The New Zealand-Australia encounter at Auckland exposes the shortcomings of the Net Run Rate computation method

Looking at the spine-chilling thriller between New Zealand and Australia played at Auckland, Arunabha Sengupta points out the potential pitfalls in using the Net Run Rate method for ranking teams with equal points.

Trent Boult‘s sensational swing, followed by Mitchell Starc‘s ripsote with bouncers and yorkers. Explosive hitting by Brendon McCullum followed by wickets falling like a pack of cards in a tornado. Boult walking out to stave off hat-trick and defeat, nine wickets down, five runs still separating the teams, Starc breathing fire, having set the stump lights flashing with two scorching yorkers at Adam Milne and Tim Southee. MATCH REPORT: New Zealand vs Australia

Add Cricket Country as a Preferred Source

And then the ice-cool Kane Williamson ended it all, with the straightest of bats and the serenest of minds, launching Pat Cummins over long on. New Zealand had clinched it by the wafer thin margin of one solitary wicket, much to the delight of the partisan Auckland crowd. And much to the delight of the cricket fans, the World Cup had at last produced a match of intense battle between bat and ball, which could not have been closer. MATCH REPORT: India beat Ireland by 9 wickets

But, hang on for a minute. Was it that close?

Apparently not, if we consider the method of Net Run Rate computation.

New Zealand finished their chase in 23.1 overs. Hence, their run rate was 6.56. That is in contrast to the Australian run rate of 3.02, since being all out implied that all the 50 overs had to be considered. It gave the winners a whopping boost of 3.54 in their Net Run Rate for the match.

In the opening match of the World Cup, New Zealand had defeated Sri Lanka by 98 runs in a largely one sided affair. But according to the Net Run Rate computation, the match had been a lot closer. In fact, the difference of run rates amounted to just 1.96, which made it way more close than the Kiwi-Kangaroo affair. ICC Cricket World Cup 2015: SCHEDULE & MATCH DETAILS

Or for example take the game between India and Ireland. India won by eight wickets with 79 balls remaining, a yawn of a one-sided romp home. However the NRR difference between the two sides amounted to a mere 1.87, making it nearly and ironically twice as close as the Auckland thriller. ICC Cricket World Cup 2015: Complete Coverage

Even more curiously, suppose Kane Williamson had not managed to clear the boundary. What if he had taken a swing and missed, and the ball had hit his stumps?

In that case, New Zealand would have ended the innings at 146. Hence Australia would have an advantage in NRR for the match of 0.10. Ridiculously, a match that could have gone either way based on one final delivery, would make the NRR swing from 3.54 to 0.10 based on the result.

The problems stem from the computation method which considers wickets sometimes and ignores them otherwise. Even though the Australian innings was considered to be of 50 overs because they lost all their wickets — and so would have the New Zealand innings if they had been all out — since the Kiwis had one wicket standing, the number of overs considered were 23.1.

Something seems to be wrong, very, very wrong.

Let us take a hypothetical scenario to demonstrate how major an effect this can have on a big tournament.

Suppose in the tournament, Australia, India, New Zealand, Zimbabwe and Bangladesh form a group. The final two matches are left to be played: between India and Zimbabwe and New Zealand and Bangladesh. Australia are already through to the next round. India and New Zealand have the same points. To make it simple, let us suppose in three matches, both teams have made 750 runs and conceded 750 runs from 150 overs. Hence, both have a net run rate of zero.

Now, the team with the bigger win will proceed to the next round.

Zimbabwe bat first and score  250 in 50 overs. India score 252 for 2 in 40 overs. Hence the NRR for India at the end of the group stage stands at 0.27. [1002/190 – 1000/200].

Now suppose New Zealand dismiss Bangladesh for 150 runs and just about manage to squeeze home at 151 for nine in 38 overs.  The Kiwis  will end with a  net run rate of 0.29 [901/188 – 900/200]. They will go through to the finals even though India have won by a much bigger margin and the Kiwis  managed to clinch a win after a great great struggle.

Thus we see that  Net Run Rate not only provides academic errors in close, low scoring matches. This method also has a big bearing on the determination of which teams will proceed to the next round as well.

The problem is obvious. Much like the early days of revising targets after rain interruptions, the Net Run Rate computations take into account only the runs scored and overs faced.  The wickets as an important factor are just ignored. This renders it prone to fallacies and especially makes it impossible to compare matches on placid tracks and difficult surfaces.

This is the exact anomaly that in the past led teams, who batted industriously to score 250 in 50 overs, end up defending 100 runs in 20 — a far easier task since the opponents have all the 10 wickets to play with.

This obvious shortcoming led to multiple methods in which the target was revised based on all the available resources — overs as well as wickets.  And the authorities opted for the most robust of them all — the Duckworth Lewis method.

It is incredible that even after the advent of DL, the NRR has been overlooked for so many years, especially with such immense stakes for which big tournaments are contested these days. It does make immense sense to reevaluate this current system of ranking teams with equal points based on Net Run Rate. There needs to be a fresh method which evaluates the difference of the performance of the team and its opponents based on runs scored versus resources expended, where the resources are both wickets and overs.

Just like the Duckworth Lewis method, this will probably not have a simple solution derived by the four basic operators of arithmetic. With two causal variables, there needs to be a degree of sophistication in the involved mathematics to derive fair results. And therefore, there will be plenty of uncomprehending voices clamouring about the sometimes counterintuitive results that this method will generate.

However,  this needs to be one of the major steps required to do away with potentially unjust results in the future. It is far more important to be correct than simple.

(Arunabha Sengupta is a cricket historian and Chief Cricket Writer at CricketCountry. He writes about the history and the romance of the game, punctuated often by opinions about modern day cricket, while his post-graduate degree in statistics peeps through in occasional analytical pieces. The author of three novels, he can be followed on Twitter at https://twitter.com/senantix)