Friday, 1 March 2013

Batting first in ODIs revisited

Ravi Bopara - better batting second
My first ever post on this blog was about the benefits of batting or fielding in ODI matches. There's been over 2 years pass since that article so I though it was time to look back and revise the numbers. It's also interesting to see if any of the trends have changed with the innovations in the game in the last 2 years.

I've had to change my criteria somewhat as this search is over roughly 2 years, rather than 3, and hasn't included world cup qualifiers, so a lot of teams have played significantly fewer games. As a result I decided to look at any team who had played at least 15 matches. It means that the results for Ireland, Zimbabwe and Bangladesh are not quite as significant as I would have liked, due to the small sample sizes. However, it's better to have them in, and be able to draw conclusions than to not have them there at all.

Here are the numbers:

TeamMatchesW/L BattingW/L FieldingW/L Ratio
England 530.782.000.39
New Zealand 380.641.160.55
Bangladesh 290.420.720.58
Sri Lanka 660.751.280.59
Australia 601.572.330.67
South Africa 311.281.800.71
India 581.441.900.76
Pakistan 521.552.000.78
West Indies 530.640.760.84
Zimbabwe 230.370.331.12
Ireland 160.800.501.60

If the results were emphatic last time, they are remarkably so now. No team who has played more than 23 matches has a better record batting first than fielding first. Of the top 8, only West Indies does not have a winning record when bowling first.

New Zealand and England are the two teams that have the biggest difference between batting and fielding, so it's unsurprising that in the recent series every game was won by the team batting second.

The question also arises, is the difference due to the batting or the bowling. Are teams better at batting second, or are they better at bowling first. To assess this I looked at the whole team batting index for each team batting first and second.

TeamMatchesBatting firstBatting secondRatio
Bangladesh 2916.0023.340.69
Australia 6026.8533.890.79
England 5328.0632.640.86
Ireland 1620.9123.350.90
Sri Lanka 6625.6028.580.90
West Indies 5321.9224.020.91
Pakistan 5223.3025.410.92
New Zealand 3828.0227.391.02
India 5834.4230.281.14
South Africa 3132.7126.481.24
Zimbabwe 2323.8718.381.30

We can see that New Zealand and South Africa are both actually better at batting in the first innings, despite being better at winning when batting second. This is very interesting, as it implies that for these teams the big difference is not the batting, but rather the bowling. Most teams are still better at batting second, but the difference isn't nearly as profound.

It also made me wonder about the batsmen. Who had the biggest difference between batting first and second in the last couple of years. I looked at every batsman who had scored at least 200 runs in each innings during the time period. As a result I found the batsmen who favoured each innings the most.

Bating first:
NameBatting 1stBatting 2ndRatio f/s
V Sehwag (India) 101.3614.866.82
SR Tendulkar (India) 52.2822.422.33
RA Jadeja (India) 35.6016.322.18
JH Kallis (SA) 39.3618.712.10
MS Dhoni (India) 70.6234.192.07
IJL Trott (Eng) 49.9027.221.83
BRM Taylor (Zim) 48.0229.591.62
JP Duminy (SA) 39.7024.521.62
Mushfiqur Rahim (Ban) 31.8119.851.60
Mohammad Hafeez (Pak) 30.6119.531.57
Younis Khan (Pak) 27.7319.361.43
HM Amla (SA) 58.6541.511.41
HDRL Thirimanne (SL) 27.7719.811.40
LRPL Taylor (NZ) 46.0932.981.40
KA Pollard (WI) 34.2625.761.33

Batting second:

NameBatting 1stBatting 2ndRatio s/f
RS Bopara (Eng) 15.9345.582.86
SR Watson (Aus) 28.4974.072.60
AJ Strauss (Eng) 22.2352.432.36
MJ Guptill (NZ) 22.5752.072.31
V Sibanda (Zim) 19.2039.172.04
Misbah-ul-Haq (Pak) 21.8944.002.01
RR Sarwan (WI) 18.6336.451.96
F du Plessis (SA) 18.1932.921.81
MN Samuels (WI) 18.6233.621.81
Shakib Al Hasan (Ban) 20.7036.611.77
EJG Morgan (Eng) 28.6147.121.65
AB de Villiers (SA) 64.6298.641.53
V Kohli (India) 35.9053.501.49
LD Chandimal (SL) 20.0828.381.41
DJ Bravo (WI) 15.2421.421.41

There is a theory that it takes a lot of experience to learn how to bat well in the first innings. Having Sehwag, Tendulkar and Kallis all in the top 4 probably backs that up. The numbers put up by Sehwag in the first innings are extraordinary. He's scored 655 runs, at an average over 80 and a strike rate over 120.

While the adage might be that 9 times out of 10 you should bat, the truth is that it's probably more sensible to have a bowl.


  1. Two points, though both minor:
    Firstly, I think the adage for batting first only applies to tests, as it gives you the best chance of forcing the other team to either chase for the win or defend for a draw, both very difficult on wearing pitches.
    Secondly, rather than counting runs scored in each innings, wouldn't it be better to use the number of innings themselves as a cut off point? Say, 5 in each? Some batsmen may be so poor at batting in one innings rather than the other that they have been cut out of your study through not scoring 200 runs in either first or second innings.

  2. You make a couple of good points. That saying is mostly about test cricket, but it doesn't stop it being wheeled out regularly by TV/radio people at toss time.

    With the batsmen I wanted to make sure that I canceled out the effects of one big innings. The limit of 200 runs was low enough that most actual batsmen made the cut.