The Model Likes Aldosivi More Than the Market Does, and That Is the Whole Story

Where the numbers actually stand

Start with the forecast, because everything else is commentary on it. Lemeister's model gives Huracan 42%, the draw 33% and Aldosivi 25%. MeisterIQ, our conviction measure, sits at 53 out of 100. Read that second figure carefully. A 53 is a shrug with a raised eyebrow. It is not a warning siren and it is not a green light. It is the model telling you it has a lean but not a grip.

Now put the market next to it. Market implied probabilities read Huracan 56%, draw 28%, Aldosivi 17%. Line those up against ours and the shape of the disagreement is obvious. The market is more confident in the home side by 14 points than we are, less confident in the draw by five, and less confident in the visitors by eight. The single largest divergence is the away price, and the model's biggest disagreement with the market is away by 8.7 points. That is the number I keep returning to. Eight point seven points on an away side that the market has priced as a long shot is not a rounding error. It is a substantive claim about how this match is likely to be played.

So the piece writes itself in one direction: either the model is overrating Aldosivi, or the market is underrating them. Everything below is an attempt to work out which.

First, a note on what these numbers are. A forecast is a probability distribution, not a prediction of a single outcome. Saying Aldosivi are 25% is not saying they will win. It is saying that if this exact fixture, with this exact information set, were played a hundred times, roughly a quarter of those matches end with an away win. Huracan at 42% are the most likely single outcome, and they are still more likely to not win than to win. Anyone reading 42 as "Huracan should win" has misread the sentence. Anyone reading 25 as "Aldosivi have no chance" has done the same thing in the other direction.

The home side and the weight of a record

Huracan come into this with a long, well documented footprint in the Lemeister archive. Two hundred and twenty nine recorded matches, spanning 2018 to 2026. Eighty one wins. Seventy six draws. Seventy two defeats. Two hundred and thirty one goals scored, two hundred and nineteen conceded. Let us sit with those for a moment, because they are more interesting than they first appear.

Eighty one wins from 229 is a 35% win rate. That is a shade under the 42% we are assigning them on Sunday, which tells you the model is not simply pasting the base rate onto the fixture. It is nudging Huracan up relative to their own long run history, and it is doing so while still refusing to hand them a majority. The draw column is the one that catches my eye. Seventy six draws from 229 is 33.2%. Our Sunday draw number is 33%. That is not a coincidence and it is not laziness. It is the model recognising a side whose historical distribution has a genuinely fat middle. Huracan, across eight years of recorded matches, have drawn about a third of the time. Nudging them away from that would require evidence, and the evidence we have is not strong enough to justify it.

The goal columns are almost perfectly balanced. 231 for, 219 against, a goal difference of plus twelve across 229 matches. That is the statistical signature of a side that has spent most of a decade living in the middle of the table, winning a little more than they lose, drawing a lot, rarely blowing anyone away and rarely being blown away. There is nothing wrong with that. Most clubs in most leagues are that club. But it matters for how you read a 42% home probability. We are not talking about a side with a habit of turning home fixtures into three points. We are talking about a side whose natural state is a tight match that could go either way, and the model is saying Sunday looks like that.

The market, at 56%, is telling a different story. A 56% home probability implies something closer to control, a side expected to take the game to the visitor and finish the job more often than not. Some of that will be situational. Markets price current form, squad availability, travel and the simple gravitational pull of home advantage, and they are usually good at it. But the gap between 42 and 56 is wide enough that it has to be explained by more than the ordinary home bump.

Why the away price is the argument

Here is the case for the model being right, and I will put it plainly. The 8.7 point disagreement on the away side is the largest single signal in this forecast, and the model rarely generates a gap that size without a reason. The most common reason is that the market has priced a favourite at a level the underlying numbers do not support, and the away side has quietly become value in probability terms even though nobody watching the buildup would call them that.

Consider what the market is implicitly saying with Aldosivi at 17%. That is roughly a one in six chance. To believe 17%, you have to believe the visitors' route to a result is narrow: an early goal, a red card, a goalkeeping error, a set piece against the run of play. One in six is the price of a side that needs something unusual to happen. Our 25%, one in four, is the price of a side that has a normal, repeatable path to a result. That is a meaningful difference in kind, not just in degree. It is the difference between "they will need luck" and "they can play their way into this".

Now, the counterargument, and it is a fair one. Markets are not naive. When a price sits at 17% and a model says 25%, the honest first question is what the market knows that the model does not. Team news late in the week, travel arrangements, a manager's press conference hint, a knock picked up in training. Our information set is what it is. MeisterIQ at 53 is the model's own admission of this. A 53 is not a high conviction reading, and I would not dress it up as one. When conviction is middling, the correct posture is humility, not certainty, and that cuts both ways. It means we should not be triumphalist about the away gap, and it also means the market should not be triumphalist about its 56.

There is also the draw, which both sides of the table agree is live. Our 33, the market's 28. The market is five points lighter on the draw than we are, and that five points has to go somewhere. In practice it has gone to the home side, which is the most common direction for it to travel, because markets tend to compress draws toward the favourite when they expect a favourite to impose themselves. Our 33 is closer to the historical Huracan draw rate of 33.2 than the market's 28 is. That is a small point in the model's favour, and small points are what you collect when you cannot collect large ones.

What the model is not saying

I want to be very clear about the limits of this forecast, because this is where most football numbers writing goes wrong. A 42% home probability is a favourite with a minority of the distribution. Huracan are more likely to win than to draw and more likely to draw than to lose, but they are not more likely to win than to draw or lose combined. Forty two against fifty eight. The single most likely outcome, if you must name one, is a Huracan win, but the second most likely outcome is a draw at 33, and the third is an Aldosivi win at 25, and the gap between first and third is seventeen points. Seventeen points is not dominance. It is a lean.

MeisterIQ at 53 reinforces that. If this were a 75 conviction fixture, I would write a different piece. I would talk about a model that had found something the market had missed and was prepared to say so loudly. At 53, the model is saying something quieter: the market is probably a little too confident in the home side, and the away side is probably a little too cheap, but none of this is a strong enough signal to declare the market wrong. It is a signal to watch, not a signal to act on. Nobody should read this piece as a recommendation, and I would be failing at my job if I let anyone infer one. These are probabilities. They describe uncertainty. They do not remove it.

What Sunday actually turns on

So what do we watch for? Three things, and they are all about whether the match takes the shape the model expects or the shape the market expects. First, the tempo of the opening twenty minutes. If Huracan come out pressing high and pinning the visitor, the market's 56 starts to look prescient. If the game settles into a cautious midfield exchange, the draw number climbs in real time and our 33 looks better by the minute. Second, the first goal. In a fixture where the model gives the away side a genuine quarter share, the identity of the opening scorer shifts the whole distribution. An Aldosivi opener does not win them the match, but it moves them from 25 toward something closer to even, and it drags the home number down well below 42.

Third, and this is the one the numbers care about most, how Huracan manage the middle of the second half. Their archive tells you they draw a third of their matches. Draws are not accidents. They are what happens when a side that is marginally better than its opponent cannot convert that margin into a goal, and the clock becomes an ally of the underdog. If this match is level at the seventieth minute, the historical Huracan and the forecast Huracan agree, and the 33 looks less like a number and more like a description.

My honest read, and it is a read rather than a prediction, is that the market is leaning too hard on home advantage and the model is right to push back, but not right enough to be loud about it. The 8.7 point away gap is real and it is the most interesting thing in this forecast. Whether it survives contact with the team sheet is another matter, and MeisterIQ at 53 is the model's way of saying exactly that. Watch the away price. If it shortens through the week, the market is coming round to our view. If it drifts, the model is probably missing something. Either way, this is a 42% favourite, not a sure thing, and anyone who tells you otherwise has stopped reading the numbers and started reading the badge.