Asian Games 2026: Indonesia's Four Doubles Scorelines and One Dirty Data Cell

I read Bola.net's match report on the men's team semifinal at the 2026 Asian Games in Aichi-Nagoya and stopped at four numbers: 21-13, 21-19, 21-15, 21-13. Both Indonesian men's doubles pairs won in straight games. In a three-singles, two-doubles format, that is the loudest signal a team can send.

Then I reached the fifth line, and that line is not clean. The report says a player "won game one" while the score cell next to it reads 16-21. In badminton, 16-21 is a loss. There is no other reading. Sports reports written on feeling drop data in places like this often enough that I am past being surprised. What matters is that most readers will slide past that line and build their conclusions on a wrong cell.

The four doubles scorelines are the opposite problem: they are too clean. And the cleaner the number, the harder I have to ask what it is hiding.

The path and the format

Indonesia reached the final by beating Malaysia 3-2 in the quarterfinal and Thailand 3-1 in the semifinal. The women's team stopped in the semifinal against China. The final opponent was undetermined when the report was published: in the other semifinal, China led India 2-1 with the fourth doubles rubber still in progress.

The Asian Games men's team format runs singles – doubles – singles – doubles, with the third singles played only if the tie reaches 2-2. That detail decides how the whole match should be read, so I separate it out.

Under that order, a doubles-strong team locks the tie with two doubles points plus one singles point and never touches the third singles. A singles-strong but doubles-thin team must win two of three singles, meaning it has no room for error. Two entirely different risk profiles, both capable of producing the same 3-1 on the scoreboard.

The 3-2 quarterfinal against Malaysia is the warning. A one-rubber margin means Indonesia had to use the fifth match in the previous round. The 3-1 semifinal did not.

The table and what it says

| Rubber | Event | Player / pair | Score | Point differential | |---|---|---|---|---| | 1 | First singles | Jonatan Christie | 17-21, 14-21 | -4, -7 → -11 | | 2 | First doubles | One of the two men's pairs | 21-13, 21-19 | +8, +2 → +10 | | 3 | Second singles | Alwi Farhan | 21-15, 21-13 | +6, +8 → +14 | | 4 | Second doubles | The other men's pair | 21-15, 21-13 | +6, +8 → +14 |

One thing must be said clearly about this table: the source does not attach each scoreline to a specific pair. Fajar Alfian – Muhammad Shohibul Fikri and Leo Rolly Carnando – Daniel Marthin are both named, but no line states which pair won 21-13/21-19 and which won 21-15/21-13. So in the third column I write "one of the two pairs" rather than assign it wrongly. This is the kind of error I have seen often enough in analysis: the writer attaches numbers to names because names read better than numbers.

Added up, the two doubles rubbers produced +24 points across four games, an average of +6.0 per game. The two singles rubbers produced +3. Indonesia's entire positive balance sits in doubles. That is arithmetic, not opinion.

The 21-19 game is the only one of the four doubles games decided by fewer than three points. The other three were won by 6, 6 and 8 points. At continental team level, a 6-to-8-point margin usually indicates a controlled match rather than a coin flip. But I have to say immediately: this is an inference from scorelines, not a measurement. The report contains no indicator for rallies over 15 shots, net-point conversion, unforced-error rate, or distance covered. Those metrics are what answer "controlled how."

When scorelines are all I have, I stop at the level of result and do not jump to the level of mechanism. Months ago I was asked why I would not write "tragedy" because it reads easier. The answer is the same: the word "tragedy" does not add up to points.

Two individual signals are clearer. Jonatan Christie lost 17-21, 14-21. Losing game two by 7 points is a heavier signal than losing game one by 4 — the player did not hold the middle of the match. Alwi Farhan won 21-15, 21-13, the most convincing Indonesian singles performance of the semifinal, and he was placed in the second singles slot rather than the third. Using a young player in that slot in a team tie where the side is already a rubber down is a decision with weight.

A single match is randomness, but a season is where probability exposes every truth. Here I have one match, not a season. So I call it a signal and mark the confidence level beside it.

Where I do not trust the report

The report sells a tidy story: Indonesia in the final, a first gold within reach, doubles as the pillar. I read that story at the level of hypothesis.

The hypothesis that this team is built doubles-first carries low confidence, because the evidence is one match. To talk about team-building strategy I need a run of matches with the same line-up order. The 3-2 quarterfinal against Malaysia may have followed a completely different script, and the report gives no detail.

The pressure narrative works the same way: the women's team lost to China in the semifinal, so the men carry the gold expectation. That is narrative inference. It may be true, but it is not data, and I do not let it into the conclusion.

What I strike entirely is the data quality of the other semifinal. The report says a player "won game one" while the score reads 16-21. One of the two is wrong: the wording or the number. For a match whose result determines Indonesia's opponent, I will not use that line for any inference until it is checked against official results.

One more point on the first-singles opponent's name. If the Thai player in the report is a transliteration variant of Kunlavut Vitidsarn, then the 17-21, 14-21 defeat means something different: a loss to a top-class opponent rather than a slip against a lesser one. Reading it as "Jonatan is out of form" and reading it as "Jonatan lost to a strong opponent" lead to two different selection decisions for the final. The report does not let me distinguish, so I keep both possibilities with medium confidence.

When the media calls it a miracle, I call it a sequence of probability distributions — and a distribution needs more than one data point to estimate.

Here is an old example of why I am strict about a wrong cell. In May 2026 at Lach Tray, I logged every Hai Phong shot against SHB Da Nang, computed an xG of 0.8 against the opponent's 1.9, and was told to my face that a woman knows nothing about tactics. The spreadsheet was right, the second-half prediction was right, the final score was 1-2. I drew the old rule from it: tactics have no gender, and neither does data. That only holds under one condition — the cell I enter has to be a cell I can read.

The 2026 transfer window taught me once more not to assign numbers carelessly. I ruled out the most expensive target outright because of a G-xG of minus 2.1, and recommended a 23-year-old at 40 percent lower cost. The spreadsheet was right. But a spreadsheet is only right when every cell is entered properly.

Signals for the final

Against China the calculation changes. That team has enough depth in both singles and doubles that Indonesia would likely have to win both doubles rubbers plus one singles point. The form of the first singles player from the semifinal then becomes the central selection risk.

Against India the door is wider but not easy. India pushed China to a fourth doubles rubber in the semifinal, meaning that team does not collapse mid-tie. Indonesia's door still sits in doubles.

Three things I will watch in the final, and watch through a table rather than with my eyes: whether the line-up order stays the same; whether the first singles repeats the exact pattern of a heavy game-two loss; and whether the team has a credible option for the third singles if the tie ever reaches it. Today's report answers none of those three. A final answers all of them.

Data never tells a sad story, it only points to whoever is lying to themselves. In those four scorelines, the story being told belongs to the doubles pairs. The open question left for final night: if the two pairs split one win and one loss, where does Indonesia's third point come from? Nobody in the technical meeting should fall asleep before answering that.