Search the Entire Blog

Wednesday, 31 December 2014

Putting Misbah through the numbers test By Shahan Shahid


Putting Misbah through the numbers test

By Shahan Shahid

When Pakistan hammered New Zealand by a whopping 248 runs in the first Test in Abu Dhabi in November, Misbah-ul-Haq became the country's most successful captain, four years after he took up reins in the backdrop of the disastrous spot-fixing scandal.
The 40-year-old Misbah achieved his 15th Test win with that triumph and overtook his much-storied compatriots Imran Khan (14 wins in 48 Tests as captain) and Javed Miandad (14 wins in 35 Tests).
The number 15, however, says little. It hides crucial details about the quality of opposition faced by Pakistan in these contests. If one were to qualitatively analyse these wins and compare them with wins under other captains, where would Misbah rank? How easy or difficult have his fifteen wins been?
 Misbah-ul-Haq. -AFP
Misbah-ul-Haq. -AFP
Let's put Misbah through the number test and we'll be able to gauge whether the Mianwali statesman is really the greatest captain to ever have donned Pakistani colours.
It is important to note the limits of this analysis though. It focuses exclusively on the Test matches a captain won.

The analysis

A team's strength is determined by three factors: batting, bowling and experience. For all metrics, match-date numbers [Footnote: 1] are used, the rationale being that the quality of teams can vary over time, even if they consist of the same players.
For batting quality, the match-date batting averages of the top six batsmen are averaged. Bowling quality is found by averaging the three lowest bowling averages. And as for experience, the total number of matches played, the total runs scored and total wickets taken by members of the team are summed up independently.
For example, in the first Test match of the recent series against New Zealand in the UAE, Pakistan's “Batting Quality” stood at 45.36, “Bowling Quality” was 29.35 and members of the team had played a total of 218 matches, scored 20117 and taken 125 wickets.
Similar, numbers are computed for each Pakistani captain over the years and how their victories compared to each other.
Here's how the comparison is done.
If a team performs better in a metric, it receives one point: a higher number for “Batting Quality” as compared to the opposing team will mean a score of one for the team.
The process is repeated for the remaining metrics -- “Bowling Quality”, “Total Matches”, “Total Runs” and “Total Wickets” -- to arrive at the "Easy Win Scale".
The “Easy Win Scale” returns values ranging from 0 to 5, 5 being an easy contest and 0 being a very difficult battle.
In the first Test between Pakistan and South Africa in Abu Dhabi last year, for instance, the Proteas started out as the clear favorites; as compared to the Pakistanis, Graeme Smith's men had better batting, bowling, more matches in total, more runs and more wickets.
For Pakistan, this match had an “Easy Win Scale” of 0. With odds stacked against Pakistan, this was a tough win [Footnote: 2] even if it came with one day to spare.
Before we delve into tables and numbers, let's take stock of what we intend to do. The underlying rationale is that a captain needs to be lauded for his efforts if he was able lead his team to victory in a match tipped against his side's favor. One part of what makes a great captain, is not just the number of victories but the ability to lead his team to quality, challenging wins [Footnote: 3].

The table for Misbah's 15 wins

Interestingly, the three matches against England in the UAE in 2012 all get a score of 1. Pakistan's only advantage in this 3-0 clean sweep of the English, was its bowling.
Compare this to the recent Australia series where Pakistan had better bowling as well as a better batting line up, it can be argued, therefore, that the England games were tougher than the ones against the Aussies.
If we were to follow this logic, Misbah-ul-Haq was not completely wrong in not having a clear-cut answer when asked if the 2-0 whitewash of Australia was “better” than the 3-0 thrashing of England in 2012.

Misbah's wins have an average "Easy Win Scale" of 2.4. Let's see how he compares to other Pakistani captains.



Inzamam-ul-Haq's eleven wins overshadow Misbah's victories in terms of 'quality'. 'Inzi' led his side to tougher wins as compared to Misbah. Possible explanations for Inazamam's low "Easy Win Scale" could be that at this time, Pakistan were rebuilding after a poor show in the 2003 World Cup and were among the weaker teams in the circuit. This is the time Pakistan tried the likes of Yasir Hameed, Rana Naveed, Shabbir Ahmed and Imran Farhat.
Javed Miandad and Waqar Younis rank at the bottom of the table, while Imran Khan, who is behind Misbah only by a fraction, comes in at number three.
Looking at 'experience' for the teams Misbah led in these fifteen wins is particularly interesting. The team, on average, has the lowest number of matches in total, lowest number of total wickets (by a margin of around 150 wickets) and is only second to Waqar Younis’s outfit in terms of total runs scored.
As compared to other Pakistani line-ups, the Misbah-led team is, therefore, the most inexperienced.
Leading the 'rookies' to fifteen Test victories, therefore, is a commendable feat.
Waqar Younis's numbers are also interesting. His "Easy Win Scale" nearly touches five. This is partly because all of Waqar's 10 wins came against weaker teams – 6 of his 10 wins are versus Zimbabwe and Bangladesh – and partly because the Waqar-led team has the best “Bowling Quality” (lowest bowling average) and a respectable Batting Quality of 42.

All the great men

A table of all captains with 25 or more matches, regardless of nationality, is shown below. Steve Waugh ranks at the bottom with an "Easy Win Scale" of nearly 5.
Unlike Waqar, Steve Waugh had such a talented eleven playing for him, that no team looked difficult. His team had better batting, more total runs and more total matches in all the 41 he won as captain. For the other two metrics, too, batting quality and total wickets, Waugh's team is rarely outdone: 3 times in batting and only 6 times in total wickets [Footnote: 5].
Surprisingly, Inzamam-ul-Haq ranks at the top with Michael Clarke and Michael Atherton trailing closely behind.
 Inzamam-ul-Haq. -AFP
Inzamam-ul-Haq. -AFP
Michael Clarke's numbers make it evident that the Australian squad under his leadership has been much weaker than the one Steve Waugh captained. Wins under Clarke, therefore, have been harder to come by.
India’s batting prowess and the deadly West Indies bowling attack that Vivian Richards captained are highlighted by the data.
India has the highest “Batting Quallity” and the highest total runs on average while the West Indian team of the 90s has an average “Bowling Quality” of 19.

Where Misbah ranks

Many argue, what Misbah has achieved cannot be put down in numbers. They say it is the stability that he instilled into a broken team and made it extremely efficient.
But where does this analysis leave us then?
Is Misbah the greatest captain Pakistan ever had?
An analysis of wins does not say so. Inzamam-ul-Haq led his side to victory in tougher matches as compared to Misbah.
Nevertheless, Misbah deserves all the credit for the fifteen Test wins against a fairly respectable pool of opponents.
What about Inzamam? Is he the greatest captain to ever to have captained Pakistan.
Clearly, Inzamam led his side to victory in matches that were very tipped against his team’s favor, but for us to conclude he was a 'great', or the 'greatest of all time', all his matches, regardless of result need to be taken into account.
Part II and III of this analysis will, hopefully, fill this void.

Readers are free to download and manipulate source data for this analysis which is available here
[Footnote: 1] Cricinfo shows career averages after the match. This is not ideal -- ideally pre-match numbers should be used -- but, sadly, is the best data available.
[Footnote: 2] Readers may wonder why a binary comparison was made and why how wide the disparities were, was not taken into account. For one, doing so led to wide oscillations between numbers which is a problem when averaging numbers. For example, in Sourav Ganguly's first match as captain, they played a much weaker Bangladesh side. Adding these wide differentials distorts the results for all of Ganguly’s wins. Also, it seemed to be jeopardizing the digestibility of the numbers and so a simpler approach was used.
[Footnote: 3] None of Inzamam's wins have come against particularly poor sides, such as Zimbabwe or Bangladesh. Misbah, on the other hand, has played against poor teams. I am not sure if this pulls Misbah's numbers down. If it does, this is difficult to account for.
[Footnote: 4] None of Inzamam's wins have come against particularly poor sides, such as Zimbabwe or Bangladesh. Misbah, on the other hand, has played against poor teams. I am not sure if this pulls Misbah's numbers down. If it does, this is difficult to account for.
[Footnote: 5] It can be argued that our model puts captains with strong teams at a disadvantage. This is a valid criticism. But, our model, quite deliberately, values a win that came against a tough opponent more than an easy one.

No comments:

Post a Comment