How to improve the World Test Championship

ESPN Cricinfo

ESPN Cricinfo

Author 2019-11-09 21:19:49


In this article, I will look at the ICC World Test Championship from a historical analytical perspective: where are the shortcomings and what can be done to redress them?

The WTC is, arguably, the most important programme initiated by the ICC. It was badly needed and has come not a day sooner than warranted. The ideas are terrific and the intentions sincere. Faced with various pulls and pressures, the ICC has done a commendable job. I am certain, the rest of this article notwithstanding, that the two best teams in the world will qualify for the final at the end of the first cycle, and whoever wins it will be a deserving winner. There is no room for luck in those five or six days. Only skill will count and the team that is better across those days will lift the trophy.

Unfortunately, the structure and implementation of the WTC have a lot of shortcomings. I referred to this in an earlier article, and there have been other articles written on the subject. Maybe it is too late for this cycle, but I hope the ICC makes some tweaks in the next cycle. Let us first look at the current table to understand what could be amiss.

When someone, whether a cricket enthusiast from Canberra or a baseball supporter from Boston or a football fan from Florence, looks at this table, they will be flabbergasted. India have secured 240 points from five matches (and five wins). New Zealand and Sri Lanka have played out a 1-1 draw and have 60 points each. Australia and England played out a terrific 2-2 draw and have 56 each.

ALSO READ: FAQs: All you need to know about the 2019-21 World Test Championship

Everyone knows what the fundamental problem is. The ICC has worked on the principle that a series is the basis for points allocation and allotted 120 points per series. Unfortunately, they did not standardise the number of Tests in a series, which can range from two to five. Hence, all Tests are not equal. An India-Bangladesh Test in a two-Test series will carry 60 points, while an England-Australia Test in a five-Test series will carry 24 points. This is a gross anomaly. It is not right that two wins and a draw by each of the Ashes contestants should fetch each fewer points than a Test win does New Zealand or Sri Lanka.

Australia, playing away from home, got only 56 points for drawing one of the greatest Ashes series ever. And they could get 120 points for winning a home series against Pakistan 2-0 soon. India got 120 points for winning 2-0 against an average West Indies side, and they will get only 80 for winning a tough five-Test series against England 3-1.

I have presented the current Test championship stands in the table above, based on the universally accepted 3-1-0 points system adopted in most leagues. India are the runaway leaders, but Australia and England are not as far behind as they are on the actual table.

Analysis of all Test series so far
In order to substantiate my statements that the current points system is lopsided, I will fall back on history. Until the end of the recent India-South Africa series, 754 Test series had been played in all. There have been 17 six-Test series and 62 one-Test series. Since these types are not part of the WTC schedule, let's ignore them. That leaves us with 153 five-Test series, 60 four-Test series, 272 three-Test series, and 190 two-Test series. I will analyse each of these groups of series based on the results achieved.

At this stage I'm not interested in whether a three-Test series finished 3-0 or 2-1, since both have zero draws and the full 120 points will be allocated, to either or both sides. So I will mainly concentrate on the series that had no draws. As a special look-in, I will examine the clean sweeps, since these fetch one team all 120 points.

Five-Test series (Total 153)
# DrawsNumber% of TotalDetails
0 3522.9%5-0 (10), 4-1 (15), 3-2 (10)
1 4126.8%4-0 (13), 3-1 (18), 2-2 (10)
2 3925.5%3-0 (14), 2-1 (25)
3 15 9.8%2-0 (9), 1-1 (3)
4 1912.4%1-0 (19)
5 4 2.6%0-0 (4)

Only ten series (6%) among all five-Test series have ended in clean sweeps. Nine of these have been home series for the winning team. The only team to lose 0-5 at home was England against West Indies in 1984. Only 23% of all series have had results in each Test. Out of these 35 series, 17 were decided in the first three Tests. Longer series allow teams to recover and catch up. Only four series (2.6%) ended 0-0.

The average number of results (wins or losses) per series is 3.30 (66%). The average number of draws per series is 1.70. Teams are willing to sustain a few more draws because of the greater number of matches in a five-Test series.

Four-Test series (Total 60)
# DrawsNumber% of TotalDetails
0 1626.7%4-0 (6), 3-1 (8), 2-2 (2)
1 1931.3%3-0 (7), 2-1 (12)
2 1626.7%2-0 (9), 1-1 (7)
3 711.3%1-0 (7)
4 2 3.3%0-0 (2)

Only six four-Test series (10%) have ended in clean sweeps, all at home. Only 27% had results in each Test. Longer series allow teams to recover and catch up, right up the third Test. Only two series (3.3%) ended 0-0.

The average number of results per four-Test series is 2.67 (66.7%) and the number of draws per series is 1.33. These numbers are almost identical to the five-Test series.

Three-Test series (Total 272)
# DrawsNumber% of TotalDetails
0 8732.0%3-0 (44), 2-1 (43)
1 10137.1%2-0 (64), 1-1 (37)
2 6724.6%1-0 (67)
3 17 6.3%0-0 (17)

As many as 44 three-Test series (16%) ended in clean sweeps; 37 of these were home series for the winning side. A total of 32% of series contained no draws. Out of these 272 three-Test series, 190 had a live final Test. Three-Test series seem to be the perfect length, all things considered. Only 17 series (6%) ended in 0-0.

The average number of results per three-Test series is 1.94 (64.7%) and the average number of draws is 1.06. Amazingly, these numbers are almost identical to the four- and five-Test series.

Two-Test series (Total 190)
# DrawsNumber% of TotalDetails
0 11862.1%2-0 (91), 1-1 (27)
1 5529.0%1-0 (55)
2 17 8.9%0-0 (17)

This is the most revealing table. No fewer than 91 series ended in 2-0 sweeps (where one team would have got all 120 points). Another 27 ended 1-1. Thus, a total 118 series (62.1%) ended with non-draw results in each match. Only 17 (8.9%) had a 0-0 scoreline. This indicates that these two-Test series are likely to fetch many WTC points for the teams playing.

The average number of results per two-Test series is 1.53 (76.5%) and the average number of draws is 0.47. The result percentage is way higher than in the other series types. There is greater urgency in obtaining results because of the small number of matches. That is welcome, but, in a peculiar way, it does not work in the WTC, mainly because, as I explain below, almost all the two-Test series involve lesser teams. (No offence intended.) I personally would like all series, irrespective of which team plays, to be of equal length.

WTC anomalies
The real problem is that most two-Test series are likely to be played between a team that is clearly stronger than the other, or between two weaker teams. India, Australia and England are unlikely to play two-Test series among themselves. That means there is more to be gained by winning two-Test series than the seemingly high-profile four- or five-Test series. A team could sweep the lower-level two-Test series, do poorly in longer, tougher series, and still qualify. And vice versa: a team could win the tougher series, by narrow margins, not do well in the lesser series, and not qualify.

The ICC has picked the series as the base. However, points are allotted on individual match results. This is causing imbalances. It would perhaps have been better to have picked matches as the base and allotted the same number of points for all matches.

There are two fundamental requirements for any successful competitive league: all teams should play same number of matches, and all matches should carry equal weight. Unfortunately, both conditions are not fulfilled in the WTC.


Let me give you a Premier League analogy. Normally, Manchester United and Liverpool play each other twice - once at Old Trafford and once at Anfield. Suppose they decided that in order to get more exposure and revenue, they will play each other twice more at each venue. That makes it six matches during the season. However, if the Football Association said that they could only share six points in all, each win would give them one point, not three. If Chelsea and Manchester City agreed to play two extra matches, their points per match will be 1.5. Au contraire, a match between Watford and Norwich would be worth three points.

Before anyone jumps on me, let me say that these statements have nothing to do with what has happened in the WTC so far. India have played very well and deserve their position at the top. New Zealand deserve credit for drawing a tough away series in Sri Lanka. India are the strongest side in the world and are deservedly at the top of the ICC and WTC tables. They have the most balanced bowling attack in the world, are the strongest team at home, and can compete very well abroad. It would be a great surprise if they do not qualify for the WTC final.

I would not change a word of this article if Australia had gone to West Indies after the World Cup and won 2-0 and if India stayed on in England and played out a magnificent 2-2 draw.

Thoughts on restructuring
1. The ICC should standardise all series as three-Test series. Yes, there are fundamental problems to be solved, but once the WTC is instituted, that concept should take a higher level of importance over commercial or other considerations. What is wrong with an Ashes series or an India-Australia Test series comprising three Tests, the same as a series between West Indies and Bangladesh? Maybe the crowds in Kingston or Mirpur will number the same as the ones at Old Trafford or Bengaluru then.

2. All teams should play all other teams. Thus, there will be eight three-Test series. This will mean that each team will play 24 Tests. Each Test will carry 40 points, each series 120, and a total of 960 points will be at stake. There will be no lopsided points allocation.

3. I will not get into any political discussions. However, it is certain that India will not invite Pakistan, nor will they travel to Pakistan to play a series. Let the Tests be played in the UAE. If India do not want to play Pakistan in the UAE, they can play at Old Trafford, Galle or Mirpur. Surely, India and Pakistan can play a WTC series somewhere in the world?

4. The next cycle can start in July 2021 and end 30 months later, if more time is needed. The final can be played at Eden Gardens or the MCG around New Year 2024.

Analysis of all drawn matches so far
For this, I have used the in-depth Team Performance system I created, which allocates points to the two teams based on the extent of match completion and match status.

Analysis of drawn Test matches (765 Tests)
Match points# of TestsCumulative% of TotalDetails
100.0 2 2 0.3%The two tied matches
90.00 to 99.99235237 31.0%Good progress in fourth inns.
75.00 to 89.99188425 55.6%Third inns over. Fourth started.
50.00 to 74.99277702 91.8%Second inns over. Third started.
25.00 to 49.99 49751 98.2%First inns over. Second started.
0.00 to 24.99 14765100.0%First inns in progress.

This table shows that 91.8% of drawn Tests go past the halfway stage of the match. Very few matches fall in the under-25% completion stage. It also shows that it is not correct to treat all draws in a similar manner. Incidentally, the average match-completion percentage for these 765 matches is 75.6%. This shows that on an average, drawn matches do progress into the fourth innings. This fact should make the ICC stop and think about the way it has treated draws in the WTC.

Tie/draw-related tweaks
Unfortunately, in allotting points to draws, the ICC has gone with the equivalent of the 3-1-0 point system in football leagues. However, there is a major difference between draws in football and in cricket. A football draw is 0-0, 1-1, 2-2, 3-3, 4-4, 5-5 and 6-6 (the 7-7 draw is yet to be played). Historically, there have been only seven possible scorelines. A cricket draw has a million shades of colour. In many cases, a draw in cricket produces more excitement than a dull innings win.

ALSO READ: Andrew Fidel Fernando: The Test championship points system might be flawed, but it's the best we have

A win, whether it is by one run (Team pts: 50.08-49.92) or 675 runs (Team pts: 92.82-7.18) is a win. One team has beaten the other team fair and square. We can have academic discussions on many on-field happenings, but it is certain that the team has won the match. The winning team has won with an innings to spare, or scored a run more than the other team, or scored the requisite number of runs with wickets in hand. Even though I have the computations to quantify the wins accurately, as I have indicated earlier, I will not suggest that these be considered.

However, all draws are not equal. Consider the following six draws, given in order of completion.

Ind v WI, 2011: WI 590. Ind 482. WI 134. Ind 242/9

Pak v Aus, 2018: Pak 482. Aus 202. Pak 181/6. Aus 362/8

NZ v SL, 2018: SL 282. NZ 578. SL 287/3

Ind v SA, 2008: SA 540. Ind 627. SA 331/5

SL v WI, 2010: Win 303/8

WI v Eng, 2009: Eng 7/0

The first match was drawn with scores level and India had the minutest of edges. The second was a highly competitive close draw that almost produced a result. The third was a well-fought draw, with Sri Lanka saving the Test. The fourth was the dullest of draws on a batting paradise. In the fifth, the first innings was not completed. In the sixth, ten balls bowled.

Assuming that these were all played in a four-Test series, according to current WTC rules, each team in all these Tests would get ten points. That does not seem fair. These Tests were at widely varying levels of closure and should not be treated the same. With one more ball, the first match could have ended in a tie or an India win. The fourth one might have needed three more days for a result.

If the ICC does not want to use outside inputs, a simple suggestion presents itself:

- For all ties, 66.7% (2/3) of the maximum points should be allotted to both teams. There have been only two ties in the history of the game and these deserve extra recognition. Each team will receive 20 points for a tie if the maximum points for the match are 30.

- All draws that have gone past two innings get 50% of the points on offer; that is 15 points per team. To discourage teams from making convenient declarations, the points can be 55%-45% for completed innings, 55% going to the team with the lead.

- All draws that have not completed even two innings get 30% of the points on offer - that is, nine points each. To discourage teams from arbitrarily declaring, the match referee can be given the authority to judge and act on intent to play foul. After all, it is child's play to identify a situation like "Team 1: 200/5d, Team 2: 156/6d" and give them only 30% each.

Alternatively, if the ICC is ready to accept outside inputs (not necessarily mine; others might have similar methods), it could use the type of in-depth Team Performance system I have worked out, which allocates points to the two teams based on the extent of match completion and match status. I will give below my points allocation for the six Tests I have referred to earlier.

Ind v WI, 2011: WI 49.24, Ind 50.66. Total 99.90

Pak v Aus, 2018: Pak 52.40, Aus 45.15. Total 97.55

NZ v SL, 2018: SK 30.86, NZ 43.38. Total 74.24

Ind v SA, 2008: SA 30.26, Ind 35.66. Total 65.92

SL v WI, 2010: WI 10.31, SL 12.24. Total 22.55

WI v Eng, 2009: Eng 00.36, WI 00.02. Total 00.38

The total points allocated reflect the extent of match completion accurately.

A look at the home-away situation
Virat Kohli suggested that away wins should have double the points of home ones. While the idea makes sense, double the points seems way too much. The last thing the WTC wants is for India to get 240 points for defeating a lacklustre West Indies away. However, before looking at it any further, it is essential to conduct an analysis of Test results so far. I have analysed the results of the 2365 Tests played and presented the findings below. (All the Dacca Tests before 1970 are treated as home matches for Pakistan.)

Draws - All locations: 765
Neutral results: 34
Home wins: 958 (61.2%)
Away wins: 608 (38.8%)

The first two entries above can be ignored. This leaves us with the results in Tests played in home and away conditions. The overall pattern is quite clear. The roughly 60-40 split makes the home win a 3:2 probability. That does not warrant the 100% bonus suggested by Kohli. Maybe, a 10-15% bonus would suffice and could be considered.

Alternatively, or in addition, the ICC could decide to do away with the toss in home and away matches. Let the away captains decide what they want to do - even when India tour West Indies. If the captain makes a basic mistake, as Australia did at The Oval in the recent Ashes series, his team will pay for it, but it will be their decision. If South Africa had won the three tosses in their series in India last month and decided to bat, they might still have lost 0-3 but at least there would have been a better contest. Of course, in neutral locations like the UAE, there should still be a toss.

One final suggestion: when the innings goes past 100 overs, the ICC should consider adding one more review if the number of reviews available is below two. This will allow teams to get a fresh look in long innings and have some leeway as the new ball comes into play. It is fair to say that this move will benefit the bowling team rather than the batting team.

I have done a lot of heavy analytical articles recently. So one of my next two articles will be an anecdotal one. The timing is perfect since the current decade ends next month.

Email me with your comments and I will respond. This email id is to be used only for posting comments.


Experience triple speed

Never miss the exciting moment of the game