Spain vs Belgium prediction: the Jarvis supercomputer's verdict

Who does the model lean towards in Spain vs Belgium?

Who wins when Spain face Belgium at the World Cup? If you want a quick answer before reading deeper, here it is straight from the Jarvis supercomputer model: Spain are favoured with a 59% chance of winning, 25% for the draw and 16% for Belgium. The score that best represents the model's tilt is 2–0. But there is one important asterisk I'll keep coming back to in this piece: the confidence on this read is only low. That means the model does lean, but it doesn't lean all the way, and the three result scenarios are more open than you might assume.

The main read: what the numbers say

The picture Jarvis paints is fairly clear in the summary panel: Spain are the higher-rated side, at 59% against Belgium's 16%. That gap is not small; it's almost four times the raw win probability. But the 25% draw is the detail I really want you to keep an eye on, because a one-in-four chance of the game finishing level is a big enough number not to brush aside. On goals, the expected total the model produces is around 2.69, which tells me this isn't a dull one-way scenario but a match that could see goals at both ends. The 2–0 predicted score should be read as the direction of the model's lean, not the single most likely scoreline.

The detail few people notice in these numbers

What I think few people notice is the interesting tension inside these numbers. On one hand, the model settles on a predicted score of 2–0, hinting at a clean win for Spain. On the other, the chance of the match producing three or more goals is pegged at exactly 50%, level with the chance of fewer than three. Those two signals aren't contradictory once you understand that 2–0 is only the midpoint of a wide distribution: the model sees both a Spain win by a margin and an open, goal-friendly game. Add the low confidence, and the real message from Jarvis is not to lock too tightly onto a single result.

Spain's form: five World Cup games without conceding

Spain come into this one with a rare statistic: five matches at the World Cup and not a single goal conceded. In the group stage they drew 0–0 with Cape Verde, beat Saudi Arabia 4–0 and then edged Uruguay 1–0. In the knockout rounds they beat Austria 3–0 before getting past Portugal 1–0. Four wins, one draw, nine goals scored and a zero in the goals-against column.

That record sits neatly with the 59% win probability Jarvis gives Spain, and with the 2–0 predicted score, a win without conceding. But 59% also means that more than four times in ten the result won't land in their win column. The goalless draw with Cape Verde in their opener is a reminder that a watertight defence doesn't win games on its own, and a reason not to turn a lean into a certainty.

Belgium's form and the chances that remain

Belgium have taken a very different road. They opened the tournament with two draws, 1–1 against Egypt and 0–0 against Iran, before bursting into life with a 5–1 win over New Zealand. In the knockout rounds they needed extra time to beat Senegal 3–2, then beat the United States 4–1. Across those five games they scored 13 goals but also conceded five.

The 16% win probability shows Jarvis placing Belgium in the lower-rated bracket for this match. But 16% isn't a number to cross them off with. Combine their win chance with the draw chance and Belgium still hold a meaningful share of scenarios in which the game doesn't go Spain's way. They have scored 12 goals in their last three games, and with expected total goals near 2.69, the pitch stays open enough for one Belgium moment to shift the picture. The lower-rated side is not the same as a side with no chance of springing a surprise, and it's a point I'll be watching closely.

Head-to-head: Spain have won the last five meetings

The head-to-head record is heavily one-sided. The most recent meeting was a friendly in 2016, when Spain won 2–0 in Belgium. Before that, in qualifying for the 2010 World Cup, they won 2–1 on Belgian soil and 5–0 at home, and in qualifying for the 2006 World Cup they won both legs by the same 2–0 scoreline. Across those five games Spain scored 13 goals and conceded just once.

That said, the last meeting was almost ten years ago, so I treat the head-to-head as background reference. The main weight still sits with the current Jarvis read: a 59% lean towards Spain, but only at low confidence.

Total goals: a defence yet to concede against an attack with 13 goals

The over/under chart from Jarvis shows the tug-of-war mentioned above: the chance of three or more goals sits at 50%. The two teams' last five games pull in opposite directions too. Spain's matches produced nine goals in total, all of them scored by Spain, which is fewer than two per game. Belgium's matches produced 18, 13 scored and five conceded, which is more than three per game.

One side is used to keeping games tight, the other to open, end-to-end contests, and Jarvis's expected total of around 2.69 sits almost exactly between those two extremes. Whether this match goes the low-scoring or the high-scoring way will depend on which team imposes its own tempo on the other.

Score prediction and three scenarios

So how do I read this match in the end? The score that best represents the Jarvis tilt is 2–0 in Spain's favour, and I'll use it as a reference point to sketch three scenarios.

The first scenario, and the one the model leans towards: Spain control the game and win by a margin, somewhere around 2–0. The second scenario, coming from that far-from-small 25% draw: the sides share the points with a level scoreline such as 1–1, especially as expected total goals near 2.69 suggest goals could arrive at both ends. The third scenario, Belgium's 16%: the lower-rated side use an open game to snatch a surprise result. With low confidence and a 50% chance of three or more goals, I lean slightly towards Spain but leave all three scenarios open. This is statistical analysis, not betting advice.