ICC Cricket World Cup 2011 / Features

Is batting first such an advantage in Tests?

Since the beginning of 2000, teams batting first have won 217 Tests but lost 250, which goes against the bat-first philosophy that has always existed in cricket

S Rajesh

November 22, 2013

Comments: 10 | Text size: A | A

Graeme Swann trapped Harbhajan Singh lbw for his 200th Test wicket, India v England, 2nd Test, Mumbai, 2nd day, November 24, 2012
Eight of the last nine Tests in India have been won by the side batting second, including two by England last year © BCCI

"Win the toss and bat first" is a refrain which is almost as old as cricket itself. Experts have always reckoned that making first use of a pitch is almost always the right move in a Test match, for it gives the batsmen the chance to bat when the track is supposedly at its finest, and the bowlers the opportunity to make use of a wearing surface on the last day.

In the early days of Test cricket, that was undoubtedly the best way to win a Test match: in 274 games till the end of the 1930s, teams batting first won 117 Tests and lost 81, a win-loss ratio of 1.44. Till the end of the 1970s, that theory still held good: in 867 Tests, the side batting first won 309 and lost only 236, a ratio of 1.30. (Click here for the decade-wise list.) When captains won the toss and batted, they won 268 times and lost 202. It isn't a surprise that the experts still reckon that when in doubt, think about it and then decide to bat anyway, for that used to be the winningest strategy when most of them played the game.

In the last three decades or so, things have changed quite significantly. It could be a combination of several factors: perhaps pitches have more juice at the start than they used to; perhaps they stay true longer; in the last decade and a bit, Tests have also been regularly finishing inside four days, which means the fifth-day pitch doesn't come into the equation at all. Perhaps it's a combination of these factors, and a few others as well, but the upshot is that teams batting first no longer carry a significant advantage in Tests. In fact, since the beginning of 1980, they've lost more than they've won: 390 wins, against 443 defeats, a win-loss ratio of 0.88 for the team batting first. Thanks to this reversal, the overall lead in victories for the team batting first has been cut down to 20: 699 wins, and 679 losses.

The period when it all changed was the 1980s: from a win-loss ratio of 1.42 for the teams batting first in 1970s, the ratio slipped by almost half to 0.72 in the 1980s. More specifically, the change came in the three-year period between 1984 and 1986, when teams batting first won only 15 Tests but lost 37, a ratio of 0.41. (Click here for the year-wise results between 1970 and 1989.) Australia and England were the teams that struggled the most when batting first during this period: Australia had a 1-10 record, losing four against West Indies and three each against New Zealand and England, while England were slightly better at 2-11. West Indies, expectedly, didn't have a problem in either circumstance, with a 7-0 record in 11 Tests when batting first, and 12-2 in 16 when fielding first.

In the 1990s the results evened out, but since 2000 teams batting second have again had an edge. In fact, captains who've won the toss and fielded first have achieved much better results since 2000 (win-loss 88-69) than those who've batted first (148-162).

The percentages column is interesting too: in the 1980s captains were a lot more willing than earlier to field first - the percentage almost doubled - and the move produced willing results too, with the win-loss numbers being 43-23 when captains put the opposition in to bat. Since the 1990s, the decade percentages have stayed around the same, with teams inserting the opposition in to bat roughly once every three Tests.

Win-loss ratios for teams batting first in Tests down the decades
  Batting first Winning toss and batting first  
Period Tests Win-loss Ratio Tests Win-loss Ratio Percentage*
1960s 186 54-43 1.25 156 47-32 1.46 83.87
1970s 198 67-47 1.42 153 52-36 1.44 77.27
1980s 266 60-83 0.72 157 37-40 0.92 59.02
1990s 347 113-110 1.02 228 75-77 0.97 65.71
2000s 464 159-191 0.83 306 107-121 0.88 65.95
2010s 158 58-59 0.98 104 41-41 1.00 65.82
Win-loss ratios for teams fielding first down the decades
  Fielding first Winning toss and fielding first  
Period Tests Win-loss Ratio Tests Win-loss Ratio Percentage*
1960s 186 43-54 0.79 30 11-7 1.57 16.13
1970s 198 47-67 0.70 45 11-15 0.73 22.73
1980s 266 83-60 1.38 109 43-23 1.86 40.98
1990s 347 110-113 0.97 119 33-38 0.86 34.29
2000s 464 191-159 1.20 158 70-52 1.34 34.05
2010s 158 59-58 1.01 54 18-17 1.05 34.18
* Percentage of Tests in which the team winning the toss chose to bat/ field first

During the recording of Match Point, ESPNcricinfo's live show over the India-West Indies Test series, Sanjay Manjrekar, one of the experts on the show, made an interesting observation: he pointed out that the advantages of winning the toss and batting first was over-rated, especially in India. Explaining further, he said that the early moisture on Indian pitches help bowlers - both fast and spin - take a few early wickets before flattening out into a batting surface. The teams batting first thus tend to recover from the early setbacks to post reasonable, but not daunting, totals. The teams batting second play when the pitch is at its best, post huge totals, and the lead is usually enough to take the fourth-innings dangers out of the equation.

The last few Tests in India have followed exactly that template: West Indies lost both Tests batting first (though it's arguable if they would have fared any better batting second, or batting three times in each match), as did Australia four times earlier this year. Even India lost twice to England last year when they batted first. Of the last nine Tests in India, the team batting first has lost eight. (Click here for the list of Test results in India since 2000.)

The table below shows the win-loss record for teams batting first in each country since 2000, and India is right at the bottom of the list: teams batting first have won only 16 Tests and lost 31 during this period. India have done better - they have a 11-6 record - but most of the other teams have struggled. (That's also partially a reflection of how dominant India have generally been at home during this period - their win-loss ratio of 3.27 has been bettered only by Australia at home.)

Surprisingly, the country where teams batting first have won the most is England, where the record is 41-34 in favour of the team batting first. In no other country has the team batting first won more matches than they've lost.

Win-loss ratios by host country for teams batting first in Tests since 2000
  Batting first Winning toss and batting first  
Host country Tests Win-loss Ratio Tests Win-loss Ratio Percentage
England 98 41-34 1.20 62 28-19 1.47 63.27
Australia 78 33-33 1.00 57 23-25 0.92 73.08
UAE 14 5-5 1.00 11 4-5 0.80 78.57
West Indies 66 21-21 1.00 43 14-13 1.07 65.15
Sri Lanka 72 25-27 0.92 48 19-17 1.11 66.67
Zimbabwe 29 12-13 0.92 15 6-7 0.85 51.72
New Zealand 54 16-19 0.84 20 7-5 1.40 37.04
South Africa 69 25-34 0.73 39 17-20 0.85 56.52
Bangladesh 41 14-20 0.70 29 9-15 0.60 70.73
Pakistan 32 9-13 0.69 23 6-8 0.75 71.87
India 69 16-31 0.51 64 15-28 0.53 92.75

S Rajesh is stats editor of ESPNcricinfo. Follow him on Twitter

RSS Feeds: S Rajesh

© ESPN Sports Media Ltd.

Comments: 10 
. Your ESPN name '' will be used to display your comments. Please click here to edit this.
Comments have now been closed for this article

Posted by Tony on (November 23, 2013, 16:55 GMT)

The only way to conclude will be to compare how many wickets fell in first session viz a viz, total number of wickets fell in all sessions except may be last two wickets of each team. That will tell whether putting opposition in will have the right effect.

Posted by Dummy4 on (November 22, 2013, 20:05 GMT)

Basically this seems to be a case of correlation confused as causation. Playing second leading correlates with winning, but that may not be a causal link. There may be another variable like "being a better team" or "being a home team" that may well be the causal connection. It is also possible that I am missing something basic.

Posted by Tim on (November 22, 2013, 11:59 GMT)

I'm (German and) new to cricket. I didn't really understand that myth in the first place since I thought to know the target in the 4th innings must be the most crucial advantage - so batting 2nd would be best.

Posted by Nirjhar on (November 22, 2013, 9:59 GMT)

Excellent work but a few suggestions/observations:

The author must remove obviously weaker teams like Bangladesh and Zimbabwe. They may lose whether batting first or second.

Secondly the best of WI, Aussie domination must also be separated. Include in total numbers but must talk about it. As an Indian, while we weathered australian domination better than most, England and India suffered most at the hands of WI.

In the 80s when I watched, I do not think it mattered if WI were bowling on Day1 or Day3...they were fast and the pressure was relentless. Gavaskar and co would fight but often lose.

I think for a battery of 4 pace bowlers, the pitch plays no role at all. As long as WI batsman do not collapse, WI would normally win. Pitch and theories matter only when you have 2-3 good bowlers and one of then is a spinner who needs assistance.

With 4 pace bowlers, just bring it on....does not matter. They will win 80%. Pace works in all conditions.

Posted by Rajan on (November 22, 2013, 6:29 GMT)

In test cricket, 2nd and 3rd days are best to bat. Day 1, the pitch is fresh and lively. Day 4 and 5, the pitch should start to crack and therefore support spin. So, this will be a good philosophy for captains to adopt "Bat Second, Bat Long, Bat Once".

Posted by Andrew on (November 22, 2013, 6:27 GMT)

Interesting stats ... Australia had the philosophy to bat first under Steve Waugh and they could post a huge total and then have the opposition skittled out twice ... but then take the games in South Africa where <50 has been posted ... all the teams batting first posted those totals and South Africa on each occasion batted second and won the match ... including when they were skittled out in Cape Town in their first innings to have a deficit ...

Posted by Aaron on (November 22, 2013, 4:51 GMT)

I think that part of it is the psychology of choosing to field and the batsmen getting on top. It destroys the confidence of the team when you choose to field first because you believe the pitch favours bowling, but the bowlers can't get anything out it.

Posted by Mark on (November 22, 2013, 4:33 GMT)

Without a doubt, winning the toss and batting first provides an advantage for the team, but it's up to the team to then capitalise on the advantage. If a team fails to do this, they give up their advantage. The stats merely show that in many cases, teams have given up their advantage of winning the toss and batting first. So this analysis is flawed in my view, because it fails to take into account what really matters, which is whether the team batting first actually took advantage of their position or not.

I would suspect that if a team won the toss, batted first and scored in excess of say 450, the win-loss ratio would be in favor of batting first.

Posted by Tom on (November 22, 2013, 3:56 GMT)

"Surprisingly, the country where teams batting first have won the most is England". Perhaps not that surprising, especially when you consider that Australia is #2 on that list. England and Australia play more Tests than anyone and have done so for longer (especially in the first period you mention to 1930), so the "bat first" mantra will have developed based on those wickets, where it seems to be statistically legitimate.

Moreover those countries also produce a disproportionate number of experts and commentators, and have for much of Test history had the leading international first-class systems so even people who call other places home will have been influenced by Anglo-Aussie domestic strategy. It seems it's not that the mantra's outright wrong; it's just wrongly assumed to be universal.

Email Feedback Print
S RajeshClose
S Rajesh Stats editor Every week the Numbers Game takes a look at the story behind the stats, with an original slant on facts and figures. The column is edited by S Rajesh, ESPNcricinfo's stats editor in Bangalore. He did an MBA in marketing, and then worked for a year in advertising, before deciding to chuck it in favour of a job which would combine the pleasures of watching cricket and writing about it. The intense office cricket matches were an added bonus.

    How Bangladesh is finding and developing its talent

Mustafizur, Mosaddek, Mehidy, Nazmul - where did they all come from? By Mohammad Isam

    It's time to rediscover Test-match batting

Mark Nicholas: England's recklessness in the name of positivity is a sign that the art of batting in the longest format is no longer given due attention

Is it possible for a Pakistani to be a fan of Ian Botham?

Imran Yusuf ponders an age-old question
The Cricket Monthly

    Nottingham's the charm

On tour in the UK, Firdose Moonda witnesses a fine comeback, visits the country's oldest pub, and squeezes in some yoga lessons

News | Features Last 3 days

No stories yet

World Cup Videos