Use case · office pools and Kicktipp rounds
How to fill in a World Cup prediction pool with probabilities
Most people enter a pool with their heart, a grudge and a vague memory of a friendly two years ago. A probability will not tell you the result, but it will tell you where the guessing is worth doing and where it is a waste of an entry.
Free, and the numbers are on every fixture.
First, find out what your pool actually rewards
Pools score in one of three ways, and they call for completely different behaviour.
- Tendency. A point for picking the right side, or the draw. Nothing else counts.
- Tendency plus goal difference. More points if you also got the margin right.
- Exact score. The full points only for the precise result, usually with partial credit below it. Kicktipp rounds are normally shaped like this.
The reason this matters is that the reliable information and the valuable information sit in different places. The tendency is where a probability is genuinely strong. The exact score is close to a lottery for everybody, and the numbers below show how close.
Why exact scores are a lottery, with figures
Across all 104 matches of the 2026 World Cup, the single likeliest scoreline averaged 14.3%. It cleared fifteen percent in 19 matches and twenty percent in only 8.
The most extreme case makes the point better than any average. The most one-sided match of the whole tournament was Germany against Curaçao on Sun 14 June 2026, where the market gave the home side 92.6 percent. Overwhelming. And yet the likeliest exact scoreline was 3:0, worth 13.1 percent. Even when you are almost certain who wins, you are nowhere near certain by how much.
So in an exact score pool, do not spend effort hunting the perfect scoreline. Spend it making sure the tendency is right and the scoreline is a plausible one.
A small set of scorelines does most of the work
The likeliest scoreline was not scattered across the whole grid. Out of 104 matches it was:
| Scoreline | Matches | Share of 104 |
|---|---|---|
| 1:1 | 37 | 36% |
| 2:0 | 24 | 23% |
| 1:0 | 13 | 12% |
| 0:2 | 11 | 11% |
| 0:1 | 10 | 10% |
| 3:0 | 5 | 5% |
1:1 and 2:0 between them led 61 of the 104 matches. If your pool gives partial credit for the right goal difference, those two results are doing an enormous amount of the work, and entering an exotic 3:2 is costing you far more often than it pays.
The draw is the pick nobody makes
The market gave the draw 22.2% on average, and in 28 of the 104 matches it was worth at least twenty-seven percent. In most pools, almost nobody enters draws. That gap between how likely a draw is and how often it is picked is the clearest structural edge available in a pool.
| Match | Home | Draw | Away | Likeliest score |
|---|---|---|---|---|
| Portugal v Spain6 Jul, 19:00 UTC | 23.5% | 34.1% | 42.4% | 0:0 17.4% |
| Algeria v Austria28 Jun, 02:00 UTC | 26.8% | 33.6% | 39.6% | 0:0 16.1% |
| Belgium v Egypt15 Jun, 19:00 UTC | 16.8% | 33.2% | 50% | 0:1 19.8% |
| Paraguay v Australia26 Jun, 02:00 UTC | 33% | 33.2% | 33.8% | 0:0 14.8% |
| Ivory Coast v Ecuador14 Jun, 23:00 UTC | 29.6% | 32.8% | 37.6% | 1:1 14.5% |
| Egypt v Iran27 Jun, 03:00 UTC | 36.9% | 32.7% | 30.4% | 1:1 14.6% |
Reading the bar: home windrawaway winPercentages are the fair market consensus after the bookmaker margin is removed.
One honest caveat, because this is the point where guides usually oversell. Backing draws does not raise your expected number of correct picks: by definition the most likely single outcome is the most likely single outcome. What it does is put you on results the rest of the field does not hold. That is a variance play, and it is the right one only when finishing second is worth the same to you as finishing tenth.
When to follow the favourites and when not to
- Leading, with matches left. Follow the probabilities. Pick the highest of the three every time and let the field beat itself.
- Behind, near the end. You need picks other people do not have. Look for the matches with the highest draw probability, or the underdog with the least hopeless number, and take those. Losing by more does not cost you anything.
- A big pool. The more entrants, the more the winner will be somebody who took low-probability picks and got them right. Copying the consensus in a hundred person pool is a good way to finish comfortably in the middle.
- A small pool. Consistency wins. Take the favourites and stop being clever.
A workable routine
- Read the tendency off the probability split. Take the largest of the three unless you have a deliberate reason to be contrarian.
- Pick a scoreline that matches the tendency and the goal expectation. If over 2.5 is below forty percent, a 1:0 or a 1:1 fits. If it is above sixty, a 2:1 or a 2:0 does.
- Do not chase the perfect score. Your best guess is right about one time in seven at best.
- Decide on draws once, before the round. How many you are willing to enter is a strategy decision, not a match by match feeling.
What none of this does
It does not tell you the result. A ninety percent favourite loses about one time in ten, and that is the number working correctly rather than failing. This page is about entering a prediction pool for fun. It is not betting advice, no bets are taken here or in the app, and nothing on this site links to a bookmaker.
Related