As I wrote in my last article, a few readers pointed out that I have a Test championship system that allocates points for each Test match based on Result, Team Performance and Location/Team Strength. They wanted me to simulate a Test championship for recent years based on this system. This is an excellent idea. It allows me to figure out how to handle the problems faced by the ICC - differing number of Tests, differing number of series, home/away imbalance, etc.
This article will simulate a Test championship covering the 130-plus Tests of the last three years - from January 2017 to December 2019. First, let me briefly describe the Test championship system. For each Test, I work on the following indices:
1. Result Index (Maximum of 3.0 points)
Irrespective of the teams, location and match performance, these are the points allocated for the result achieved. The allocation of points is as follows:
Win: 3.0 points
Tie: 2.0 points
Draw-1: 1.5 points (At least two innings are completed)
Draw-2: 1.0 points (Two innings are not completed)
Loss: 0.0 points
2. Team Performance Index (Maximum of 1.0 point)
The allocation here is based on complex performance analyses of the scores in the match. A few typical point allocations are given below, but it must be remembered that the actual point allocations depend on the scoring patterns in the Test. Two one-wicket wins will differ from each other in their points allocations, and so will two 13-run wins.
50.0 - 50.0 for a tie (the anchor value - Australia v West Indies, Brisbane, 1960-61)
50.1 - 49.9 for a one-run win (Australia v West Indies, Adelaide, 1992-93)
50.4 - 49.6 for a one-wicket win (South Africa v Sri Lanka, Durban, 2018-19)
50.6 - 49.2 for a draw with scores level and nine wickets down (India v West Indies, Mumbai, 2011-12)
53.9 - 46.1 for a 13-run win (Australia v England, Sydney, 1886-87)
58.1 - 41.9 for a five-wicket win (Ireland v Pakistan, Malahide, 2018)
71.2 - 28.8 for a nine-wicket win (New Zealand v Bangladesh, Christchurch, 2016-17)
90.6 - 9.4 for a win by an innings and 198 runs (Pakistan v Australia, Sharjah, 2002-03)
94.2 - 5.8 for a 675-run win (Australia v England, Brisbane, 1928-29)
43.4 - 30.9 for a three-innings draw (New Zealand v Sri Lanka, Wellington, 2018-19)
0.3 - 0.1 for a ten-ball draw (West Indies v England, North Sound, 2008-09)
A detailed description of the process will take too much space. It has been done earlier. Suffice to mention that a very complex process allocates points for a match, teams, innings, functions and players. The base for all allocations is that all results (including the two ties but excluding South Africa-England in Centurion, 1999-00, and England-Pakistan at The Oval, 2006, for obvious reasons) will get allocated 100 points. All draws will get allocations below 100, depending on the extent of completion of the match. Then follow allocations by level, right up to the player.
All innings wins are allotted 75 or more points. However, depending on the margins and match scoring patterns, wins by runs could get more than 75 points. For instance, Australia received 75.1 points for their win by an innings and five runs over Pakistan in Brisbane and 76 for their 296-run win over New Zealand in Perth.
3. Location-Result-Team Strength Index (Maximum 1.0 point)
This index is based on the location of the match (home/away/neutral) and the result (Win/Loss/Tie/Draw). These two factors result in ten combinations (Tie + three Locations x three Results). Each combination is assigned a suitable weight and this weight is indexed by the Relative Team Strengths of the two teams, taking into account the location. Location and Team Strengths get relatively higher weights since Result has already been covered. The result is an index value from 0.00 and 1.00. Afghanistan got the maximum points of 0.99 for their away win against Bangladesh last September and many strong teams got 0.01 points for home losses to very weak teams.
Match Rating Points (Maximum 5.0 points)
The MRP value is the sum of the above three index values. The theoretical range is 0.0 to 5.0. The actual highest value was secured by Afghanistan for their win in Chattogram. They secured 4.71 points. The fewest points were secured by Pakistan (0.11) when they scored 59 and 53 against Australia's 310 in 2002.
The fewest points for a win - 3.78 - were awarded to the strong Australian team, playing at home, when they defeatde a weak Indian team by only 48 runs in Adelaide in 2014. This was the match in which Virat Kohli scored a century in each innings. Before anyone jumps on me, let me say that this was indeed a very weak Indian team: the highest away batting average was 39 and the bowlers' away averages were quite high.
For readers to understand the Test Performance Analysis system, I have provided a table containing the detailed points allocation break-up for seven selected Tests - most of these are recent ones.
1. TS refers to Location-based Team Strength, TPP is Team Performance Points and MRP means Match Rating Points.
2. It can be seen that a strong England's close win over India at home gets only 3.93 points while India's win over South Africa is valued slightly more because of the margin of win.
3. On the other hand, Afghanistan's away win over a Bangladesh team twice as strong fetches 4.71 points - the highest ever. Australia's away win over England is rated at just above 4.0 because of Australia's higher Team Strength.
4. The tie is differentiated only in the second decimal - that too because of Australia's marginally higher team strength.
5. The first draw is that of a weaker England over a stronger Indian team away. Hence England gets about 15% more points.
6. The second draw is in the recent Rawalpindi match, which did not even reach the halfway stage. However, Pakistan get slightly more points than Sri Lanka because their home Team Strength is undefined and they did better in the match. The two teams get below 1.5 points.
Simulated Test Championship 2017-1019
The teams are allotted the MRP values for each Test they played during the three-year period. The total is divided by the number of Tests each team played and the average value is used to determine the top-performing teams during the three-year period. I have strictly gone by the dates. In other words, only Tests starting between January 1, 2017 and December 31, 2019 are considered. To me, the concept of "series" is secondary. However, as can be seen later, the display is by series and some tweaks have been made across years. During this three-year period, 134 Tests were played. Out of these, 12 Tests featuring Zimbabwe or Afghanistan or Ireland have been ignored. Thus this analysis covers 122 Tests.
Let us see how I have addressed the shortcomings that exist in the ICC's WTC system.
1. All qualifying Tests are considered. The number of Tests played by a team varies from 34 by Australia to 19 by Bangladesh. This is an acceptable spread and is likely to occur in any considered period.
2. All Tests carry equal weights - a maximum of 5.0 rating points. The mean Match Rating Points value is a fair method to determine the best-performing team.
3. The patently unequal and unbalanced series concept is given a miss. It is understood that a series win will automatically follow if Tests are won. I have checked and can confirm that there is not a single series in history in which the winning team secured fewer MRPs than the losing team. Even a series with three narrow wins and two big losses or four draws and one close win is likely to produce 12.0-10.0 or 10.0-8.0 MRP scorelines.
4. Away results are given due weight. At the same time, strong away teams playing weak home teams do not get undeserved credit.
5. Team strengths are considered. A weak team defeating a stronger team is given due credit.
6. Match performances are considered and emphatic wins given due weight. However, the base points are the same whether it is a one-run win or 400-run win.
7. Ties are given two-thirds weight. Draws that have progressed well are given half weight. Finally, draws that have not even gone past the halfway stage are given only a third weight.
Thus it can be seen that all the points I had mentioned in my critique of the WTC are addressed effectively in this simulation.
Before I get to the table of team performances, let me show the results in graphical form. There are four graphs: one for each year and one for the entire period. I will provide only highlights of each graph, leaving it to readers to peruse and draw insights from the visual presentations.
The colour codes depicting results are quite distinct. The location (home/away) is indicated by the small circle above the square. The Tests of a series are grouped together for easier perusal. If a series overlaps two years, it is shown in the year in which a majority of the Tests of that series were played.
The highlights of 2017 are presented above. India's narrow win over Australia after losing the first Test in Pune can be clearly seen. England's thumping of South Africa at home and their Ashes disaster down under are also prominent. In between come Bangladesh's two home draws against Sri Lanka and Australia. New Zealand's good year can be seen with clarity.
In 2018, South Africa followed a tight win over India with a rout of the Australians. India's poor tour of England stands prominent. England continued in winning vein, this time away in Sri Lanka. Finally India's famous away win over Australia stands tall. Tucked in there is Bangladesh's 2-0 win over West Indies. Australia had an awful 2018, losing all three Test series they played.
In 2019 what stands out is India's run of seven consecutive Test wins - all part of the WTC. As one looks at the graphics, the ups and downs of the topsy-turvy Ashes series come to mind. Australia had a great 2019. Sri Lanka's momentous away win over South Africa stands out like a beacon. Pakistan's disastrous year is also obvious. The lack of home matches is taking a toll on them. Although I must say that the resumption of home Tests in Pakistan is wonderful and they have resumed playing at home very well.
This is a representation of all 122 matches. It is clear that India and New Zealand have been the best performers during these three years. New Zealand's last three Tests stand out, indicating how they have fallen off. Australia and England have been inconsistent. However, Australia's recent run has been impressive. South Africa started strongly but have fallen off badly recently. Sri Lanka have had their moments both home and away. The absence of home matches for Pakistan stands in stark contrast to the others. The rare moments of glory for West Indies stand out.
Now, on to the Team Performance Table.
It is no surprise that India have finished at the top of the table. Their average of 2.91 points per match indicates their average performance falls in between a draw and a win. On the maximum value of 4.71 points, this represents a very creditable 62%. India's standout performances were their wins against Australia (both home and away), their away sweeps against Sri Lanka and West Indies, and their late sequence of home series wins. These more than made up for the two losses against South Africa and England. Their 20 wins during this period are a clear indication of their overall success.
New Zealand, second to India in the ICC's Test rankings at the end of 2019, have also come in second in my simulation. Their placement is well-deserved. (One thing to note, though: my table includes the first two Tests of the Australia-New Zealand series, played in December 2019, while the ICC rankings at the end of the year don't take them into account. If you took those matches to be a two-Test series and calculated the rankings points for them, Australia would have been marginally ahead of New Zealand in the table at the end of the year.) In my simulation, New Zealand led the table until December 15. Their sub-par performance against Australia pushed them down to second place. The difference between a win and loss for a team that has played around 20 matches is about 0.15, which is good enough to move a team down. However, it must be said that New Zealand played seven out of their ten series at home. And they won six of those seven, including two against England. It must also be said that these home wins get around 8-10% fewer points.
Australia are comfortably placed in third position. It is clear that they have played hot and cold during these three years. While they won quite a few series, they lost at home to India, could only draw in Bangladesh, and could not convert the 2-2 draw in the Ashes series to a 3-1 win. They started and finished well but had a very poor 2018. Their four wins on the trot at the end of the period, all achieved inside four days, in addition to giving them 200 WTC points, pushed their average MRP up by over 0.20 points. Over the next year they are likely to mount a strong challenge to India's position. Anyone putting down money on any scenario other than an India-Australia WTC final could very well end up saying goodbye to their hard-earned money. A combination of lopsided match points, scheduling, and a slew of home series for the stronger teams has made the WTC's League phase a virtually dead one only six months after the start.
The next two teams, England and South Africa were highly inconsistent. South Africa started strongly but had a horrid 2019, including an awful home loss to Sri Lanka. England alternated good wins with poor losses and were fortunate to draw their home Ashes series thanks to the heroics of one player. A series of good results would certainly move these teams up. South Africa created some daylight between themselves and England with their Boxing Day win. It is clear that England's red-ball form is scratchy - as is South Africa's. Only the top three teams are strong in both forms of cricket.
The rest of the table is predictable. Pakistan's decline is alarming but understandable since they do not play many Tests and virtually played nothing at home - except towards the end of the year. Kudos to Sri Lanka for playing the series in Pakistan. I hope other teams also start following suit.
Among these teams, three have had their moments - Bangladesh's draws against Australia and Sri Lanka, Pakistan's win over England, and West Indies' win over England. But these have been few and far between. They have to play out of their skins to move to the top half. Sri Lanka have had the most draws and losses.
It is interesting to see that the Overall Result Index %, which works on the no-frills 2-1-0 basis, also matches the Ratings points average totally.
It is nice that the top two teams in the ICC Test rankings table finished in the same positions here as well. It is a vindication of the simpler ICC rankings methodology and my more complex simulation process. I would also say that the much maligned ICC Test rankings process is not that flawed. (Michael Vaughan, please note.)
It looks like the series to savour is the forthcoming one between New Zealand and India. India's resurgent pace attack could cause the New Zealanders some problems, although the home side will be confident of defending their fortress; if only they can find an opening batsman who averages more than 8.5 in his last ten outings. Poor Jeet Rawal. He is playing the way David Warner played in England. Since writing this, it is nice to see the emergence of Tom Blundell.
At the end of the current WTC cycle in 2021, I will do a follow-up article on the WTC using this methodology to see how the standings in the two very different systems compare. I will make sure that I include only the WTC series of Tests.
An important note to readers. I will be starting major revamp of my Golden Willow 25/Red Cherry 25 work in the new year. Most of the tweaks are finalised and have all been summarised in my previous articles. I am not sure when this work will be over - considering that it has to be done in parallel with my regular work. I will keep you informed of the progress or lack of.
I have also decided this is the year when I finally give up on the T20 format. It's good that the ICC has given international status to all matches played between countries, however, when I see that over 300 T20Is were played in 2019 (80 in 2018), I get the shivers. And when I see that matches between Luxembourg and Turkey, Samoa and Vanuatu and Gibraltar and Portugal are T20Is, or when I see Czech Republic winning by 257 runs or Austria taking 16 balls to score 33 and winning with 104 balls to spare, I throw my hands up and give up. My T20I database has stopped at the count of 749 matches and will not move further. I will not be doing any more T20I analyses, even around the next T20 World Cup.
My next article will be a comprehensive analysis of the results achieved in Tests across the years. In addition, I will also look at the duration of the decisive Tests during the past two decades, from points of view of both days played and overs bowled. Let us see whether the conclusions there will support the proposal for four-day Tests as envisaged by the ICC.
Email me with your comments and I will respond. This email id is to be used only for posting comments.