Explainer
ATP rankings explained, and where the WTA rules differ
How tennis rankings are really calculated: the rolling 52-week window, the best-of-N count, what defending points means, and how the draw follows from it.
By CricketTaken EditorialPublished Explainer19 min read
A player loses in the quarter-finals on Friday and climbs four places on Monday. The one who won the title drops a place. Nothing has been miscounted, and the tournament that has just finished explains almost none of it.
Most attempts at getting the ATP rankings explained stop at one sentence: points from the last 52 weeks, best results counted. That sentence is true and it leaves out the two things that actually move the number, which are what fell off the back of the window and what was already sitting in the slot the new result had to displace.
The ranking is not a scoreboard that ticks upwards. It is a ledger with a moving twelve-month boundary, and on any given Monday a player's total is what arrived minus what expired. Once that is clear, the rest of the system stops looking arbitrary: the mandatory tournaments, the two competing lists, the protected rankings, the reason the top spot can change hands while both players are asleep.
The ranking is an entry queue before it is a merit table
Tennis has no league. There is no fixture list handed down from above, no division a player is promoted into, and no season-long competition that everybody enters by default. There is a calendar of independently promoted tournaments, most of them owned by somebody other than the tour that sanctions them, played by self-employed athletes who choose their own schedules and are not obliged to turn up anywhere.
Every one of those tournaments has to settle the same two questions weeks before a ball is hit. Who is in the draw. Who plays whom.
The ranking exists to settle them without an argument. It is a queue, ordered by a published formula, which the tours, the majors, the promoters and the players have all agreed in advance to accept. Its other job, the one the public cares about, is to say who is having the best year. Almost every design decision serves the first job, which is why several of them look wrong if you assume the system is only a list of who is good.
Hold onto that through everything below. A ranking built purely to rate players would fade old results gradually, weight the strength of the opponent, and keep the surfaces apart. Tennis does none of those three things. It does not do them because a tournament director filling a draw six weeks out needs a number that is simple, published, unarguable and calculated identically for a teenager from Perth and a former champion from Michigan. Nothing else in the way professional tennis is organised can do that job.
- 52Weeks a result stays on a player's ranking
- 52Ranking lists published in a year
- 4Grand slams inside the fixed part of the count
- 2Parallel systems, the rolling ranking and the year-to-date race
ATP rankings explained: why it is a window and not a season
The window is the first thing people get wrong, usually by assuming it works like a league table that empties at the end of the year.
It does not empty. It slides. Each Monday the tours recompute every player's total from the tournaments played in the previous 52 weeks, which means one week's worth of year-old results drops out at the same moment the current week's results go in. The list is not a record of the season. It is a photograph of a twelve-month period that ends on the Sunday just gone.
There are three good reasons for building it that way, and only the first is obvious.
The first is administrative. Entry lists for January's tournaments are drawn up in November and December. If the ranking reset on 1 January, the sport would spend its opening weeks with either no working list or a list with nothing on it, and the first major of the year would be seeded off a fortnight of results. The rolling window means there is always a full, current, twelve-month list available on any Monday of any month, including the ones with no tennis in them.
The second is about fairness of sample. Fifty-two weeks is exactly one trip round the calendar, so every player has been offered the same menu: the same majors, the same surfaces, the same clay swing, the same indoor autumn. Shorten the window to six months and you would be ranking a clay-court specialist on their hard-court season, which is not a measurement, it is a joke. The length is not arbitrary. It is the shortest period that is surface-complete, and it is one of the few places where the design quietly acknowledges that surfaces change the sport enough to create specialists.
The third reason is political. A rolling window means there is never a moment when the ranking is meaningless, and therefore never a moment when a tournament can argue it should be allowed to pick its own field.
What the window does not do is fade anything. This is the part worth dwelling on. A result is worth one hundred per cent of its value for fifty-one weeks and nothing at all in week fifty-three. There is no decay curve. A semi-final from eleven months ago counts precisely as much as a semi-final from last Tuesday, and then one Monday it counts for nothing whatsoever. Other ranking systems handle this differently, and it is instructive to see how: the world athletics system averages a set of best performances and applies a sliding discount as a result ages, so a rating drifts downwards rather than falling off a shelf.
Tennis chose the shelf. The cost is that the list is jumpy and lags form badly. The benefit is that anyone with a calculator can reproduce it, and nobody has to argue about the shape of a decay curve in a tournament director's office.
There is a timing detail underneath this that even engaged fans get wrong. Points do not expire on a strict anniversary date. They expire when the current year's edition of that tournament is complete, which is how the tours keep the calendar aligned when an event moves by a week or two. If a tournament is not held at all in a given year, the rulebook determines when the old points come off instead, and if a tournament moves several weeks, the drop moves with it. This is not a footnote. It is the difference between a projected ranking in April being right and being nonsense.
The count: best of N, and the moment an extra tournament stops helping
A ranking total is not the sum of everything a player has done. It is the sum of their best results, counted up to a fixed maximum number of tournaments.
Both tours use a cap. The two caps are different, both have been changed more than once, and the current figures are in the rulebooks. What matters far more than the exact number is the shape it produces, because the cap is what stops the ranking from becoming a measure of frequent flyer miles.
The count is not a free choice of a player's best results, either. It is partly reserved. The four majors and the top tier of tour events occupy fixed slots, and only the remaining slots are filled by whatever else a player has done. So a top player's ranking is mostly a report on how they got on at about a dozen specific tournaments they were required to enter, plus a handful of results from wherever else they chose to play.
Now the arithmetic that almost no explanation bothers with.
Once every slot in the count is full, a new result does not add its face value. It has to displace the smallest result currently counting, and the ranking moves by the difference between the two. A player whose weakest counting result is worth ninety points gains nothing at all from a tournament that pays them fifty, and gains one hundred and sixty from a title that pays two hundred and fifty.
Read that third pair of bars again, because it is the whole economics of a top player's schedule. For someone whose count is full of majors and top-tier results, an early loss at a small tournament is worth precisely nothing, while the flight, the fatigue and the risk of tearing something are real. The system does not merely permit the best players to skip small events. It makes skipping them the rational choice, and the tours then have to compel attendance by other means, which we will come to.
Turn the same rule upside down and you get the other end of the game. A player ranked outside the top two hundred has empty slots. Nothing is being displaced, so every point they win is additive, and a decent week at a small event moves them dozens of places. That sounds encouraging until you notice the loop it creates: a low ranking restricts you to tournaments that pay few points, and few points restrict your ranking, and your ranking restricts your tournaments. Climbing out means winning consistently at a level where the prize money frequently does not cover the travel. The ranking is self-correcting at the top and self-reinforcing at the bottom, and that asymmetry is the biggest structural reason the professional game narrows so sharply below the main tours.
Mandatory, optional, and why the ranking is really a calendar contract
The tours sell their biggest tournaments on a promise: the best players will be there. The awkward part is that the players are independent contractors who cannot be ordered to play anywhere.
So the tours use the only lever that reaches every player at once. They make the top-tier events mandatory for anyone ranked inside a threshold, and they enforce it through the ranking itself. A missed mandatory tournament does not simply go uncounted. It occupies one of the counted slots at zero, which means it displaces a result the player would otherwise have banked. The penalty is not a fine deducted from a total. It is the loss of a slot, and against a full count that is worth more than a fine.
That distinction is the entire mechanism, and it is why the exemption list matters so much. Both tours grant relief for genuine injury, and both grant a form of long-service exemption to players who have been on tour for many years and played a great many of these events, on the reasonable view that a thirty-four-year-old body should not have to play every hard-court fortnight to keep its ranking honest. The precise thresholds and the precise treatment of a missed event have been revised repeatedly by both tours and live in their rulebooks rather than being stable enough to quote.
Above the mandatory tier there is a further layer of obligation. The ATP requires its highest-ranked players to enter a set number of the tier below as well, with at least one of them falling late in the season, which exists to stop the autumn events being abandoned once the majors are done. The WTA operates an equivalent commitment further down its own tier list. There is also one long-standing exception at the top level: the Monte Carlo tournament has for many years carried top-tier points without being compulsory, a survival of the sport's older governance that nobody has managed to tidy away.
Put the mandatory structure together with the best-of-N count and you can see what the ranking really is. It is not principally a rating. It is the enforcement clause of a calendar, and the reason the tennis season is as crowded and as strangely shaped as it is is that the ranking makes roughly a dozen fortnights unskippable, then leaves players to fill the gaps between them with whatever their bodies can stand.
How the points scale, by round and by category
Two scales run at once. One sets what a tournament is worth. The other sets how that value is shared out among the players in it.
The first is easier than people expect, because the tiers are named after it. At tour level, the number in the category name is the champion's points: a 500-level event pays its winner five hundred, a 250 pays two hundred and fifty, and the top tier pays a thousand. That single convention does most of the work, and it is worth knowing because it turns an apparently opaque system into something you can read straight off a schedule. The majors sit above the top tier at a figure published by the tours, and the season finale is scored differently again, awarding points for each round-robin match won and then for the knockout rounds, so an undefeated champion there banks a total no ordinary tournament can match while a winless qualifier banks almost nothing.
The second scale, the one that shares out a tournament's value by round, is steeper than most people picture. Going backwards from the title, each round is worth something in the region of half to two thirds of the round in front of it. The champion earns far more than the runner-up, the runner-up far more than the beaten semi-finalists, and by the time you reach the opening rounds of a large draw the amounts are close to rounding errors. Winning a tournament outright is worth substantially more than a season of respectable third-round exits.
Draw size then decides how many rungs the ladder has. A larger draw does not raise the top of the scale, it adds rounds at the bottom, which is why reaching the third round of a big event and the quarter-finals of a small one can be worth similar amounts. Where a draw is not a power of two, the leading seeds receive first-round byes. A bye pays nothing by itself. What it does is deliver a seeded player straight into the round where the points begin, which is a substantial competitive advantage disguised as a scheduling convenience.
Qualifying has its own small table. Winning through qualifying pays something, losing in the final qualifying round pays a little less, and a qualifier who then loses in the first round of the main draw has had a hard week for a total that a player two tiers down might have collected by winning three matches. None of this is unfair exactly, but it explains why the calendar is full of players making decisions that look eccentric from the outside.
What defending points actually means
This is the phrase that does the most work in tennis commentary and gets the least explanation.
A player is said to be defending points when the tournament taking place this week is one where they earned points in last year's edition, and those points are now at the end of their fifty-two weeks. The word implies a contest against their own past self, which is roughly the right feeling and exactly the wrong mechanism.
The old points are not at stake. They are already gone. They expire when this year's edition finishes, and nothing the player does can preserve them. Entering does not save them. Winning does not save them. A player who withdraws with a broken wrist loses precisely as much as one who plays and loses in the first round.
What actually happens is a subtraction followed by an addition, and the net change is this year's result minus last year's.
- Monday, before the tournamentThe total reads 4,000. Inside it sits a 500-point block earned by winning this event last year, which has now reached the end of its 52 weeks.
- The old points are already condemnedNothing this week can save that block. It expires when the current edition finishes, whether the player enters, withdraws, or never intended to play at all.
- The draw is madeEntry was decided by the ranking on a cut-off date weeks earlier, and seeding by a later snapshot. Neither is the total the player will hold on Monday.
- The player loses in the second roundThis year's edition pays a small figure for reaching that round. Call it 50 in this constructed example.
- The tournament finishes on SundayThe expiring block and the new result are applied in the same update, not one after the other.
- Monday, after the tournament4,000 minus 500 plus 50 leaves 3,550. The player has fallen 450 points in a week in which they won a match.
Constructed example. The player starts the week on 4,000 points, of which 500 came from winning this same tournament a year ago. Every figure is invented to make the arithmetic legible.
Run the same week with different results and the shape of the system becomes clear. Win the title again and the player collects five hundred, refills the hole exactly, and finishes the week on the same four thousand they started with. They have defended perfectly and gained nothing. Reach the final for three hundred and they still finish two hundred down, having played better than almost everyone in the draw. Skip the tournament and they finish on three thousand five hundred.
There is only one way for a defending champion to gain points at their own tournament, which is to have lost points somewhere else. The best possible outcome of a title defence is to stand still.
That single fact drives more scheduling behaviour than any other. A player whose previous season was heavily front-loaded faces a punishing spring in which the number falls almost every week regardless of how well they play, while a player who was injured until August has nothing to defend until August and can climb by doing very ordinary things. Neither is a statement about current quality. Both will be described as form.
It also produces a genuinely strange incentive. Playing a tournament you won last year is a bet with limited upside and a long way to fall, whereas playing one you missed last year offers addition and nothing else. Players and their teams read the twelve-month-old calendar as carefully as the current one, because the schedule of expiries is public, dated and known a year in advance.
Why the number one ranking can change hands when nobody wins anything
Follow the subtraction and the paradox dissolves.
Suppose the leader is defending a large result this week and the player behind them has nothing on the board from the corresponding week a year ago. The leader plays and loses early, or does not play at all. Their total falls by whatever expired. The challenger's total does not move, because there is nothing to expire and nothing being added. On Monday the order swaps.
Nobody won a title. Nobody had to be in the same country. One number went down and the other stayed still, and the sport has a new world number one.
The same arithmetic works in reverse and rather more cruelly. A player can put together the best fortnight of their career and still slip a place, because the rival above them had a bigger week or a smaller expiry. The ranking is a difference between two moving quantities and only one of them is under any individual player's control.
This is also why speculation about somebody losing the top spot at a particular tournament is not really speculation. The expiry schedule is arithmetic anyone can do weeks ahead. Both tours publish the totals, and the points coming off in any given week are simply last year's results at that week's tournaments. Where two players finish level, the tiebreak is a rule in the book rather than a matter for debate, which is the same principle as everything else here: the system would rather be predictable than clever.
Two lists at once: the ranking and the race
Running alongside the rolling ranking there is a second table, and confusing the two accounts for a good share of the arguments on tennis forums.
The rolling ranking is the working document. It fills draws, sets seeds, decides who gets into qualifying, and is the number quoted next to a player's name.
The race is a calendar-year table. It resets to zero on 1 January, accumulates only what has been won since, and its job is to decide which eight singles players and which eight doubles teams qualify for each tour's season finale. The ATP names its race after whichever city is hosting the finals that year, and the WTA runs the equivalent for its own championship.
The reason for having both is straightforward once stated. A tournament billed as gathering the best players of this season cannot select its field from a window that still includes last November. The race exists so that qualification for the finale measures the season the finale is about.
The two lists converge in the autumn. By the closing weeks of the year, the rolling fifty-two weeks and the calendar year cover very nearly the same set of results, so the tables end up describing the same thing. That is why the year-end number one is the player who tops the race, and it is why year-end number one and weeks at number one are genuinely different achievements. The first is a calendar-year competition. The second is a count of Mondays on which a rolling twelve-month total happened to be the largest. A player can accumulate a great many of the second without ever winning the first.
Where the ATP and WTA rules diverge
The two tours are separate organisations with separate rulebooks, separate boards and separate commercial interests, and neither of them owns the four majors, which run to their own book under the international federation. That the two ranking systems look as similar as they do is the product of convergence rather than common authorship.
They agree on the fundamentals. Fifty-two weeks. A cap on counted results. Publication every Monday. Fixed slots for the majors and the top tier. Points awarded by round reached. A calendar-year race for the finale. Anyone who reads across from one to the other will not be lost.
They differ in the details, and the details are where people trip up.
The caps are different numbers, and both tours have moved theirs. The tier structures are not identical either: the WTA's top category comes in two draw sizes with different points tables, so two tournaments carrying the same category name can be worth materially different amounts, which has no clean equivalent on the ATP side. Below tour level the ladders diverge again, the ATP feeding through its Challenger circuit and the WTA through its own second-tier events, with different points and different entry rules.
The commitment obligations are shaped differently as well, because the two calendars are shaped differently and because the two tours have different amounts of influence over their own events. The terminology diverges too: what the ATP calls a protected ranking, the WTA calls a special ranking, and the two are similar instruments with different limits.
And there is one substantive divergence with no counterpart at all, which is the WTA's provision for players who take time out to have children.
The honest advice for anyone who cares about a specific figure is to read the current rulebook rather than any article, including this one. The architecture described here has been stable for decades. The numbers inside it change most years.
Protected rankings, special rankings and the maternity provisions
A long injury does something to a tennis ranking that it does not do in a league sport. The player does not decline. They evaporate. Each Monday another week of results expires with nothing arriving to replace it, and after fifty-two weeks away a former top-ten player has no ranking at all, because a ranking is a statement about the last twelve months and they have not had a last twelve months.
Then they recover, and hit the queue problem. Entry is by ranking. They have none. Without some provision they would be reduced to asking tournaments for wildcards, which is a fine outcome for a former champion and no outcome at all for a good player who was never famous.
The protected ranking, or special ranking, is that provision. A player absent for a qualifying period, defined in months in each rulebook and long enough that it cannot be gamed by a short lay-off, may apply for a number derived from where they stood when the absence began.
The important part is what it is not.
It is not a rank. It never appears on the published list, it does not seed the player, and it does not restore a single point to their total. Their actual ranking remains whatever the empty year left it, and if they want it back they have to earn it again from wherever the window has dumped them. What the protected number does is precisely one thing: it lets them into draws, for a limited number of tournaments, inside a limited window measured from their return.
Those limits are not meanness. Every place granted on a protected ranking is a place taken from somebody who earned it in the last fifty-two weeks, and that somebody is usually ranked around a hundred and fifty and living on the margin. An unlimited protection would let a returning name sit at the front of the queue indefinitely. A protection that is too short returns a fit player to a schedule of small events that cannot repair the ranking fast enough to matter. The rules are an attempt to price that trade-off, they are set out plainly in both books, and they have been adjusted several times as players have pointed out where the pricing was wrong.
Which is exactly what happened with the maternity provisions. A player returning after childbirth was for years treated under the injury rules, which suited almost nobody: the absence is longer, it is planned rather than sudden, and it is not an injury. The WTA now operates a distinct provision with a materially longer window, measured in years rather than months, and its own allowance of tournaments.
The seeding question turned out to matter more than the entry question, and it is the part that took longest to fix. A returning player entering a large draw unseeded is likely to be drawn against a top seed in the first round, lose, earn nothing, and go back to the bottom of the queue. Entry without protection from the seeds is close to a formality. The WTA has moved on this, and has separately introduced a paid maternity leave programme funded through a tour partnership, which is a different thing altogether from the ranking rule and is regularly confused with it. One is money. The other is a place in a draw. A returning player needs both, and until recently the tour offered neither.
How the ranking runs the draw: entry, cut, seeds, byes and wildcards
Here is where the ranking stops being a talking point and starts being an operating system.
The entry list is built on a snapshot, not on today. Every tournament has a published entry deadline weeks before the event, and the field is taken from the ranking list as it stood on that date. A player who climbs fifty places in the intervening month does not get in on the strength of it. They are on the alternate list, hoping for withdrawals, or hoping for a wildcard. This single fact explains a great deal of apparently irrational scheduling: players enter tournaments they do not want to play because the deadline forces the decision long before they know how their body or their draw will look.
The cut is where the list runs out. Direct acceptances fill the main draw from the top of the entry list downwards until the places are gone. The player one place below the cut is an alternate and will play if somebody pulls out in time. There is also a special exempt place, reserved for a player still competing in the previous week's tournament who could not therefore appear in this week's qualifying, a small mechanism that exists purely because the calendar overlaps itself.
Qualifying is a second, smaller queue. It has its own entry list built the same way, its own cut and its own draw, and the winners join the main draw. If a main-draw player withdraws after the draw is made, a beaten qualifier takes the place as a lucky loser, chosen by ranking or by lot depending on the rules in force.
Seeding uses a later snapshot than entry. Seeds are set from the ranking closer to the event, usually the Monday of the tournament or the week before, so a player can be admitted on one week's list and seeded on another. The seeds are then placed by rule: the top two are separated into opposite halves, the next two are drawn into the remaining quarters, and so on down in bands, with placement inside each band decided at random. Being seeded fifth does not buy a defined path, only a guarantee about which opponents cannot be met before a given round. The majors add their own layer on top of that, which is a subject of its own and worth reading separately in how the four majors decide their seeds.
Byes go to the seeds when the draw is not a power of two. No points, no match, and a free pass into the round where the scoring starts.
Wildcards are the deliberate hole in the system. A tournament may award a capped number of main-draw and qualifying places to whoever it likes: a local teenager, a returning champion, a name that sells tickets, and some of the majors operate reciprocal arrangements to send each other's players into their draws. Every wildcard is a place removed from the ranking queue and handed out on judgement, which is exactly the sort of discretion the ranking was invented to avoid. The sport keeps them anyway, because a system with no discretion at all cannot readmit a great player after a year off, and cannot give a home crowd anybody to watch.
- Direct acceptance from the ranking24
- Qualifiers4
- Wildcards4
Illustrative rather than a specific tournament. Real splits vary by event and by tour, and the permitted number of wildcards is set in each rulebook.
Show the numbers
| Item | Value |
|---|---|
| Direct acceptance from the ranking | 24 |
| Qualifiers | 4 |
| Wildcards | 4 |
What the ranking is genuinely good at
Given how much criticism the system attracts, it is worth being precise about what it does well, because it does the job it was built for better than any alternative anybody has proposed.
It measures availability and consistency across a full year on every surface. That is a real quality and it is not a lesser one than peak level. A player who reaches the second week of all four majors, turns up at every mandatory event and reaches a couple of finals has demonstrated something a player with one spectacular fortnight has not.
It is nearly impossible to fluke, because the fixed slots mean everybody at the top is measured on the same core of tournaments. There is no soft schedule available to a top-ten player.
It is computable by anybody. No committee sits in judgement, no panel votes, no method is kept private. Given the published tables and a list of results, two strangers will arrive at the same number, which is why seedings are almost never litigated and entry lists almost never disputed. Sports that rank by committee spend a remarkable share of their year defending the committee.
And it is fast. The list is complete and correct every Monday, which is what a calendar with tournaments on three continents in the same fortnight actually requires.
ATP rankings explained: what the number cannot see
Four blind spots, and none of them is a bug. Each is the direct cost of a decision made in favour of simplicity.
It cannot see form. No decay inside the window means a quarter-final from last October and a quarter-final from last week are worth the same to the total, right up until the October one expires without warning. A player in the middle of the best three months of their life can be ranked below one who has not won a match since spring. Independent rating systems built on match results, which update after every match and weight recent form, disagree with the official list constantly, and they are usually the better guide to who will win on Thursday.
It cannot see the opponent. Points are paid for the round reached, not for who was beaten to reach it. A quarter-final is a quarter-final whether the path ran through the third seed or through three qualifiers, and a player who loses a three-hour match to the eventual champion in round two collects the same as one beaten by somebody ranked two hundredth. This is the largest gap between the ranking and any serious attempt at rating strength, and it is not an accident: opponent weighting would require the ranking to be calculated from itself, recursively, which is fine for a rating system and hopeless for a document that has to fill a draw sheet on a Wednesday. Other sports accept the complexity for the accuracy. Cricket's rating system weights the quality of the opposition and ages results gradually, and is correspondingly harder to reproduce on the back of an envelope.
It cannot see the surface. One list covers a sport played on clay, grass and hard courts, in conditions ranging from a Melbourne afternoon to an indoor hall in Paris. A clay specialist is seeded at Wimbledon on the strength of results that happened almost entirely on dirt. Everybody knows this is unsatisfactory, and the fix, seeding by surface, has been tried and largely abandoned because it hands discretion back to the tournaments and generates an argument every time it is used.
It cannot tell absence from decline. A player injured for eight months falls hundreds of places without having played badly once. The number reports what the last fifty-two weeks contain, and if they contain nothing then nothing is exactly what it reports.
To that list add one more caution. Ranking points are not comparable across eras. The tiers have been renamed, the tables rewritten, the counts changed and the calendar restructured more than once, so a total from one decade and a total from another are measurements in different units. Positions are roughly comparable. Points are not, which does not stop them being compared in every other article on the subject.
None of this makes the ranking dishonest. It makes it a specific instrument with a specific purpose, in the same way that the scoreline of a match tells you who won without telling you who played better, which is a separate and equally deliberate feature of how a tennis match is scored.
How to read a ranking properly
If you want to know where a player is heading rather than where they stand today, four things tell you nearly everything, and all four are public.
What they are defending over the next three months. This is arithmetic, not prediction. Last year's results at the coming weeks' tournaments are the points about to expire, and they set the direction of the number before a ball is struck. A player defending very little will rise by playing competently. A player defending a great deal has to repeat a career-best run just to stand still.
How full the count is, and what the smallest counting result is worth. Together those give the marginal value of the next tournament. A young player with empty slots gains the full face value of everything. An established one gains only the difference, which is why the same result moves two players' rankings by wildly different amounts and why the tennis press so often describes one of them as underachieving.
How much of the total came from the fixed slots. A total built on the majors and the top tier is more durable than one of the same size assembled from smaller events, because the big tournaments come round again with the same points available, while a fluke run at a 250 has to be replaced from somewhere.
The gap between the ranking and the race. If a player sits far higher in the calendar-year race than on the rolling list, they are improving and the ranking has not caught up. If they sit far lower, the ranking is a memory of a year that has already ended, and it is about to be corrected one Monday at a time.
Read those four together and the weekly ranking stories mostly stop being interesting, which is the point. The number is not an opinion about who is better. It is a receipt for the last fifty-two weeks, it is reissued every Monday, and the only genuinely useful question about it is which parts of it are about to run out.
Common questions
How do the ATP rankings work?
A player's ranking points are the total of their best results over the previous 52 weeks, counted up to a fixed maximum number of tournaments rather than added up without limit. The four majors and the top tier of tour events occupy fixed slots in that count, and the rest of the slots go to a player's best remaining results. The list is recalculated and published every Monday, so a result earned a year ago drops out on the Monday after the current edition of that same tournament finishes.
What does defending points mean in tennis?
It means the points a player earned at this week's tournament last year are about to expire, and that their ranking will fall by that amount unless this year's result replaces them. The old points come off whether or not the player enters, so someone who is injured loses exactly as much as someone who loses in the first round. Defending successfully by winning the same title again produces no gain at all, because the new points simply refill the hole the old ones left.
What is the difference between the ATP rankings and the ATP race?
The ranking is the rolling 52-week list used for entry into tournaments and for seeding the draws. The race is a separate table that resets on 1 January and counts only the current calendar year, and it decides which players qualify for the season finale. They converge in November, because by then the rolling window and the calendar year cover almost the same set of results, which is why the year-end number one is the player who leads the race.
What is a protected ranking in tennis?
A protected ranking, called a special ranking on the WTA side, is a number granted to a player who has been absent for a long period through injury, derived from where they stood when the absence began. It is a key to the door rather than a rank: it lets the player enter a limited number of tournaments within a limited window after returning, and it does not appear on the published list or generally earn a seeding. Their actual ranking stays wherever the empty 52 weeks have left it.
Why did a player's ranking change in a week they did not play?
Because a ranking total moves by subtraction as well as addition. Every Monday the window slides forward and the results from the corresponding week a year ago expire, so a player sitting at home can lose points, and a player above or below them can lose more. This is how the world number one position can change hands in a week when neither player won a match.
Filed under Tennis·tennis · atp tour · wta · rankings · seeding