Seeding exists to protect the draw, not the favourites

What an unseeded bracket would do

A purely random draw would regularly place several of the strongest entrants in the same quarter, which would eliminate them against each other before the rounds that carry the most attention. The tournament would still produce a champion, but the closing days would frequently feature players who had faced a far weaker path than others knocked out early. Seeding prevents that by distributing the highest-ranked entrants across separate sections, so the bracket collapses towards a small group of leading players rather than an arbitrary one.

This is a structural objective rather than a favour, because the value of a tournament rests heavily on the quality of its final rounds. The design is explicit about it, which is why seeding positions are fixed by rule and only the placement within a section is left to chance.

How seeds are placed inside the bracket

The top two seeds are separated into opposite halves so they can only meet in the final, and the next two are placed to fall into separate quarters. Beyond that, seeds are grouped into bands and drawn randomly within their band, which preserves the separation principle while keeping the draw genuinely uncertain. Unseeded entrants are then distributed across the remaining positions, so a dangerous player outside the seeding list can land anywhere in the bracket.

That residual randomness is what produces the early upsets spectators enjoy, and it is deliberately left in rather than engineered out. A fully deterministic bracket would remove the draw ceremony entirely and turn the tournament into a ladder, which is a different competition format with different appeal.

When ranking and seeding come apart

Some events adjust their seeding list to account for a surface, on the reasoning that a general ranking accumulated across different conditions is a poor predictor on one of them. The adjustment usually weights recent results on the relevant surface more heavily, which promotes specialists and demotes players whose ranking was built elsewhere. This is defensible as a way of making the seeding do its job better, since the objective is to protect the later rounds rather than to reward accumulated ranking points.

It is also contestable, because a formula that reorders players introduces a discretionary element into what is otherwise a mechanical process. The debate is really about whether seeding is a reward for the season so far or a tool for building a coherent bracket, and the two answers point in different directions.

Byes, wildcards and the edges of the field

Where the field is not a clean power of two, the highest seeds receive byes into the second round, which shortens their path and is defended as a reward for ranking. Wildcards fill a small number of places at the discretion of the organiser, typically for returning players, promising locals or entrants whose ranking understates them. Both mechanisms are exceptions to a merit-based entry list, and both are justified by the tournament interest in a compelling field rather than a strictly ordered one.

The tension is obvious, since every wildcard displaces a player who earned entry through qualifying or through ranking accumulated over a full season. Organisers manage the criticism by keeping the number small and by publishing criteria, which converts an arbitrary decision into a constrained one.

Protected rankings and the returning player

A player absent for a long period through injury would normally lose a ranking built over years, which would force a return through the lowest levels of the calendar. Protected ranking schemes allow a limited number of entries at the ranking held before the absence, which lets a returning player rebuild without starting from nothing. The entitlement is capped in both time and number of uses, because an unlimited protection would let a player bank a peak ranking indefinitely.

Protected entries usually do not carry seeding, which is the compromise that admits the player to the draw without distorting the bracket around a result long past. The distinction between entry and seeding is the cleanest illustration that the two mechanisms exist for entirely different reasons.