South Korea's Five Medals, Four Empty Rows: Reconstructing the Aichi-Nagoya Esports Numbers

At eleven at night last week I opened a spreadsheet. Thirty rows, four columns. One column for the event, one for the format, one for the result, one for the source. Thirty information points pulled from the only published recap of the Aichi-Nagoya Asian Games esports programme. Five cells on the medal table were filled: two golds, one silver, two bronzes.

Four cells on the right edge were empty. That is the first rule of my work: an empty cell is data too. A recap that celebrates five medals and lists nine events is obliged to account for four failures.

I opened the spreadsheet. 3,800 matches later, the pattern was already there. That line is from 2026, when I was an economics student at Baruch, scraping five seasons of shot data across Europe's top five leagues and building my first expected-goals model in R. Shot volume was noise; xG per shot separated real dominance from lucky scorelines. The habit never left: numbers before narrative.

Context: nine events, eleven titles, one medal table

The 2026 Aichi-Nagoya esports programme awarded medals in nine events — League of Legends; a Fighting Games composite of Tekken 8, KOF XV and Street Fighter 6; Puyo Puyo Champions; Gran Turismo 7; Pokemon UNITE; PUBG Mobile; Honor of Kings; Identity V; and eFootball. This is not a publisher circuit. It is a national-team competition under the Olympic Council of Asia umbrella, and the same medal table rewards radically different skill sets: MOBA team cohesion, racing consistency, battle-royale survival.

That single sentence frames everything. A national roster is not a club squad. South Korea's League of Legends side combined three returning Hangzhou gold medalists — Faker, Zeus, Keria — with three debutants in Canyon, Zeka and Gumayusi. Formats differed event by event: LoL ran group stage into single elimination, a 2-0 semifinal and a 3-0 final; the fighting-game event ran a set-based team relay; Gran Turismo 7 aggregated points across three races; Pokemon UNITE ran single elimination with a shared bronze; PUBG Mobile ran a points-based battle royale.

My second old habit applies here. At the 2026 World Cup I live-tweeted Germany's group-stage collapse. Against Mexico on June 17 they took 26 shots for 1.9 xG — possession without penetration. In Kazan on June 27 they took 28 shots for 2.7 xG and scored none. — Root: Germany. That taught me to read the format before drawing conclusions from the result.

On May 16, 2026 the Bundesliga returned to empty stadiums. Across the first 83 matches behind closed doors, the home win rate fell from 43 percent to 33 percent and home penalties dropped sharply. The empty stadium didn — that is where I started hunting structural breaks, the quiet moments when a rule of the game changes. Aichi-Nagoya has one, and its name is format heterogeneity.

Core 1: five medals, three different weights

I set a simple weight — gold 3, silver 2, bronze 1. South Korea's raw score came to 10. Then I adjusted twice, for format variance and field strength.

The Pokemon UNITE bronze was shared. Both semifinal losers received bronze with no bronze match played, so the competitive separation inside that medal is roughly zero. The Gran Turismo 7 bronze came from a three-race points aggregate, where one mistake can sink the whole total — a consistency prize, not a peak-speed one. The League of Legends gold came with an 8-0 record and a 3-0 final sweep.

All three sit in one medal table, and none of them ever weighed the same. Counting medals is easy. Weighing them is hard, and most recaps skip the weighing.

Core 2: roster architecture — continuity plus injection

Faker, Zeus and Keria brought institutional memory. Canyon, Zeka and Gumayusi brought first-time peak form, and all three won gold on debut. That is not an accident; it is selection-committee architecture. Memory in mid lane, continuity in top lane, fresh carry form in the damage roles is the lowest-risk national-team build. A fully new roster carries chemistry risk; a fully old roster carries form risk. Korea split the risk in half.

Eight games, eight wins, a 3-0 final: that is evidence of chemistry, not of talent. The talent was already on paper.

One caveat matters. Group B listed Chinese Taipei, India and Hong Kong. The semifinal was Saudi Arabia. The final was Chinese Taipei. China appears nowhere. For a continental Games, that is a conspicuous gap, and it directly discounts the value of the 8-0 record. Without the bracket I will not use the word unbeatable.

Core 3: the fighting-game gold had three heroes, not one

Korea trailed 1-2 in the final and won 5-2. Bae Jae-min lost the opener and then took two sets. Lee Kwang-no won two matches. Yeon Je-gil avenged an earlier defeat to Kojiro Higuchi.

That is a distributed-hero pattern — three separate clutch points rather than dependence on one star. It is the hardest shape to prepare against, because blocking one player leaves the other two to close the remaining sets. Set-based relay formats amplify momentum, which makes this result the least reducible to stronger nation wins.

Core 4: Japan's host dividend is real, and it splits the region

Kang Dong-shin lost the Puyo Puyo Champions final 2-10 to Yuki Kurihara. A double-digit set deficit suggests a structural matchup problem rather than a narrow form dip. Japan eliminated South Korea from Pokemon UNITE with a 2-0 semifinal. Japan was the opponent in the fighting-game team final.

The regional picture is bifurcated, not hierarchical: Korea leads in team and MOBA titles, while Japan is competitive in fighting, racing and puzzle formats. Host dividend is not only crowd support — it is event selection, preparation advantage and quota design.

Core 5: the rows that are missing

Honor of Kings: quarterfinal exit. Identity V: quarterfinal exit. eFootball: group-stage exit. PUBG Mobile: sixth place on 71 points. All part of the same nine events.

The market prices the story. The spreadsheet prices the mistake. Five podiums out of nine events means four empty hands, and a framing that stops at breadth keeps those four in shadow.

Core 6: why format design fills the medal table with noise

Longer series suppress upset variance and reward the stronger side — hence Korea's clean sweeps. Set-based relays reward clutch capacity over raw numbers. Three-race aggregates reward the fewest mistakes rather than the highest speed. Single elimination means one bad match and the medal is gone. Points-based battle royale sometimes rewards consistent top finishes more than a single win.

Five entirely different competitive philosophies sit inside one medal table. Add them together and the resulting number measures event design, not skill.

Core 7: my field-strength discount

I cut one point for the shared Pokemon UNITE bronze, because no victory match backs it, and discounted the LoL gold from 3 to 2.5 given China's absence from the listed bracket. The adjusted score lands near 8.5. The direction matters more than the number: Korea's result is real, but it hangs between excellent and unchallenged, and I will not use the second word until the field is verified.

Contrarian: a superpower claim standing on two samples

The LoL dynasty claim rests on two Asian Games cycles. Two data points can form a pattern; they cannot prove invincibility. A model that writes certain after two rows is not a model, it is a slogan.

Format heterogeneity is itself a noise generator: nine formats means nine skill sets, and folding them into one table inflates the medal count while thinning the meaning of each medal.

The most uncomfortable point is provenance. The tally comes from a single source, and the author bio attached to it describes a sponsorship and money-flow reporter — yet the text contains not one financial figure, no contract, no prize pool, no allocation. That is a beat mismatch, and beat mismatch usually signals syndicated or templated copy. It does not make the results false; it makes them unverifiable. I don’t trust narratives. I trust rows that survive a filter. These rows have not passed a second filter.

And the human ledger. When Christian Eriksen collapsed on June 12, 2026, my models had nothing to say. I spent that night counting Denmark's rows instead — 0-1 to Finland, 4-1 over Russia, 2-1 extra-time defeat to England at Wembley. It became my most-read piece because it was about the limits of numbers. Kang Dong-shin's 2-10 is a row, but he is a person who prepared for years and lost ten sets in a final. The model says X, but here is what it cannot see. A scoreboard does not measure mood, sleep or pressure.

Takeaway

Watch individual-title pipeline investment. If Korea brings home no gold in fighting, racing or puzzle formats next cycle, the breadth story breaks and the empty cells multiply. And whether China returns to the bracket will set the real value of the LoL record: the 8-0 stays the same number, its meaning does not.

An xG map is not a verdict. It is a hypothesis generator. So is a medal table. Before the next Asian Games, one question is worth asking: are we measuring Korea's strength, or measuring who was not there?