World Test Championship explained: points, final and flaws

Test cricket spent nearly a century and a half without a table. Not a bad table, or a disputed one. There simply was not one. Two countries agreed to play each other, played, and the result went into an archive. Nothing accumulated. Nothing was at stake beyond the series itself, and when the series ended the sport reset to zero and waited for somebody to arrange the next one. The World Test Championship, explained in its simplest terms, is an attempt to attach a destination to all of that, and the argument about whether it succeeded is now one of the better arguments in cricket.

To understand why the format looks the way it does, you have to start with the problem rather than the mechanism. The mechanism is odd. It scores a percentage rather than a points total, hands out different numbers of matches to different teams, penalises sides for bowling too slowly, and settles two years of cricket with one game in England. Every one of those choices is a response to a constraint, and most of the constraints are older than the competition.

The problem: a sport with no destination

For most of its history, Test cricket was organised as a series of private arrangements. Two boards agreed a tour, agreed how many matches it would contain, agreed who paid for what, and played. The Ashes ran to five Tests because both boards wanted five Tests and could sell them. A tour between two smaller boards might run to two, because two was what the money supported.

Nothing linked those arrangements together. A side could win a series in one country and lose one somewhere else three months later and the two results had no relationship at all. There was a ranking table, generated by a weighted formula that almost nobody could reproduce from memory and that updated in ways that felt arbitrary to anyone watching, but a ranking is not a competition. Nobody chases a ranking. It is a description of the past, not a race, and it produced no moment at which someone was handed something in front of a crowd.

The consequence showed up in the middle of the schedule rather than at the top of it. The marquee contests looked after themselves. A five-Test series between the two most watched sides in the game sells out on its own history, and the long rivalry that built the Ashes needs no table to give it meaning. What died was everything else. A two-Test series in an autumn gap, in a half-full ground, between two sides who had already played each other twice that year, was cricket that existed only for the people in the ground and the broadcaster who had already paid for it. When the series ended it left no trace on anything.

That is a slow, quiet failure, and it is much harder to fix than a bad rule. Audiences do not abandon a sport because a single match disappoints them. They drift away from a format that never seems to be going anywhere. By the middle of the 2010s the drift was measurable in attendance and viewing figures for anything outside a handful of fixtures, and it was arriving at the same time as two other pressures.

The first was the franchise calendar. Twenty20 leagues were taking a larger share of the year and a larger share of the best players, and they had something Test cricket did not: a season, a table, a knockout stage and a winner, all inside seven weeks. Test cricket's pitch to a player choosing between the two was that it was the highest form of the game. True, and no longer sufficient on its own.

The second was money, and specifically the way money is distributed in cricket. Without a competition, the boards with the most of it played each other most often, and the boards with the least played whoever would have them and for however long the tour budget stretched. A championship at least imposes an obligation that everybody must be scheduled against somebody.

The idea had been circulating since the 1990s and had been shelved several times, once in favour of a knockout play-off that would have displaced an existing white-ball event and did not survive contact with the broadcast arithmetic. What eventually launched, with the cycle that opened in 2019 immediately after that year's World Cup, was smaller than the versions that failed. That is why it works the way it does: it was designed to be laid on top of the calendar cricket already had, because a design that required a new calendar had already been tried and had already failed.

• 9Teams in the championship
• 6Series each team plays
• 3Of those, played at home
• 2Teams that reach the final

Structural counts defined by the championship regulations rather than figures from any particular cycle. The number of matches inside those six series is not fixed, which is the source of nearly every complication that follows.

What the championship actually is

Nine teams take part, the full members who play Test cricket regularly. Ireland, Afghanistan and Zimbabwe sit outside it, which is one of the format's least comfortable features and gets its own section below.

Each side plays six series inside the cycle, three at home and three away. It does not play everybody. Nine teams cannot complete a round robin in two years without demolishing every other commitment in the sport, so each team faces six of the other eight and misses two. Which six is not decided by the championship. It falls out of the schedule that already existed.

A series counts whether it contains two Tests or five. That single sentence is the origin of the percentage system, the strength-of-schedule complaint, and most of what people find strange about the table.

At the end of the cycle the top two meet in a one-off final at a neutral venue. Everyone else stops, the table resets, and the next cycle begins. The winners take the ICC Test Championship mace, which before the competition existed was simply held by whichever side happened to top the rankings on a particular date each year, which tells you something about how badly the sport wanted a trophy that somebody had to win rather than merely occupy.

Why the table is a percentage and not a points total

Start with the constraint, because the answer is forced by it.

Teams in this competition do not play the same number of matches. Not approximately the same, and not within a match or two. One side might play sixteen Tests across its six series and another ten, because the series lengths are set by the boards involved rather than by the competition. If you rank a table on raw points, the side that played sixteen is competing on a scale that goes half again as high as the side that played ten. Every additional Test is another chance to add to the total, so the team that was scheduled most wins the table, and the table stops measuring quality and starts measuring itinerary.

That is not a subtle bias you could correct at the margin. It is the dominant term.

The fix is to divide. Count the points a side has actually won, count the points that were available in the matches it played, and express the first as a proportion of the second. Now every side is scored on the same scale, from nought to a hundred, no matter how many matches it played. A side that wins two thirds of what was on offer sits above a side that wins half of it, whether the first played nine Tests or eighteen.

Percentage-of-available is a normalising device, and cricket is not the only sport that reaches for one. Tennis solves a related problem in a different way: instead of dividing by what was available, it counts only a fixed number of a player's best results inside a rolling window, so a player who enters forty events cannot out-accumulate one who enters eighteen. The mechanics of how the ATP and WTA rankings handle that are worth comparing, because the two sports have arrived at opposite solutions to the same underlying unfairness. Tennis caps the numerator. Cricket inflates the denominator.

Both approaches have the same weakness, and it is worth naming now rather than at the end. Normalising for the number of matches does nothing about the difficulty of them. A percentage treats a Test in Australia and a Test against a side in transition as identical units of opportunity. More on that below, because it is the criticism with the most substance behind it.

The first cycle's method, and why it was abandoned

The original design did not use a percentage. It allocated a fixed pot of points to every series, and divided the pot by the number of matches in it.

Each series was worth 120 points in total. A two-Test series therefore paid 60 points for a win in each match. A three-Test series paid 40. A five-Test series paid 24. Draws and ties took proportionate fractions of those figures. The reasoning was the same as the reasoning behind the percentage that replaced it: no side should gain an advantage from having more cricket scheduled, and every series should be worth the same to the table.

It was a defensible idea that produced results almost nobody could defend.

The first problem was that it valued the wrong thing. Beating a strong opponent four times in a five-match series earned less than beating a weaker one once in a two-match series. The competition was, in effect, paying a bonus for playing short series, and the short series were mostly the ones between the sides with less at stake. A format designed to reward achievement was rewarding scheduling.

The second problem was internal to the long series. If a series is a fixed pot, then the fourth Test of a five-match series that has already been decided is worth exactly the same slice of that pot as the first. In theory that is elegant. In practice it meant the marquee series, the ones with the most Tests, had the least valuable individual matches in the competition, which is close to the opposite of what a championship wants.

The third problem was not the format's fault at all. The pandemic tore the schedule apart. Series were postponed, shortened and cancelled, sides ended up having played wildly different numbers of matches through nobody's decision, and a system built on the assumption that everybody would complete six full series had no way to compare a team that had played its programme with one that had played half of it.

The percentage table arrived first, as an emergency repair while that cycle was still running, because it was the only method that could rank teams who had played completely different amounts of cricket. The flat per-match points value came at the start of the following cycle, once it was clear that dividing a pot by series length had been the wrong idea to begin with. The two changes together are the system as it now stands: every Test worth the same, and the table ranked on the fraction of what was available.

What each result is worth

Every Test in the championship carries the same allocation, whatever series it belongs to and whether it is the first match of a five-Test tour or the last match of a two-Test one.

Result	Points
Win	12
Tie	6
Draw	4
Loss	0

A tie, where both sides are dismissed and the scores finish level, is worth half a win. It is one of the rarest results in the sport and can be safely ignored when thinking about how the table behaves.

The draw is the interesting number. Four points is a third of a win, and that ratio is doing more work than it appears to.

Set it too high and the competition rewards survival. If a draw were worth six or eight, a side arriving in hostile conditions would have a rational path to a good percentage that involved never trying to win a match. Set it too low, at one or two, and the last day of a losing position becomes meaningless, which removes one of the things Test cricket is uniquely good at: a side batting through a final session with nine wickets down and nothing to gain but the refusal to lose.

Four is the compromise. It is enough that batting out the fifth day is playing for something real. It is not enough to build a cycle on. Three draws pay the same as one win, so a side that draws everything finishes with a percentage in the low thirties and finishes nowhere. If you have never had to think about why a draw is a legitimate result rather than a failure, the structural differences between Tests and the limited-overs formats are the place that logic comes from.

Win12

Tie6

Draw4

Loss0

The allocation per match under the current playing conditions, identical in every series regardless of its length. The ICC publishes the playing conditions in force for each cycle.

Show the numbers

Championship points for a single Test, by result

Item	Value
Win	12
Tie	6
Draw	4
Loss	0

How a series result becomes a number on the table

The arithmetic is not difficult, but the order of operations is where people lose the thread, particularly with over-rate deductions, which are applied to a match after it has been scored rather than to a season total at the end.

1. 1The Test is played and a result is recordedWin, loss, draw or tie, under the ordinary Laws. The championship does not change how a match is decided, only what happens to the result afterwards.
2. 2Each match is scored on its ownTwelve points for the win, four for the draw, six for a tie, none for a defeat. The series it belongs to is irrelevant. A dead rubber pays the same as an opener.
3. 3Any over-rate deduction is applied to that matchPoints are docked for each over the fielding side finished short of the minimum, after allowances. The deduction reduces the points earned; it never reduces the points available.
4. 4The points earned go into the running totalThis is the numerator. It only ever moves upwards, and only on a result better than a defeat.
5. 5The points available go into a second totalTwelve for every Test played, win or lose. This is the denominator, and it moves after every single match without exception.
6. 6The table is re-sorted on the fractionNumerator divided by denominator, expressed as a percentage. Sides who have played twelve Tests and sides who have played eighteen sit on the same scale.

The sequence applied to every championship Test. The denominator moves whether the match was won, drawn or lost, which is why a defeat costs a side twice: nothing is added above the line and twelve is added below it.

The worked example below is invented. The three teams, and every figure attached to them, are constructed to show the arithmetic and are not drawn from any real cycle.

Northland have played fourteen Tests: seven wins, three draws, four defeats. That is 84 points from the wins plus 12 from the draws, so 96 earned. Fourteen Tests at twelve points each is 168 available. 96 divided by 168 is 57.14 per cent.

Eastvale have played twelve: six wins, four draws, two defeats. That is 72 plus 16, so 88 earned, from 144 available. 61.11 per cent.

Southport have played ten: six wins, one draw, three defeats. That is 72 plus 4, so 76 earned, from 120 available. 63.33 per cent.

On raw points, Northland lead comfortably with 96, Eastvale are second on 88 and Southport are last on 76. On percentage, the order reverses completely.

Southport, 10 Tests63.33%

Eastvale, 12 Tests61.11%

Northland, 14 Tests57.14%

Constructed example, not real standings. Raw points earned are Northland 96, Eastvale 88, Southport 76, in exactly the opposite order to the percentages shown here. Northland lead on volume of cricket, not on results.

Show the numbers

Three invented teams, and why the raw total misleads

Item	Value
Southport, 10 Tests	63.33%
Eastvale, 12 Tests	61.11%
Northland, 14 Tests	57.14%

That reversal is the whole argument for the percentage in one picture. Northland have won the most matches in the competition and have the largest pile of points, and they are third, because four defeats in fourteen is a worse rate than three in ten. Whether that is fair depends entirely on who Northland played and where, which the percentage does not know and does not ask.

The over-rate penalty, which has decided placings

This is the part of the championship that gets skipped in almost every explanation of it, and it has moved teams in the table.

Test cricket sets a minimum over rate, worked out across the match with allowances for wickets falling, drinks, injuries, reviews and other stoppages. A side finishing below that minimum is short, and the shortfall is counted in whole overs. Under the championship the sanction is not only a fine on the players. Championship points are deducted, which is a heavier punishment by a distance, because a fine is a private matter between a board and its players and a deduction is public and permanent and sits in the table until the cycle ends.

The original rate was two points per over short. Applied to a match worth twelve, that is punishing. A side could win a Test, be six overs short across five days, and walk away with nothing. From a later cycle the deduction was halved to one point per over short, and the calculation of allowances was revised so that time genuinely lost to circumstances outside a side's control is treated more sympathetically.

Work through what even the reduced version does. Take an invented case: a side wins a Test and finishes three overs short. Twelve points become nine. The denominator does not move, because twelve points were still available in that match, so the side has effectively converted a win into three quarters of a win. Do that three times in a cycle and nine points have gone, which on a fourteen-Test programme is a shift of more than five percentage points. That is the difference between second and fourth in a tight table.

The tension in the sanction has never been resolved, and it is a genuine tension rather than a complaint.

Slow over rates are a real problem and not a manufactured one. They shorten the amount of cricket a paying spectator gets for the price of a ticket, and they turn the last hour of a day into an exercise in arithmetic, with the fielding captain visibly working out how few overs he can get away with. Something has to police that.

But the penalty falls on the fielding side in a game where the batting side controls much of the tempo. A batter who takes his time between deliveries, calls for a change of gloves, or has the physio on is costing the fielding captain overs he will later be docked for. And the sanction falls hardest on precisely the attack Test cricket most wants to see: four fast bowlers, long run-ups, frequent field changes, in heat. A side with two spinners gets through its overs and is never troubled by the rule. There is a version of this argument in which the championship is quietly taxing pace bowling.

Points can also be docked for other offences, most obviously a pitch rated as unfit under the ICC's pitch and outfield monitoring process, and for serious code of conduct breaches. Those are rare enough that they are not part of anybody's planning for a cycle. Over rates are not rare. They are a running cost.

The final: one match, one venue, two arguments

The top two at the end of the cycle play a single Test. That is the entire knockout stage of a two-year competition.

Several details of it matter, and most people know none of them until a final actually turns on one.

It is at a neutral venue, and in every cycle so far that venue has been in England. The reasoning is a mixture of the commercial and the practical: an English June sits in a gap in most international calendars, the grounds are free, and a crowd can be assembled from resident populations of every cricketing nation without anybody having to fly. The consequence is that the deciding match of a world championship is played in the conditions of one participating country, with a Dukes ball and English preparation, which is not neutral in any cricketing sense. Anyone who has watched a side arrive from a dry, low, slow surface and try to bat in overcast conditions against a swinging ball understands what a venue does to a match, and how much the surface itself dictates is the reason the choice of ground is a sporting decision rather than an administrative one.

A reserve day is scheduled. It is not a sixth day of cricket in the ordinary sense, and it is regularly misunderstood as one. It exists solely to recover playing time lost to weather during the five scheduled days, and it is used only if that lost time cannot be made up inside the normal schedule by extending play on the days that remain. If the five days are completed without significant interruption, the reserve day is not played, whatever the state of the match.

If the final is drawn or tied, the trophy is shared. No super over, no bowl-out, no countback to the league table, no award to the side that finished higher in the standings. This is the rule that most surprises people arriving from a knockout sporting culture, and the defence of it is sound: a draw is a legitimate Test result, and any tiebreak would change how the final was played. Give the trophy to the higher-placed finalist on a draw and you have handed one side a licence to bat for five days. That is a worse final than a shared one.

Three finals have been played and decided so far. New Zealand beat India in Southampton in a match extended into the reserve day by rain. Australia beat India at The Oval. South Africa beat Australia at Lord's, a result received at home as the end of a long and painful run of near misses in global tournaments.

Now the argument, honestly on both sides, because this is where most discussion of the championship becomes tribal.

For the single match. A final has to be an event. It needs a date, a venue, a crowd and a broadcast slot, and it needs to be resolvable in a fixed window, which a best-of-three is not. Three Tests would take a month, would require a host willing to give up a month, and would produce a series that could be over after two matches with a third to be played anyway. There is also a sporting argument rather than a logistical one: the league phase has already tested consistency over two years, and a final is meant to test something different, which is the ability to play well on one occasion when it is the only occasion. Every knockout sport in the world accepts that the better side sometimes loses on the day. That is not a flaw in the design. It is what a final is.

Against the single match. A Test is the least reliable unit of measurement cricket has. The toss is worth a great deal on some surfaces, and enormous amounts on a few. Weather can remove a day and reshape the game. A championship that spends two years carefully accumulating evidence about who is best, then discards nearly all of it in favour of five days in one country, has built an elaborate qualifying system for a lottery. The counter to "that is what a final is" is that in most sports the knockout stage is more than one match, or the league phase is short enough that nobody pretends it was the real test.

Both positions are coherent. What is not coherent is the position that a single final is fine and that the league table is a rigorous measure of the best side, because the two claims pull in opposite directions.

The scheduling problem underneath everything

Every other criticism of the championship is downstream of this one, and it is the part that deserves the most attention, because it is not a rule that could be amended at the next ICC meeting. It is the structure the competition sits inside.

Test fixtures are not set by the championship. They come from the Future Tours Programme, which sounds like a fixture list and is not one. It is a register of bilateral commitments, negotiated between pairs of boards over a multi-year block and then collated and published. The ICC ratifies the FTP. It does not compose it. Two boards decide between themselves whether they will play, when, and over how many matches, and the answer to all three questions is largely commercial.

The reason it is commercial is worth understanding precisely, because it explains everything else. In cricket, the home board sells the broadcast rights to its own home series. The visiting side is paid little or nothing beyond expenses. That means a home series is not a shared enterprise with revenue split down the middle; it is a revenue event owned by one of the two participants. And the value of that event depends almost entirely on who the visitors are, because the largest broadcast market in the sport is a single country, and a home series against the sides that market wants to watch is worth a multiple of a home series against anyone else.

Follow that through and the whole shape of Test cricket falls out of it.

A board hosting a lucrative opponent wants as many Tests as possible, because each additional match is additional inventory to sell. So those series are five matches, and they recur.

A board hosting a less lucrative opponent wants as few as it can decently arrange, because each additional match is additional cost, staging, staff, security and player payments, against thin returns. So those series are two matches, and they recur less often.

A board touring is not being paid much either way, so a long tour to a place with no broadcast value is a straightforward loss, and the sides with the least money are the ones least able to absorb one.

None of that is corruption or bad faith. It is what happens when the fixture calendar is assembled from individually rational commercial decisions with no central authority able to overrule them. The result is a hierarchy: the wealthiest boards play each other most often and over the most matches, and the rest fit around them.

The championship was laid on top of that. It did not replace it, and it was never in a position to. The competition takes the series the boards have already agreed and declares that six of them count. So the championship inherits, wholesale, the inequality of the schedule it was partly meant to address.

The practical consequences are severe, and they are the reason the percentage table remains contested even though the arithmetic is sound.

Sides play very different amounts of cricket. A team on five-Test series against the big draws and three-Test tours elsewhere may play sixteen or more championship Tests. A team on two-Test series throughout plays ten or twelve. The percentage handles the count. It does nothing about the fact that a side playing ten matches has far fewer opportunities to recover from one bad week, which makes its percentage more volatile without making it more or less deserved.

Sides face very different opponents. Two teams can qualify for a final having played programmes with almost nothing in common. One may have toured both of the hardest away destinations in the sport and played five Tests in each. The other may have played three home series against sides in transition and short away tours to places that suit it. Both are ranked on the same scale, as if the two tasks were equivalent, because the percentage has no concept of opponent or venue.

Nobody wants to fix it, for a specific reason. A strength-of-schedule weighting has been proposed repeatedly and has been rejected every time, and the reason is political rather than mathematical. Weighting opponents means the governing body formally publishing a number that says one member is harder to beat than another, and then paying teams differently for beating them. Every board would have to agree to a system that ranks it, and the ones ranked lowest would be asked to agree that beating them is worth less. That is a difficult vote to win, and it has not been won.

And the calendar keeps shrinking. Franchise Twenty20 leagues now occupy windows in most months of the year, boards schedule around them because their own players and their own revenue are involved, and the space left for a five-day format is under pressure everywhere except in the few places where Test cricket still pays for itself. The championship has no window of its own and no power to reserve one. It fits into whatever the FTP leaves.

This is why the criticism that the championship "does not fix the schedule" is both correct and slightly beside the point. It could only fix the schedule by taking control of it, and taking control of it would mean boards surrendering the right to sell their own home cricket. There is no version of the competition, under the current economics, that has that power. Everything else about the format is a workaround for not having it.

Three full members left outside

Ireland, Afghanistan and Zimbabwe play Test cricket and are not in the championship. They therefore play Tests with no competitive context whatsoever, which is precisely the condition the whole competition was invented to end.

There is no promotion, no relegation and no second division, so there is no route in. The nine are the nine. The stated reasons are practical: those sides play few Tests, sometimes none in a year, and a percentage table built on three or four matches would be meaningless and would distort the standings for everybody else. That reasoning holds. What it does not do is offer any path, and a competition with no path is a closed shop with an explanation attached.

A second division has been discussed at intervals and has never been adopted, partly because it raises the same question the strength-of-schedule debate raises: which members would be in it, who decides, and what does relegation do to a board's broadcast income if it can no longer promise a home series against the sides its market wants.

What the championship has changed, and what it has not

Sort the effects honestly and the picture is mixed rather than triumphant or damning.

Dead rubbers are genuinely dead less often. This is the clearest win. Under the flat per-match allocation, the fifth Test of a series already decided is worth exactly twelve points, the same as the first. That is a real change in incentive, and it applies to the sides that used to have the least reason to care: the team already 3-0 down, playing a final match in front of a thin crowd, now has twelve points on the table and a percentage that will carry into the next series. Not every dead rubber has come alive. But the arithmetic reason to play one seriously now exists, and it did not before.

The value of a draw is clearer than it was. Four points is a small, specific, publicly visible reward for the fifth-day rearguard, and it has given commentary a vocabulary for something that used to be described purely as a moral achievement. A side hanging on for a draw is now demonstrably playing for a third of a win, and everybody watching can do that sum.

The incentive to force a result is real but weaker than people claim. Twelve for a win against four for a draw is a strong ratio, and it does push captains towards declarations and towards taking a chance on the fourth evening. What it does not do is override the situation in front of them. A side one down in a two-Test series with a percentage to protect will still take the draw, because in percentage terms four points from a match is a lot better than nought, and the difference between a draw and a defeat can matter more than the difference between a draw and a win. The championship has not abolished defensive cricket. It has priced it.

Over rates are policed and still not solved. The deduction is a genuine sanction that has cost teams real ground, and it is applied consistently. Play has not obviously got faster. What the rule has done is make slow play a subject that captains are asked about, which is a change in accountability rather than in behaviour.

The table has become part of the conversation, which was the actual goal. Captains discuss percentage points in press conferences. Series that would once have been ignored feed a standing that people check. Nobody argued about whether a two-Test series in a gap in the calendar mattered before 2019, because there was no mechanism by which it could. Whether that is worth the compromises depends on what you thought the problem was. The cycle-by-cycle detail of who is playing whom, and where the table stands, is worth following alongside the fixtures rather than on its own.

It has not made Test cricket equal. It was never going to, because equality would require a scheduling authority the sport does not have.

What a fairer design might actually look like

The criticisms are easy. The alternatives deserve to be taken seriously rather than waved at, so here are the four that come up most, with what each would cost.

Weight results by opponent and by venue. The most obvious repair. Beating a top-ranked side away is worth more than beating a bottom-ranked side at home, on a multiplier published in advance and derived from the existing ranking formula. It would fix the deepest flaw in the percentage, which is that it treats all twelve-point opportunities as identical. The cost is the political one already described, plus a real cricketing objection: a multiplier derived from rankings imports whatever is wrong with the rankings into the championship, and it makes the table harder to read at a glance. Some of the appeal of the current system is that anyone can compute it.

Standardise series length. Make every championship series three Tests, and let boards schedule additional non-championship matches on top if they want them. Every side plays eighteen championship Tests, the denominators match, and the percentage becomes a raw total again. The cost is that it either kills the five-Test series or makes two of its matches decorative, and the five-Test series between the biggest draws is the single most valuable product Test cricket has. No board is voting to devalue it.

Expand the final. Best of three, at three venues, or a two-match final with the higher-placed finalist retaining the trophy on a drawn series. The second version is cheaper than it sounds: two Tests, one at each finalist's home ground or both at a neutral one, with the league position acting as the tiebreak. It gives the league phase a purpose beyond seeding and reduces the weight on a single toss. The cost is a month of calendar and a scheduling negotiation with every league and board that would have to give the window up, and a final that can end without a deciding match, which broadcasters dislike.

Add a second division with promotion. Bring in the excluded members, run a lower tier with fewer matches, promote and relegate between cycles. It solves the closed-shop problem and gives every Test in the world a context. The cost is the one that has killed it every time: relegation would remove a board's ability to promise its broadcasters a home series against the sides that generate the revenue, which for some members is the difference between solvency and not. Any workable version needs a revenue-sharing arrangement attached, and that is a much larger argument than a competition format.

Notice what all four have in common. None of them is a cricketing problem. Each is technically straightforward and each fails at the same point, which is a board being asked to accept less money or less control. That is the real design constraint on the World Test Championship, and any proposal that ignores it is a wish rather than a plan.

How to read the table mid-cycle without being fooled by it

The standings are published continuously and are misleading for most of the cycle. Five habits make them useful.

Ignore the percentage until a side has played five or six Tests. Before that, the denominator is too small and one result moves the figure enormously. A team that has played two Tests and won both sits on 100 per cent and is telling you nothing at all. Early-cycle tables are noise dressed as information.

Read the matches-remaining column before the percentage column. A side on a high percentage with two Tests left has almost no capacity to improve and a great deal to lose. A side a few points behind with eight left has all the room in the world and all the risk. The two figures only mean something together, and most published tables put them side by side for exactly this reason.

Check where the remaining matches are, not just how many. Three home Tests against a side that historically struggles in those conditions is a very different prospect from three away in the hardest place to tour. The table has no idea about this. You do.

Watch the over-rate deductions, which are applied quietly. They are announced after matches, usually in a short disciplinary note, and they go into the table without much attention. A side that has been docked several points across a cycle can miss a final by a margin smaller than its penalties, and the table will not tell you that is what happened. It shows only the net figure.

Convert the percentage back into matches before you get excited. A gap of four percentage points on a fourteen-Test programme is about seven points, which is somewhere between a draw and a win. Percentages compress differences and make close races look decided and decided races look close. If you want to know what a gap really is, multiply it back out into points, then divide by twelve. That is the number of results in it.

Do all five and the table becomes what it was meant to be, which is a running account of who has done what. Do none of them and it becomes what it most often is in the middle of a cycle, which is a leaderboard that changes for reasons unrelated to anything you watched.

The rest of the numbers underneath all this, the ones that decide the matches that decide the table, are read off a scorecard, and the conventions for pulling an innings apart column by column are the same in the championship as they are anywhere else. More on the format and everything around it sits in the cricket archive.