by Arunabha Sengupta
HOME-AWAY, SENA and other CLICHES that masquerade as ANALYSIS
So, recently we had Wasim Raja’s birthday. Hence, as in every year, the records of batsmen against that special group of West Indian pacers were being looked at. Raja of course has by far the best record against those great bowlers, leading to a lot of cognitive dissonance.
Finding Mohinder Amarnath (avg 38) absent from the list of top batters against those West Indian pacemen is always a counter-intuitive jar for most – mainly because of popular memory.
The automatic conclusion was that he did better against the West Indies in West Indies and failed at home (as indeed he did) .
But the argument that followed was because he did well away from home, his numbers should be weighted accordingly.
Fine. How does one deal with this ‘weighted accordingly’ bit?
And again, is it axiomatic that playing West Indian bowlers in West Indies was more difficult than playing them at home?
As it turns out, the answer is ‘No’
From 1976 to 1995, the fearsome WI quartets (or in occasions trios) bowled in 137 matches. We will ignore the Packer depleted attacks between 1978 and 1979…
In 59 Tests at home they conceded 24.40 runs per wicket.
In 78 Tests away from home they conceded 23.79 runs per wicket.
In other words, opposition batsmen had a slightly easier time in WI than in their own backyards against the quartet. Counterintuitive but true.
So, no one should get more credit for better performances in WI than at home.
That does not mean home performances should be given more weightage. There is no significant difference between the home-away numbers.
But definitely away performances need not be glorified as something esoteric.
Why was this so? I have not gone deep enough to get an absolutely certain answer.
But what I do understand is that most of the great fast men of WI during that generation performed better away from home. The tables show as much.
It might have been because some of their own wickets – like Georgetown and Port of Spain – were the slowest in the world.
Or the wickets did not make much difference to these really fast, really great bowlers. Some, like Holding, Roberts and Garner, had distinctly better records away from home. Marshall performed as well at home as he did away.
It does seem the great names really did not care where they were bowling.
But this exercise is also to remind enthusiasts that to arrive at proper analysis one needs to test the basic hypotheses.
There are broadly two categories of cricket fans.
One who flamboyantly declare numbers don’t reflect the entire picture and hence stick to their age-old beliefs in the face of data. They are more to be pitied than censured – and this post is not really for them. They can curl up with memories of their idol of choice and drift off to dreamland.
Then there are the ones who do go by data. Of course that is the way more preferred option. But then there are some pitfalls when the methods used becomes ritualistic, and numbers are forced to reflect what they want to see.
These are occasions when many bring in the same clerical and opinionated habits as the data agnostic souls into their ‘methods of analysis’.
Analysis is about asking questions, not creating axioms.
For example recently someone tried to argue that a particular player during a phase in 1980s,while boasting great home numbers, did not have the best numbers ‘away from home’ …
'Home Away' – another categorisation many take as axiomatic, often in a clerical manner.
Here is where clerical thought process interferes with the analytical techniques.
In this particular case, when one goes into the details he finds that during the mid-80s period in question the ‘home’ series were against West Indies (twice) and Pakistan … and the away were against Sri Lanka and post-Lillee-Chappell-Marsh Australia. It puts the fallacy in perspective.
Granted, this was argued by one of the sloweer thinkers who try to use data with as much finesse as a chimpanzee uses a joystick in a slot-machine all the while thinking the bananas are real.
But even the relatively analytical ones fall for this error.
Home and Away distinction may be a great indicator of quality in most cases. But whether it is really a differentiator in all cases, that is always the first thing to test before we draw our inferences.
If the basic assumption is wrong, then the analysis is shaky.
Cliche is not exactly scientific. “Four legs good two legs bad” is good for Animal Farm, not for analysis. Hence home-away has to be tested as a definite indicator.
Another typical example is the fallacy of SENA countries. The acronym makes great reading … perhaps sends pulses marching along, especially in the subcontinent.
Perhaps it makes sense as well, if you are dealing with the last two decades or so.
However, the acronym has its limitations.
New Zealand was never a difficult country to go to before, say, the 1980s. So, if someone tells me Prasanna was a great spinner because he took bags of wickets in New Zealand that is just poppycock. Every spinner, of note or otherwise. took wickets by the bushel at very low averages in New Zealand till the mid-1970s.
Does SENA always make sense?
Does Home-Away always make sense?
Ask these questions … Verify the underlying hypothesis. With data.
Else the opinions remain as obstinate as those of the first group of fans … only one dips one’s opinion in pseudoscience, making it more impregnable by looking at what one want to see.