Four boats in under two seconds. Why nobody is safe in Poznań 2026
- Published
- 2026-08-26T09:20:24+00:00
- Author
- PaddleStats
Full competition results: ICF Canoe Sprint and Paracanoe World Championships - Poznan
In Szeged in May, Bálint Kopasz paddled the K1 1000 metres in 3:22.33. That is quicker than the time that won him the world title three years earlier, when 3:27.46 was enough. Quicker, too, than the 3:25.68 that put him on the bottom step of the Olympic podium in Paris.
Kopasz finished second.
Fernando Pimenta won it in 3:22.25, eight hundredths ahead. Thomas Green came home third in 3:22.93, the fourth boat in 3:23.99. Four crews inside two seconds; the top three inside one.
Kopasz has not slowed down. The rest have caught up.
That single race explains why, going into the World Championships in Poznań, no paddler carries more than a 28 per cent chance of gold. When half a metre separates gold from silver, there is nothing left to build certainty on. All that remains is to measure how narrow it has become at the top, and to say how many people are crowded in there.
Pimenta and Kopasz meet over the same distance again this week.
Who leads the fifteen solo events
Fifteen solo events, kayak and canoe. In each of them, whoever the model currently rates highest:
| Event | Top-rated | Gold | Podium | Top ten | Entered |
|---|---|---|---|---|---|
| C1 men 200 m | Artur Guliev (UZB) | 10% | 34% | 85% | 32 |
| C1 men 500 m | Martin Fuksa (CZE) | 13% | 41% | 89% | 37 |
| C1 men 1000 m | Martin Fuksa (CZE) | 13% | 44% | 91% | 36 |
| C1 men 5000 m | Balázs Adolf (HUN) | 28% | 75% | 99% | 30 |
| C1 women 200 m | Liudmyla Luzan (UKR) | 27% | 75% | 99% | 34 |
| C1 women 500 m | Sophia Jensen (CAN) | 16% | 44% | 92% | 28 |
| C1 women 5000 m | Maria Mailliard (CHI) | 24% | 64% | 99% | 19 |
| K1 men 200 m | Artūras Seja (LTU) | 9% | 30% | 78% | 38 |
| K1 men 500 m | Alex Borucki (POL) | 8% | 31% | 77% | 48 |
| K1 men 1000 m | Bálint Kopasz (HUN) | 12% | 44% | 91% | 47 |
| K1 men 5000 m | Fernando Pimenta (POR) | 17% | 59% | 98% | 44 |
| K1 women 200 m | Frederikke Mørcke (DEN) | 11% | 35% | 84% | 31 |
| K1 women 500 m | Pauline Jagsch (GER) | 12% | 43% | 91% | 39 |
| K1 women 1000 m | Zsóka Csikós (HUN) | 14% | 43% | 92% | 28 |
| K1 women 5000 m | Melina Andersson (SWE) | 21% | 60% | 98% | 30 |
Fuksa leads two events and gets the same number in both. The Olympic champion is on 13 per cent over 500 metres and 13 over 1000. Two different distances, two near-identical fields of 37 and 36 boats, and each time the same number of rivals capable of beating him.
The long distance sits highest. Adolf, Mailliard, Andersson, Pimenta — four of the five biggest numbers in the table land over 5000 metres. Endurance decides there, and endurance moves more slowly than 200-metre form, so the field stretches out and the top of it is easier to see. The one exception is Luzan and her 27 per cent in the canoe 200.
The men's K1 is the most open of the lot. Forty-eight paddlers entered the 500, more than any other event on the programme, and Borucki leads it on 8 per cent. Over 200 metres the leader is on 9. In those two races the model does not really see a favourite at all.
Where the numbers come from
Before every solo event we publish three figures for each entrant: the chance of gold, of a podium, and of a top-ten finish. Each goes up before the race, with a timestamp, and stays as it was.
The model knows nothing about how anyone slept, or how the wind will sit on Malta at three in the afternoon. It knows what can be read out of previous races, and it reduces that to three questions.
How fast are they. The simplest and, as it turns out, the most telling. Time from an international regatta over the same distance, set against the rest of the field.
How close to the winner do they finish. Fifth in a European final is not fifth in a World Cup heat. The model measures the gap to the winner and compares it only within that one race. Everyone paddled the same water in the same wind, so what is left is the distance to the best man or woman on the day.
Who do they beat. A head-to-head record, weighted the way a chess rating is: beating a strong rival counts for more than beating a weak one.
Which of the three matters most was not decided in advance. The model was handed 1551 international races from before this season and looked for the balance that most often identified the actual winner. Raw speed says the most. The gap to the winner comes next. Head-to-head records say the least.
Which makes sense. Beating a slower rival tells you less than the time you had to produce.
A model tested only on what it learned from is an opinion with a chart attached. This one learned on races up to the end of 2025, then had to call 99 international finals from this season it had never seen.
Its top pick won 36 of those 99. Picking blind from a nine-boat final would have landed eleven. The actual winner sat inside the top three 68 times; blind picking gives 33.
Three times better than chance, and still wrong more often than right.
A number beside a name is not a prediction of victory. It is a measure of how often victory arrives. Thirteen per cent means one win in roughly eight starts like this one — and it is only honest if the other seven are genuinely lost. If the top picks won every time, one of two things would be true: the numbers are invented, or canoe sprint has stopped being a sport.
So after every block of finals the same reckoning comes back: what was written before the start, and what happened. Saturday evening the same as Sunday, after a good day the same as after a bad one. A forecast nobody settles up on is not a forecast.
How to read the percentages
The trap is reading a number as a verdict. Adolf's 28 per cent does not mean he will probably win. It means he wins one of these races in four, and somebody else takes the other three. And that is the highest figure of the entire championship.
A percentage always belongs to a particular field. Forty-eight paddlers entered the men's K1 500 metres. Eight per cent against that many is a different thing entirely from eight per cent in a nine-boat final. After the semi-finals the field narrows to nine, so the numbers are recalculated for that nine and republished on the morning of the finals. They rise sharply then — not because anyone found form between Wednesday and Saturday, but because 37 rivals became eight.
And one thing that gets lost in the tables: the top-ten chance is often more interesting than the gold. Tenth place is a B-final win, and world championship ranking points count one and a half times towards Los Angeles. In several events, more paddlers are fighting to get into the ten than are fighting for a medal.
There is a second side to the same honesty: the limits of what can be counted at all. The forecast covers solo boats, kayak and canoe. Crews have to be counted differently: a four goes as well as it fits together, not as well as four sets of abilities add up. That is separate work and it is under way.
A number appears beside a name only when there is something to calculate it from: three international starts over the distance and a confirmed eighteenth birthday. Anyone short of that stays in the standings without a percentage — they are racing and they take chances away from everyone else, but putting a figure beside them would be pretending to know.
Eleven finals, not once
The table says who stands highest. It does not say what happens between particular people, and that is often the better story.
Wiktor Głazunow and Martin Fuksa have met in eleven international A-finals. The record is as one-sided as records get: eleven to nothing for the Czech. The Pole came closest in Poznań in 2023, beaten by little more than a tenth. They raced four times this season and Fuksa finished first every time, with Głazunow seventh, seventh, sixth and fourth.
That run says something the table cannot. Głazunow is third in the 1000-metre standings with a 21 per cent podium chance, a solid position in a field of 36. But eleven straight defeats to one rival means his road to gold runs through a man he has never once got past. The model counts that without sentiment: high placing, low chance of winning.
The distances change the picture, though. Fuksa does not race the 5000, and there Głazunow is second behind Balázs Adolf, with 44 per cent for a medal.
A similarly one-sided run is under way in the women's canoe 200. Liudmyla Luzan has won all four international finals over the distance this season: three World Cups and the European Championships. Hence her 27 per cent, the second-highest figure of the championship and the only one on a short distance that comes near thirty.
The most open story of the week, though, is somewhere else. One event has a Pole at the top of the standings. Alex Borucki leads the men's K1 500 metres, in the race that drew more entries than any other on the programme. Eight per cent for gold sounds modest, but against 48 starters it is nearly four times what an even split would give him. His podium chance is 31 per cent, his top-ten chance 77.
Right behind him is Josef Dostál, entered only for the 500 metres in Poznań. Third is Adam Varga. Two percentage points separate first from third — less than the model's own margin of error. Which is another way of saying the model cannot tell them apart.
In the women's K1 500, Anna Puławska is second on 9 per cent, three behind Pauline Jagsch. Wiktor Głazunow is third in the men's canoe 1000, and Dorota Borowska third in the women's canoe 200 behind Luzan and Jensen.
What's next in Poznań
The championships open on Wednesday on the Malta course and run to Sunday. Medals go out from the opening day: three A-finals on Wednesday and two on Thursday. Then it thickens — eight finals on Friday, seven on Saturday, and on Sunday seven in the morning plus four 5000-metre races in the afternoon.
An updated forecast goes up before each day of finals, and a reckoning after each block. The full programme is on the championship page.
Frequently Asked Questions
Is this betting?
No. We publish no odds, work with no bookmakers and encourage nobody to gamble. It is a statistical forecast — a way of reading a field before the start.
How accurate are these numbers?
Checked across 99 finals this season: paddlers marked at 30 per cent won roughly three times in ten. The figures missed reality by about two percentage points on average, so "30 per cent" means something between 28 and 32 in practice.
What does the top-ten chance mean?
The probability of finishing the championship inside the first ten of the event — winning the B final or reaching the A. That is what counts towards the qualification ranking for the Los Angeles Games.
Why do some paddlers have no percentage?
Because after one or two international starts a number would say more about the starting point than about the athlete. Better a note than a figure with nothing behind it. They stay in the calculation, because they race and they take chances from the rest.
Does the model include neutral athletes?
Yes. They race the same events and take chances from everyone else, so they belong in the arithmetic. Their names do not go on the graphics published to social media.
What happens if the forecast is wrong?
We write about it the same day, in the same format as a hit. A model that shows only its successes stops being a model.
A forecast is not a prophecy, and it is not there to spoil the surprises. It is there so you know which surprise is the bigger one: a win by the paddler on 28 per cent, or by the one on three.
The model sees only what can be counted from earlier results. It does not see how somebody feels on Thursday morning, how the wind sits on Malta at three, or what goes through a paddler's head over the last 200 metres. Which is precisely why the races are run at all.
By Sunday night we will know how much of this held up. Some of the calls will land, most will not — that is the honest arithmetic when 48 boats chase nine places in a final. Then the same table comes back, with the places where the numbers were right and the places where the water said otherwise. The first of those reckonings belongs to Alex Borucki and his 500 metres.