The Empty Dataset in Penang and the Final Three Checks

2:47 a.m. Penang time. I reopen the analysis file from the night shift, the one I keep apart in a folder labelled unverified. On the screen, a single column of text runs down, repeated eleven times in eleven different cells: insufficient information. No tournament name. No match-up. No timestamp. No odds board. Only the skeleton of an analysis presented with great formality, complete with column headers, complete with slashes, and one warning line at the bottom: risk level high, please supply valid input data before analysis.

I read it three times. Then I poured a black coffee with no sugar and sat still for about ten minutes. Not out of confusion. Because in nearly twenty years of reading sports numbers, this was the first time I had met a dataset honest enough to be almost rude.

My job is to turn a match into sequences of numbers and then hand that match back as a story that can be verified. I live in Penang, I was born in Vietnam, I write about badminton and football for the Malaysian market, and I make a living reading what the money is saying before the referee blows the whistle. In this trade the common fear is missing data. My fear sits on the opposite side.

Penang is where I buried a part of my innocence; since then I have dug data the way others dig graves.

In 2026, I was thirty-nine and accepted a role as an analyst for a newly launched television channel. The match was in the Malaysian second tier, between Pulau Pinang and Johor Darul Ta'zim. I used self-collected xG data, sat counting every shot, and reached a result: the home side generated 2.8 xG but finished the match with a 0-2 defeat. I went on air and said Pulau Pinang had in fact played the better game in terms of chances. The reaction was fast and not gentle. I was accused of not understanding football, of bringing a spreadsheet onto grass.

A week later, the Pulau Pinang head coach was sacked for poor results. The team won four straight matches under the assistant. My data was right. But it was only when I sat alone in Penang that I understood something more important: being right is never enough, and a correct conclusion presented the wrong way is still treated as a wrong conclusion.

From then on I wrote differently. Numbers became testimony, not verdict. I never issue a judgement based only on the scoreline. In every piece I cite xG, passes into dangerous areas, second-half running distance, and only then do I allow myself to talk about the emotions of the match.

In 2026 I joined a live analysis team at the World Cup in Russia, set a tablet on the stands and collected PPDA figures match by match. In the group stage, when England met Tunisia, I noticed Harry Kane had a habit of drifting to the far post in the final five minutes. Against Panama the pattern repeated: three attempts from inside five metres, two goals. I published that observation on my personal blog. It drew fifty thousand reads, and an Asian bookmaker called to propose an analytics partnership. That was the turning point that took me from amateur writer to contracted professional.

Moscow on a World Cup night: money flowed like the Volga, and I was only a leaf. I wrote that line that year and have reused it many times. These days I limit it myself, because an image can be beautiful enough to become a habit if you are not careful. Money flow is not destiny. Money flow is a conversation, and you only hear it if you stand close enough not to be swept away.

In 2026, when the season was suspended by the pandemic, I threw myself into the question of home advantage without crowds. I collected data from 145 Bundesliga matches after the restart. The home win rate fell from 43% to 31%. The over-under rate rose 12%. I published on a small account and was attacked by Western analysts for using too small a sample. I did not argue. I tracked 98 more matches in Hungary and Portugal. The results pointed the same way. In the end, major outlets cited that research as one of the rare studies on pandemic football.

In 2026, at a major European tournament, I ran a data column for a regional sports newspaper. Before the final between Italy and England, I analysed PPDA: Italy allowed opponents only 11.2 passes per defensive action, England 13.8. I predicted a card-heavy second half as England's press broke, and bet accordingly. The match closed with six yellow cards. The 120 million dong I won went into building software that measures pressing fatigue from each player's running distance.

Behind all those figures sits something smaller: I had been in commentary studios for major events since 2026, from a table tennis World Cup to badminton's Sudirman Cup. That cross-sport experience taught me something people who follow one sport tend to miss: the way data is produced always carries the fingerprints of the rules, the calendar, and the habits of the person doing the counting.

So what does an empty table mean inside my process?

My process has three layers. Layer one is supply: odds boards, line movement, schedules, line-ups, commercial data. Layer two is primary source: video, notes taken at the venue, numbers I count myself. Layer three is cross-checking: at least three independent sources for the same fact, and one of the three must be something I collected with my own hands.

When layer one is empty, most writers panic and fill the gap with guesswork. I choose the opposite. An empty table is not a wrong table. It is simply a table with nothing to say yet. My job then is to downgrade the question: instead of asking who wins, ask what I can observe that nobody can dispute. That is the only way to keep a conclusion from being contaminated at the source.

In football, the undisputable things are usually the speed of ball circulation in the opponent's final third, the number of passes into dangerous areas, the number of turnovers within ten seconds of winning the ball, and running distance between the 60th and 75th minute. PPDA tells you how many passes a team tolerates before intervening. Late-match running distance tells you which team is about to break and where. For me, minutes 60 to 75 are the single most important window of any match.

In badminton the list changes. I count average rally length, the tempo of serve changes, reaction speed after a low serve, the rate of coming to the net within the first three beats of a point, and how a player handles 18-all. Badminton is the fastest micro-market in Asia. Every serve change, every run of points, every reflex instantly becomes a live betting line. I read those small movements the way I read an order being filled.

There is a line I use as a professional principle: Players do not listen to the crowd, they play like machines; but bookmakers have never been mechanical. What that means is that when the table is empty, I do not go looking for data where the crowd is thickest. I go back to where the fewest people are looking.

And I still hold one sentence as an oath: I do not trust a single statistic that cannot be used to arrange. This needs to be stated plainly here, because the term is easily misread: arranging, in my usage, means reordering the narrative and the priority of the numbers so that a scattered fact becomes a meaningful link in a chain. It belongs to method. It is not interference in a match result, and anyone reading that word in my work should understand it that way.

Living between two homelands gives me an advantage global models never touch. Money flow in Vietnam and in Malaysia does not move to the same rhythm. Time-zone gaps, exchange-rate gaps between dong and ringgit, and gaps in player psychology: players on the two sides of the strait react to the same news at two different speeds. Those gaps create price zones that rankings and standard models never see. To read them you have to sit exactly where the two meet, not at the centre.

But then the empty table taught me the opposite of my own instinct.

The dangerous thing in this trade is not too little data. The dangerous thing is too much. A packed file makes a writer feel entitled to conclude, while most of the numbers inside have never had their sources checked, have never been read in context, and have been copied over from another season. The empty table in Penang that morning was far more honest than many beautiful files I have received.

That leads to the hardest part: correlation is not causation. In 2026 it would have been easy to tell myself a flattering story that my data got the Pulau Pinang coach fired and the team won four matches afterwards. Wrong. Results got him fired. The data was only testimony read out before the outcome arrived. Confusing the two is the fastest way for an analyst to turn into a teller of fairy tales.

The same error appears in subtler places. Possession share is the most deceptive metric in modern football. Many teams grind their way to 60% of the ball with sideways passes carrying no intent to break a line at all. On a stat sheet they look in control. On video, they are passing the ball to their own fear.

Another field I track shows the same disease: esports betting. There, competitive integrity erodes faster than in traditional sport, simply because the rulebook trails reality by too far. Matches are numerous, money cycles fast, and integrity staffing is thin. When the data structure cannot keep pace with the speed of money, the person reading the numbers has to build their own fence.

If the empty table taught me one thing, it is this: the honesty of data lies not in how many cells are filled, but in how many times you are willing to delete a cell because you could not verify it. Read twice, publish once. No exceptions, even when the line is live and people are waiting.

That empty table is still in my unverified folder, and I have no intention of deleting it.

What to watch in the next cycle is not on the scoreboard. It is in the second-half running distance of teams that have just played three matches in eight days; in how a player handles the serve at 18-all after four consecutive matches; in the lag between two exchanges when news crosses a border twenty minutes later on one side than the other. Whoever sits long enough in those places sees the signal before it becomes a headline.

Three months living with a World Cup taught me this: money never runs straight. It runs in loops, off-centre, through places people do not think are roads. The job of the one counting is not to guess where it turns, but to plant enough markers to know which stretch he is standing on.

This empty table will fill up eventually. But the question I carry into the next live shift is not when it will fill. It is whether, once it is full, I will have the nerve to delete it again.