Eden Gardens, World Cup semi-final, 2023. At 119-4 in the 31st over, 250 did not seem improbable for South Africa, but Travis Head struck with two consecutive balls. This would have hurt any strong side, but it probably hurt South Africa more – for it brought Gerald Coetzee to the middle.

While not exactly a walking wicket, Coetzee is not Don Bradman either (at the time of writing, he is yet to reach 25 across 22 international innings), but he was the best of South Africa’s bottom four.

Miller now had little option but to nurdle the ball for singles and eliminate all risk. He eventually made 101 in 116 balls, the slowest of his ten highest ODI innings till date. South Africa reached 212. They had Australia down to 193-7, but an injured Kagiso Rabada could not complete his quota, and Mitchell Starc and Pat Cummins seized the advantage.

South Africa’s eight and nine were outplayed by Australia’s eight and nine that day. South Africa had a specialist bowler more, but Head and Maxwell compensated that gap.

Three days later in the final at Ahmedabad, India went from 76-1 to 81-3. They had been 2-3 against the same opposition earlier in the tournament, but on that occasion they were chasing only 200. Virat Kohli and KL Rahul could afford to take their time.

It was different this time. Not only were India batting first but they had also been forced to leave out Shardul Thakur after Hardik Pandya’s mid-tournament injury. Mohammed Shami’s superlative bowling had hidden Kuldeep Yadav’s promotion from nine to eight. Kuldeep has a first-class hundred, but he does not have an international fifty. He can at best hang around. India were also forced to pick Suryakumar Yadav, a T20 giant but of little ODI credentials.

Kohli and Rahul, both capable of scoring quickly, thus had little option but to take it deep. The match was lost – because India did not have a No.8, let alone a No.9.

Twice in the space of fours days could Australia choke batting line-ups with strong, explosive top sevens but little thereafter. They made early inroads, then cut off run-scoring areas in the middle overs. If these triumphs were not obvious, it was probably because they happened in the first halves of matches.

Of course, there are myriad reasons for a team losing World Cup knockout games. Yet, India and South Africa decided that these needed to be addressed. The Champions Trophy squads revealed that.

Not only have India included Axar Patel but they have often used him up the order, even if it was to break the monotony of right-handers or win match-ups against specific spinners. They also have Washington Sundar in the squad. South Africa, too, need not end at seven anymore: along with Marco Jansen, they are now fielding Wiaan Mulder who is in the side effectively ahead of Tabraiz Shamsi. An all-rounder, who is neither a top class top-six batter nor an elite frontline seamer is preferred to the specialist spinner.

Both teams have fallen back upon the same blueprint, and there is nothing new about it. The strongest ODI sides have bolstered batting even if at the cost of a bowler. Whenever they had a choice, they always preferred to bat deep.

The one constant in ODIs

The West Indies were 50-3 in the 1975 World Cup final, worse than India’s start in 2023. But their No.6, while still new, was one Viv Richards, followed by Keith Boyce, Bernard Julien, and Deryck Murray. Clive Lloyd had no hesitation in going berserk despite knowing his side had 60 overs to bat. He smashed 102 in 85 balls, the West Indies made 291-8, and that was that.

Four years later, England’s ploy to bolster their batting (and use part-timers as the fifth bowler) failed spectacularly in the 1979 World Cup final, first against Richards and Collis King, then due to the questionably slow opening stand.

But teams did not deviate. Balwinder Sandhu, the Indian No.11 at the 1983 World Cup, averaged more than 30 in Test cricket, while Mohinder Amarnath and Kirti Azad shared fifth-bowler duties. In 1987, Steve Waugh, Simon O’Donnell, Greg Dyer, and Craig McDermott (promoted to two-drop in the final) gave Australia a solid cushion. After promoting Wasim Akram in the 1992 final, Pakistan had Saleem Malik, Ijaz Ahmed, and Moin Khan from seven to nine. And in 1996, Sri Lanka had Test opener Roshan Mahanama at seven, followed by Kumar Dharmasena and Chaminda Vaas. One can go on.

In each of these cases, the teams could have added a bowler. In each case they chose not to.

Yet again, this is a sample of four games spread across 13 years, over the course of which ODI cricket changed substantially. To dig deeper, we need to go beyond the games, or even beyond the tournaments. Let us begin with 1999, the year when the ICC assumed control of the World Cup.

Let us gauge every top team, not only at the World Cups but in the 730 days (two years) prior to that. It is a fair assumption that two years provide an indication of how the teams had been doing as well as the team structure at the World Cup (Kenya made it to the semi-finals in 2003, but had played only a handful of ODIs against the best sides ahead of the World Cup).

Let us now ask ourselves two questions.

Did the best batting sides end up in the semi-finals?

What do we notice? With the exception of 2011 (a year with quarter-finals, where one defeat could deny a top side a semi-final berth), the top four have often almost always been from the best five batting sides in the world, if not from the top four.

A case in study is England. Until 2015, their average was always below 30 and strike rate below 85. Then, in the two years ahead of the 2019 World Cup, the numbers shot up to 41 and 101 – and the results showed. Ahead of 2023, they did keep up with the strike rate, but the average fell drastically. They were still scoring as quickly as before, but they were also losing wickets.

Is it true for the best bowling sides as well?

Not quite, as we can see. To begin with, the difference in bowling averages (or economy rates) is not as stark as the batting numbers.

Little separated New Zealand, semi-finalists in 1999, from the West Indies, who did not finish in the top six. India, runners-up in 2003, actually bowled worse in the lead-up than several sides. The 2011 numbers seem to go the other way. The 2015 numbers are too uniform for a conclusion. South Africa, the best bowling side before 2019, did not even come close to the semi-final. And excellent pre-2023 bowling numbers helped neither Sri Lanka nor Bangladesh.

So what is all this about?

The idea is to establish that batting strength is more relevant to a team’s success than bowling, When there is a choice – it is a fair assumption that everyone saves their best strategies for the World Cup – the most successful teams swap bowlers for batters, or at least bowling all-rounders for batting all-rounders.

They have been doing that for decades now. That is also why India and South Africa have bolstered their squads with all-rounders this time.