The bowlers to fear with the new ball

Last week's column had analysed the performances of teams and bowlers during the most frenetic stage of a one-day international - the last ten overs. This week the focus shifts to the start of the innings, an aspect which is likely to be vital in the World Cup on tracks which are unlikely to provide much cheer to the bowlers. The days of openers playing themselves in and exercising early caution have long been replaced by a whole assembly line of dashing, throw-caution-to-the-wind style of openers, making the new-ball bowlers' task a pretty tough one as well. The introduction of the Powerplay rule - with only two fielders allowed outside the cordon in 20 out of 50 overs - has only made the bowlers' job even more unenviable.

If the last ten overs are primarily about keeping the runs in check, the key to using the new ball is the ability to nail wickets. The team that has managed this better than any other over the past year is also the side that recently usurped the top spot in ODIs - South Africa concede just 27 runs per wicket in the first 20 overs, a clear four runs better than second-placed Australia.

England may be on a high after their recent CB Series triumph, but they clearly have plenty of work to do as a bowling unit at the start of an innings. Not only have they been sorely lacking in wicket-taking abilities, their bowlers, along with those from Pakistan, concede the most number of runs in the first 20 overs.

The next table illustrates why South Africa have been such an effective bowling unit in the last year. Their new-ball attack of Shaun Pollock and Makhaya Ntini has been more than a handful for most opposition batsmen, with Pollock in especially devastating form. The Numbers Game column a couple of weeks back had celebrated his return to the top of his form as an ODI bowler, and he has kept that form going. As the list below shows, Pollock and the new ball form a particularly irresistible combination. His average is impressive enough but his economy rate, at 2.76 runs per over, is truly staggering.

The highest number of representatives in the top ten - in terms of averages - is, surprisingly, from India. Munaf Patel leads the way, while even Irfan Pathan sneaks in, thanks to his solid displays in the early part of the period under consideration.

Australia only have Nathan Bracken in the top ten but Brett Lee and Glenn McGrath both make it into the top 15. McGrath hasn't been at his wicket-taking best recently, but his nagging accuracy still makes him the second-most economical bowler, after Pollock.

And now on to some of the not-so-smart performers at the start of an innings. Rana Naved-ul-Hasan has had plenty of moments to savour when bowling at the death (click here for more on that), but with the new ball he has been anything but impressive. The top three names in this list all have similar bowling styles - they like to pitch it up, and swing the old ball appreciably. Those are ingredients for effective spells at the death but early in the innings all three have been creamed for runs. It's also interesting that most of the bowlers with high averages have poor economy rates as well, suggesting that the best way to keep the runs in check is to get the wickets.

In response to last week's piece, Mike from New Zealand had asked for a list of run-rates of batting teams in the last ten overs of ODIs, so here it is. Once again, South Africa's excellent form over the last year shines through - they average more than seven per over in the last ten, and nearly 30 per wicket. New Zealand haven't done too badly either, but the numbers for Sri Lanka and West Indies suggest a lack of powerful hitters during the slog overs, something that could cost the two teams in the World Cup.

The team to watch out for, though, is clearly South Africa. They have the firepower in their bowling attack, they have the firepower with the bat, and they have the momentum. All they need to guard against now is their infamous ability to choke on the big occasion.

After last week's column, there was feedback asking about the reasons for excluding Bangladesh and Zimbabwe from the analysis. The reason is this: since 2006, Bangladesh have only played 11 out of 33 ODIs against the top eight teams, while for Zimbabwe the figure is eight out of 34. Playing with minnows considerably skewed the stats and boosted the numbers for the Bangladesh and Zimbabwe players, which is why those games have been ignored completely.