Remember the semi-final of the 1992 World Cup between South Africa and England? Near the end of the match, rain interrupted play and when the players returned, South Africa were asked to score 22 runs of one single ball in order to win the match. That was a silly rule, and it has been changed since. The 2007 World Cup still contains a silly rule, and I hope the following set of circumstances don't come up to show it. I also hope the ICC read this and they change the rule for next time.
Let's have a look at the situation in group B after three matches:
Let's further assume that:
Not a very unlikely scenario, is it?
Then the three Asian sides will all finish on 4 points and the NRR rule will decide who will qualify to the Super Eights.
Now don't misunderstand me, I agree that it is right that the margin of winning against Bermuda should decide the group. If Bangladesh scrape through in the 48th over, they don't deserve to be called a test side, let alone a super eight one. So let's assume that Bangladesh also thrash Bermuda. How many runs would they need?
In order to calculate this, we need of course to also know the scores of the two other matches that have to be played. So let's assume that both those matches end like the Bangladesh - India one: the side batting second winning with 9 balls to spare. In fact, let's assume that the scores are exactly the same.
In that case, all three sides have scored and conceded 383 runs in 98.3 overs in the two "Asian" matches.
India then have a NRR of : 796/148.3 - 539/148.3 = +1.731
Sri Lanka have a NRR of : 704/148.3 - 461/148.3 = +1.636
So Bangladesh need a NRR of +1.637 in order to qualify. It seems logical to decide that they should need to win by 244 runs (one more than Sri Lanka), but it's not as easy as that.
Because suppose that Bermuda win the toss, and that they have the audacity (or, dare I say it, the Sri Lankan Rupees) to bat first, then what will happen?
OK, have a guess: suppose that Bermuda score just 78, as they did against Sri Lanka, in how many overs should Bangladesh reach that target in order to qualify? Have a guess?
no, you guessed wrong, it's fewer
even fewer than that
minus one ball !
yes, count with me: Bangladesh would have conceded 461 runs in 148.3 overs (3.104), so in order to have a NRR of +1.637, their own runs have to have come at a rate of 4.741. They have 383 already, and they can score a maximum of 84 (a six when the scores are tied), so they need to do so in 98.2 overs. And they have played 98.3 already!
The reason for this is that the three matches are not counted equally. Completed matches are counted for 100 overs (or almost), but rain affected matches, or matches in which the target is quickly reached, only count fractionally.
A solution? Yes, easily: if all results are translated into a run difference over 50 overs, then all we need to add are the run differences. Bangladesh would be credited with 198 runs in their match against India (192/48.3*50) and a win of 7. (so would the other two matches). In the last match, Bangladesh would need to outscore Bermuda by 244, but they could do that either in real terms or by scoring the 80 runs needed in 12.2 overs (50/(244+80)*80).
In fact, I would think some form of D/L adjustment would be preferable to a simple runs/over formula, but you understand what I mean.
Here's another example:
After two matches each, South Africa have performed better than Australia. They've beaten Scotland (the RSA total at the end of their match would have beaten a Scotland total of 457 in a D/L sense) and the Netherlands more heavily, as is witnessed by their respective NRR at this moment (+5.009 for RSA, +4.320 for AUS). However, if they would happen to tie their match, look what would happen to their RR's:
| RSA and AUS tie their match with runs each: | 200 | 220 | 225 | 230 | 250 |
| RSA NRR: | 2.838 | 2.872 | 2.880 | 2.889 | 2.992 |
| AUS NRR: | 2.880 | 2.880 | 2.880 | 2.880 | 2.880 |
So, while RSA are better than AUS over the two other matches, when the tie is added, RSA can drop below AUS!
Written : 2007-03-20 - Last modified 2007-03-21