Match Watchability Index - Padel API

A tournament day on the professional tour is forty or more matches across five or six courts, most of them starting whenever the previous one ends. Somewhere in there is the one you should be watching. The order of play will not tell you which, and the round names barely help: a round of 32 between two seeded pairs is often a better watch than a semi-final that ends 6-2 6-1. So every match in the API carries a watchability index: an integer from 0 to 100 that answers one question, how much is this match worth watching, before a single point is played.

One number per match

Sort a day, a court or a whole tournament by it and the best matches come out on top.

Known before it starts

It is set as soon as both pairs are known, so it works for previews, alerts and programming.

Never rewritten

Once the match is played the number stays as it was. What you promised is what you can audit.

Where to find it

Every match resource carries a watchability field, and the match list endpoint accepts it as a sort key.

curl 'https://padelapi.org/api/matches?after_date=2026-08-10&before_date=2026-08-10&sort_by=watchability&order_by=desc' \
  -H 'Authorization: Bearer YOUR_API_TOKEN'

{
  "data": [
    {
      "id": 11293,
      "name": "Coello/Tapia - Chingotto/Galan",
      "round_name": "Finals",
      "category": "men",
      "played_at": "2026-08-10",
      "watchability": 93,
      "schedule_label": "Starting at 6:00 PM",
      "court": "Centre Court",
      "status": "scheduled",
      "...": "..."
    }
  ]
}

That single request is the whole feature: the first item is the best match of the day. Narrow it with the filters the endpoint already accepts, all of which combine with sort_by=watchability:

What you want	Add this
The best match of today	`after_date` and `before_date` set to the same day
The best women’s match	`category=women`
The best match on a given court	`court=Centre Court`
The best match of the weekend	a two or three day date range
The best match of a round	`round=4` for quarter-finals

You can ask the same question in plain English through Padel Analyst: “which is the best padel match today?” or “what is the most interesting women’s match this weekend?”. The MCP match search tool takes sort_by=watchability too.

How to read the number

The index is ordinal: it exists so you can rank matches against each other. It is not proportional, so an 80 is not “twice as good” as a 40, in the same way a 200 point Elo gap is not “twice” a 100 point one.

Value	What it usually looks like
90 to 100	Exceptional. Two of the best pairs on tour, evenly matched, in a final or a semi-final
80 to 89	Rare. Two strong pairs late in a draw: the fixture you would pick if you could only watch one
60 to 79	A strong match: elite pairs in a final or semi-final with one side favoured, or a very even quarter-final just below the top
40 to 59	The solid middle of the draw. Even matches between main draw regulars, or good pairs meeting early
Below 40	Predictable, or early, or both. A seeded pair against a qualifier, a lopsided round of 32

Most matches land in the middle two rows, which is the point: the top of the scale is meant to be scarce. Most tournaments produce nothing above 80 at all, so treat the top band as an exception rather than a target to filter on. For a concrete anchor, take the 2026 London P1 as the API recorded it. The men’s final scored 93 and the women’s final 91. The winning pair’s own quarter-final, two rounds earlier against a much weaker side, scored 50 — the same players, forty-three points lower, because the opponent and the round both changed.

Values are comparable across categories, tournaments and seasons, because the ingredients are. What they are not is a measure of quality of play: they rate the fixture, not the ninety minutes that follow.

What goes into it

Three things, and nothing else. Each is a number the API already publishes, so the index is auditable rather than editorial.

How good the players are

The average Elo rating of the four players on court, read against the range the professional tour actually occupies. This is the heaviest ingredient: a final between two mid-table pairs is not the same event as a final between the top two.

How close it should be

The calibrated win probability for the matchup. A 50/50 contributes everything it can, a 95/5 contributes almost nothing. This is what stops a superstar-against-a-qualifier fixture from ranking above a real contest.

What is at stake

The round. A final counts for more than a semi-final, which counts for more than a quarter-final, down to the opening rounds. Enough to break ties between comparable matches, not enough to push a dull final above a great quarter-final.

Notice what the three do together. Level alone would just rank the seeds, closeness alone would put two even qualifiers above every final, and the round alone is the draw sheet you already have. A match has to score on more than one of them to reach the top of a day.

This is a pre-match index, which is what makes it useful for programming. It is not the same family as the excitement indexes published in other sports, which are computed after the fact from how the win probability swung during play. Those tell you what a match turned out to be. This one tells you what to watch tonight.

When it is set, and when it stops moving

It appears when the line-up is complete

A match needs four known players to be rated. Bracket slots waiting on a previous result carry null, and they pick up a value as soon as both pairs are decided.

It follows changes to the draw

If a pair changes before the match is played, through a withdrawal, a replacement or a re-published draw, the index is recalculated for the new matchup.

It freezes when the match is played

Once there is a result, the value stops moving, permanently. The ratings behind it keep evolving, but the match keeps the number it had going in. That is deliberate: a number you published on Friday still means the same thing on Monday, so archives and back-tests stay honest.

Unrated matches sort last

Matches without a value never crowd the head of a list. sort_by=watchability&order_by=desc gives you the best matches first and the undecided slots at the end.

What it does not know

The index is built from level, closeness and round, so anything outside those three is invisible to it:

Rivalry and narrative. A grudge match, a former partnership meeting for the first time, a player’s last tournament. None of it reaches the number. If your audience cares, layer head to head records and pair stats on top.
Home players. A local wildcard in an opening round can be the match of the day for that crowd and rate in the twenties.
Form and fitness. Elo does not decay with inactivity and does not know who retired mid-match last week. See what the prediction model does not know.
New pairs. Ratings are individual and averaged, so a pair playing together for the first time reads the same as one with two seasons of joint titles.
What actually happened. A 90 can end 6-1 6-2 and a 30 can go to a third set tie-break. Padel still gets to decide.

Treat it as a shortlist, not a verdict. Its job is to cut forty matches down to three worth a human look, and it is very good at that.

Availability

The watchability field is available on paid plans. On the Free plan it is returned as the string hidden_free_plan on every match, so parse it expecting an integer, null or that marker.

What people build with it

Match of the day widgets

One request, one match, rendered on a homepage or a newsletter without an editor picking it.

Push alerts

Notify users only when a match above a threshold is about to start, instead of on every fixture.

Stream and watch guides

Rank the day’s order of play so a viewer with two hours knows where to spend them.

Automated previews

Index plus probability plus head to head is enough for a publishable preview of the right match.

A practical pattern for a live product: sort the day by watchability, take the top few, then use the estimated start times to know when each one goes on court and the WebSocket channel to follow it point by point.

Frequently asked questions

How do I find the best padel match of the day?

Request the day’s matches sorted by watchability in descending order and take the first result: GET /api/matches?after_date=2026-08-10&before_date=2026-08-10&sort_by=watchability&order_by=desc. The index combines how good the four players are, how even the matchup is expected to be and which round it is, so the top of that list is the day’s most attractive fixture.

Which padel match should I watch today?

The highest rated one that fits your schedule. Anything above 80 is a rare fixture worth rearranging your evening for, the 60s and 70s are the strong end of an ordinary day, and below 40 you are usually watching a formality. Cross the list with the estimated start times to pick what is actually playable in your day.

What is a watchability index?

A single number that rates how much a fixture is worth watching before it is played, built from the strength of the players involved and how competitive the match is expected to be. Other sports publish similar indexes for weekly schedules. This one runs on Elo ratings and calibrated win probabilities, on a 0 to 100 scale, on every professional padel match in the database.

Is it available before the match starts?

Yes, that is the point. It is set as soon as both pairs are known, which for most matches is a day or more before they go on court, and it is fixed by the time play begins. It is never computed after the fact from the result.

Does a high value guarantee a good match?

No. It rates the setup: two strong pairs, evenly matched, late in a draw. A great fixture can still end in straight sets. Over many matches the high values are clearly better watches than the low ones, which is all an ordinal index claims.

Why is the value of a played match different from what I would compute today?

Because it was frozen when the match was played, using the ratings as they stood then. Recomputing it now with today’s ratings would give a different number and would quietly rewrite history, so we do not. Historical matches keep the value they carried into the court.

Does it work for women's padel?

Yes, and values are directly comparable between categories. The ratings behind it are computed the same way for both, and the round and closeness components do not depend on the category at all.

Why does a match have no value?

Because its line-up is not complete. Bracket slots still waiting on an earlier result have no four players to rate, so they return null until both pairs are decided, and they sort to the end of a watchability-ordered list.

Can I filter by value instead of sorting?

Not directly, but sorting plus pagination covers it: request the range you care about ordered by watchability descending and stop reading when the values drop below your threshold.

Next steps

List matches endpoint

Full reference: date, round, category and court filters, plus sort_by=watchability.

Padel Elo rating

The player level that feeds the index, and how to read its scale.

Match prediction model

The calibrated win probability behind the closeness component.

Match schedule estimates

When the match you picked actually goes on court.

One number per match

Known before it starts

Never rewritten

​Where to find it

​How to read the number

​What goes into it

​When it is set, and when it stops moving

​What it does not know

​Availability

​What people build with it

Match of the day widgets

Push alerts

Stream and watch guides

Automated previews

​Frequently asked questions

​Next steps

List matches endpoint

Padel Elo rating

Match prediction model

Match schedule estimates

Where to find it

How to read the number

What goes into it

When it is set, and when it stops moving

What it does not know

Availability

What people build with it

Frequently asked questions

Next steps