The archive rarely flatters, but Stuttgart arrive armed with 95 matches of memory

The numbers that frame the night

There is an old librarian's habit, when faced with a fixture that does not quite fit the calendar, of checking the shelf twice. So let me check it twice for you. Bayern München, in the Lemeister archive, have 15 recorded matches. Eleven wins, three draws, one defeat, 53 goals scored and 22 conceded, and every one of those matches was played between 2025 and 2026. VfB Stuttgart, by contrast, arrive with 95 recorded matches. Forty seven wins, seventeen draws, thirty one defeats, 183 goals for and 126 against, and that ledger runs from 1984 to this very weekend.

So the archive is lopsided in a way that should make any historian pause. One club is a recent addition to the record books, a precocious newcomer with a perfect little run of form that suggests a machine finely tuned. The other is a veteran of the shelf, dusty and dog-eared, with nearly a century of football folded into its pages. And on Friday evening at the Allianz Arena, these two volumes finally sit side by side.

The Lemeister model, which does not read history the way I do, sees a close contest. Bayern are given a 41% chance of winning, Stuttgart 33%, and the draw sits at 26%. MeisterIQ, the model's confidence in its own forecast, is a modest 52 out of 100. That is not a number that suggests certainty. It suggests a coin that is only slightly weighted, a fixture that could slip in any direction.

I have been asked, across many seasons, to make sense of such numbers. The model looks at recent form, expected goals, squad depth, injury lists, the shape of the last six matches. I look at the archive. And the archive, for all its uneven distribution, tells a story that the model cannot quite capture.

The young giant and the old warhorse

Bayern's 15 matches in the Lemeister archive are a study in controlled dominance. Eleven wins from fifteen is a 73% return, and that solitary defeat is the kind of scar that makes a record human. Fifty three goals in fifteen matches is a shade under three and a half per game, which is the sort of scoring rate that used to belong to video game settings rather than actual Bundesliga football. They have drawn three times, which is worth pausing on. This is not a side that gets beaten often, but it is a side that can be held. Three draws from fifteen suggests that, on their off nights, they meet resistance rather than surrender.

Stuttgart's 95 recorded matches tell a longer, stranger, more human story. Forty seven wins is a 49% return, which is the profile of a club that has never quite been a giant but has never been a passenger either. Seventeen draws, thirty one defeats. One hundred and eighty three goals scored, 126 conceded. That goal difference of plus 57 across more than three decades of football is the signature of a side that tends to outscore its problems rather than suffocate them. Stuttgart do not win pretty, and they do not lose quietly. They are a team that lives in the chaos of open matches, and the archive has the receipts.

The head to head between these two, in the Lemeister archive, is a quiet page. There are no recorded meetings between the two clubs in the database, which is a peculiarity of how the archive has been built rather than a statement about the real world. Bayern's 15 matches have come against other opposition, and Stuttgart's 95 matches have done the same. So I cannot reach for the famous 2024 cup final, or the 2007 title race, or any of the other fixtures that live in the collective memory of German football. The archive is silent on that front, and I must respect its silence.

What the archive does hold is a profile of each side's behaviour across many matches, and that is where the real value lies.

The machine versus the memory

Bayern's recent recorded form, across those 15 matches, has the rhythm of a metronome. They win, they win again, they score in clusters, and then occasionally they stall. The three draws suggest that their attack can be frustrated, that there are defensive shapes which can slow them even when they dominate possession. One defeat in fifteen is a remarkable record, but it is worth remembering that a single defeat in a short sample can be a warning as much as a blemish. Every losing run starts with one bad afternoon, and the model's 41% forecast reflects a certain humility about Bayern's capacity to convert dominance into victory.

Stuttgart's 95 matches, spread across four decades, carry a different lesson. This is a club that has been through relegation battles and European nights, that has seen generations of players come and go, that has learned to survive the gaps between its glory days. The 49% win rate is not the record of a giant, but it is the record of a survivor. And survivors, in my experience, are the most dangerous opponents in a one off match. They do not panic when the machine in front of them hums. They have seen the machine before, in various forms, and they know that machines can stall.

The MeisterIQ score of 52 is the model's way of saying that it is not sure. A score of 70 or 80 would indicate a forecast the model trusts implicitly. A score of 30 or 40 would suggest a coin flip. Fifty two is the sound of a model clearing its throat and admitting that the data points in multiple directions. Bayern's recent form is better, but Stuttgart's historical resilience is a counterweight. The model sees a home favourite, but not an overwhelming one.

The shape of the match

Let me walk you through what the numbers suggest Friday evening will look like, not as a prediction but as a probability map. Bayern will almost certainly dominate possession. Their 15 match record, with 53 goals, indicates a side that creates chances in volume and expects to convert a healthy proportion. The Allianz Arena will be a wall of noise, and the early exchanges will likely see Stuttgart camped in their own half, absorbing pressure and looking for a release valve.

But Stuttgart's 183 goals across 95 matches tell a story of a side that scores wherever it goes. They are not a team that parks the bus and hopes for a point. Their record suggests they will seek to create chances of their own, probably on the counter, probably through quick transitions and direct runs. The 49% win rate is not built on defensive solidity alone. It is built on the willingness to take risks, to leave space at the back in pursuit of goals at the front.

The draw probability of 26% is the third number in this equation, and it deserves more attention than it usually gets. In a match between a machine and a survivor, the most common outcome is often a grinding stalemate. Bayern can dominate and still fail to break through, particularly if Stuttgart's goalkeeper has one of those evenings that become legend. Stuttgart can create chances and still find themselves frustrated by Bayern's defensive record, which has conceded only 22 goals in 15 matches.

The model sees a home win as the most likely single outcome, at 41%, but it does not see it as a certainty. The other 59% is split between a draw and an away win, and that is a substantial chunk of probability. Friday evening is not a coronation. It is a contest, and the archive suggests both sides know how to contest.

The weight of the archive

What does the history actually tell us, once we strip away the recent form and the model's probabilities? It tells us that Bayern München, in the Lemeister archive, are a side that wins often and scores heavily. It tells us that VfB Stuttgart are a side that wins nearly half their matches, scores more than most, and has been doing so for over forty years of recorded football.

It tells us, crucially, that these two clubs have never met in the Lemeister archive. That is a quirk, a gap in the record, and it means I cannot offer you the traditional head to head statistics that usually fill a preview like this. I cannot tell you that Bayern have won the last five meetings, or that Stuttgart have not won at the Allianz since 2010, because those facts live outside the archive I am bound to respect.

What I can tell you is this. Bayern are the favourites, but the margin is slim. Stuttgart are the underdogs, but the model gives them a 33% chance, which is not the chance of a side that expects to lose. The draw is a live possibility, more live than most fixtures of this type. And the MeisterIQ score of 52 is the model's honest admission that this is a match which could go any way.

The archive, in its uneven way, has prepared us for this. Bayern's 15 matches show a side that has not lost often, but has been held. Stuttgart's 95 matches show a side that has rarely been blown away, that has consistently found ways to score, and that has a habit of making matches uncomfortable for more decorated opponents. The young giant and the old warhorse. The machine and the memory.

A quiet forecast, honestly made

I do not deal in predictions the way the model does. I deal in probabilities, in the long arc of the record, in the patterns that emerge from hundreds of matches played across decades. And the pattern here is clear, even if the outcome is not.

Bayern will dominate. Stuttgart will resist. The match will be decided by a single moment, a lapse in concentration, a piece of individual brilliance, a refereeing decision that could go either way. The model's 41% reflects that Bayern are the more likely side to find that moment. The model's 33% reflects that Stuttgart are more than capable of finding it themselves. The 26% draw is the quiet possibility that neither side finds it at all.

The MeisterIQ score of 52 is the model telling us not to get too attached to any one outcome. It is a forecast, not a certainty. It is a map of the territory, not the territory itself. And on Friday evening, the territory will be a green pitch in Munich, two sets of players, and ninety minutes of football that the archive will one day record.

Whenever I write a preview like this, I remind myself that the archive is a living document. These 15 matches and 95 matches are not the end of the story. They are the beginning. Friday evening will add one more line to whatever record exists, and in doing so it will shape the next preview, the next forecast, the next attempt to make sense of a game that resists certainty.

Bayern München versus VfB Stuttgart. 41% against 33%, with a 26% draw sitting between them. A model that is only 52% sure of itself. An archive that is quiet on the head to head but loud on the profiles. Two clubs, one cold and efficient, the other warm and stubborn. Friday evening will tell us which version of history gets written.