Saturday, 1 January 2011

9 times out of 10 you should bat first?

I once head an old cricket follower suggest that there was a saying that if you won the toss, 9 times out of 10 you should bat first, and the tenth time you should think about putting the other team in, and then bat first.

However, as someone who has repeatedly watched the New Zealand cricket team bat first, struggle to post a competitive score, and been badly beaten many times, this saying didn't ring true to me.

So I did some digging into the stats to see first if my memory of New Zealand doing poorly when batting first was accurate or not, and secondly to see if New Zealand were any worse at batting first than everyone else.

The first thing that I did was look up all countries results over the last three years, in all formats (I excluded any team that either batted or bowled less than 20 times in the last 3 years). The results were interesting.

When comparing wins per loss for batting first and batting second the results came out like this:

Team W/L Batting W/L Fielding Ratio
Kenya 0.16 0.5 3.125
West Indies 0.24 0.52 2.167
Bangladesh 0.32 0.55 1.71
England 1.2 1.77 1.475
India 1.4 2.05 1.46
Zimbabwe 0.44 0.58 1.32
Ireland 1.33 1.41 1.06
New Zealand 0.84 0.83 0.99
Australia 1.75 1.62 0.93
Sri Lanka 1.46 1.33 0.91
South Africa 3.09 1.89 0.61
Pakistan 1.34 0.73 0.54



The interesting thing here is that most teams are actually better at fielding first, rather than batting first.

My theory was the there was some players who found it difficult to know how to bat when they hadn't seen another team play on a pitch. The next step was to look at team batting averages batting first and second. This time I only included matches between teams that had played test cricket during this period (excludes Zimbabwe). I also excluded innings 3 and 4 from my calculations, as I was more interested in seeing if the psychological impact of not knowing how a pitch was going to play could be measured statistically.

I found the teams batting average when batting first, then subtracted that teams bowling average bowling second. I then did likewise for batting second and bowling first. The larger the difference the better a team is performing.

TeamDifference batting firstDifference batting secondDifference
Sri Lanka5.82-1.267.08
Pakistan0.99-5.156.14
Bangaldesh-17.38-16.89-0.49
England3.524.13-3.28
New Zealand-4.14-2.45-1.69
India0.774.05-3.28
Australia5.149.29-4.15
West Indies-14.88-9.89-4.99


This shows only Sri Lanka and Pakistan being better at batting first, although both of them are much, much better at batting first. (Interestingly Sri Lanka is better at batting second away from home, but they score so well batting first at home that it skews the figures)

The results were quite surprising. They showed that New Zealand were actually roughly in the middle of all teams when it comes to the difference between batting first and batting second. However it does make the whitewash of NZ by India a little more understandable, given NZ batted first in 4 of the matches.

1 comment:

  1. Very interesting analysis. I am not sure that all types of the game present the same challenges when it comes to the decision whether to bat or field first. I wonder if you could produce the same stats just for Tests?

    ReplyDelete