Head-to-head records between clubs are quoted constantly and explain very little. The version worth checking is the record between the two coaches, because that one describes something that actually repeats.
A club record has no continuous subject
A run of meetings stretching back a decade involves squads with almost no overlap, several changes of coach at both clubs and playing styles that bear no resemblance to each other.
The only continuous element is the name on the shirt, which does not run, press or defend a corner. There is no mechanism by which those results could influence the next one.
What people are usually noticing is a coincidence that a long enough sequence will always contain somewhere. Given twenty fixtures a season, some pairing will have a striking record.
A coaching matchup does repeat
Two coaches meeting again bring the same core principles, the same preferred structures and the same instincts about when to change a match.
If one habitually sets up in a way that exploits the other's build-up pattern, that advantage carries across clubs and across seasons because it belongs to the method rather than the squad.
This is why a manager can arrive at a new club and continue an unbeaten record against a particular opponent. The relevant variable moved with him.
The signal is still small
Coaches meet each other a handful of times, so even a striking record rests on a sample too small to be confident about.
Methods also evolve. A coach who was repeatedly beaten by a high press three seasons ago may have rebuilt his build-up entirely since.
The record is therefore a prompt to look for a mechanism rather than evidence in itself. If no mechanism is visible, the record is probably noise.
What a real matchup effect looks like
Genuine tactical mismatches are describable. A team that defends narrow struggles against one that overloads the wings, and the pattern is visible in where chances were created.
The test is whether the same shape of problem recurred across the meetings, not whether the results happened to fall the same way.
Where the shape recurs, the record has predictive value. Where the results are similar but the matches looked nothing alike, it has none.
How to use it before a fixture
Check whether either coach has faced the other recently, and if so, what changed between those meetings in personnel and structure.
Then check whether the losing side has since acquired the players or changed the setup that would address the problem.
A record that survives that examination is worth carrying into a preview. Most do not, which is the useful finding rather than a disappointing one.