national team coach. I wrote against the consensus: do not blame the coach, look at the five mistakes the players made themselves. I cited data — twenty-three misplaced passes in the last fifteen minutes, a lead striker touching the ball only eight times in ninety minutes. The piece drew over a million reads in twenty-four hours. Several players later said publicly they read it and re-examined their play.

I tell this not to brag. I tell it to say I know the power of a number placed correctly. But I also know the destructive force of a number placed wrongly. Same number, two outcomes. What decides is context.

And the context here — with a six-team and eight-team sample, with an unnamed statistics source, with not a single specific patch cited — is a lack of context.

Where the real worry sits

If I had to point at the real worry in this story, it is not Faker's metrics. It is that two people posted low numbers at the same time.

In systems analysis, when two theoretically independent variables move in the same direction together, you must look for a shared cause. Not two separate causes. One shared cause.

The shared cause could be scrim quality. Could be the coaching staff's patch understanding. Could be a schedule so dense that practice time compresses. Could be something inside the team's structure changing that outsiders cannot see.

And here is what I want to stress: if it is a systemic problem, it can be fixed far faster than an individual one. A player out of form needs time. A system out of rhythm can be fixed in a week of correct practice.

That is why I am pessimistic about the number but optimistic about the structure. Those two are not contradictory. They are two layers of the same problem.

On brand and value beyond the stage

There is one aspect of this story I think few notice, and it says more about the industry than the match.

Faker is not merely a mid laner. He is a brand beyond the discipline's borders. There are signs that industries outside esports are paying attention to him — including large technology companies. That is a signal financial analysts should care about more than match analysts.

If a player's commercial value can decouple from his competitive value, then the whole logic of "performance determines value" that we still use to judge esports is being challenged. A team can lose many games and keep sponsorship. A player can post low metrics and remain the centerpiece of a global campaign.

This means pressure on players will change shape. Pressure no longer comes only from match results. It comes from being a commercial asset that must be present, must speak, must appear.

I do not say this to excuse anyone. I say it to widen the analytical frame. If we only read ranking tables and conclude about people, we are missing most of the story.

On regional context

One more thing to put in the frame: this is a period when a multi-sport event with an esports program is approaching. That means some players must balance club and national team duties. Schedules fragment. Practice time splits. And for a team with many national-team players, this is a pressure factor no ranking table reflects.

In Southeast Asia, where I grew up and where part of my audience still is, T1 and Faker remain a cultural icon beyond the discipline. That means the story about them is read with an emotion other markets do not have. And that emotion, added to a small data sample, creates a very flammable mix.

Closing: what I will track instead of what I will predict

I have no prediction about what T1 will do at Worlds. Predicting a team with a history of rebounds is a game where both outcomes can be judged correct. I do not play that game.

What I have is a set of criteria to track, and I will make those criteria public so anyone can check me later.

I will track official pick/ban data across patches to see where the jungle role truly sits in the priority order. I will track T1's metrics over a larger sample — at least the full regular season — to separate variance from trend. I will track official coaching statements about roster or approach changes. I will track any information about player health or time budget, because that is the hidden variable that could falsify my entire analysis.

And if, after all that, larger-sample data still shows these two veterans systematically below same-position peers, I will be the first to rewrite this piece. Not because I fear being wrong. Because an error corrected is a lesson, while an error defended is a belief.

What I keep after all this is not a conclusion about a team. It is a question about how we do this work. When does a number become a truth? When it is measured correctly, normalized correctly, and placed in a sample large enough to bear the weight of a conclusion. Those three conditions. Not one less.

And if you are a fan reading a playoff table right now, I want to ask one question with no right answer: are you reading a number about a player's present, or a number about a team's recent past? Those are different. And the gap between them may be the whole story.