The new, improved batting average

Top Contributors
M

mas cambios

Active Member
The new, improved batting average

The new, improved batting average

The batting average is a simple and convenient way of putting a number to a player’s ability with the bat, but often it doesn’t give the entire picture. One major problem with the conventional average – which is calculated by dividing the total number of runs scored by the number of completed innings – is the way it deals with not-outs.
Click to expand...

From Cricinfo blogs.

Have been having a really good read through the cricinfo blogs and have found some really interesting posts.

Take the one above for instance. It's another attempt to make averages a truer reflection and to take away the benefits some players gain from a large number of not outs.

You'll need to read the blog to fully understand the thinking behind it but it does seem to be a sound way of doing things. I guess that the big question is 'Should this be the norm'? I think so, what about you?
 
Re: The new, improved batting average

man, i dont know about it. it seems strange to add runs on to balance out a not out score.

not much changes and its a lot of of ifs and buts. maybe it can help in settling some arguements but, im not sure.
 
Re: The new, improved batting average

You could be right and you have picked up something which troubled me. Saying that, I guess their would be no point in this exercise without the correction added.

It would be interested to see how this effects lower order players with averages in the low 40's as opposed to the 25 shown in that table. We may well see some startling changes, especially if they have one big score to their name.
 
Re: The new, improved batting average

thats crazy. how can one predict how many runs the batsman would have scored if he stayed not out. there is no way to predict it mate.the current system is good enough.
 
Re: The new, improved batting average

That is very complicated method I dont think its easy to calculate average like that. Statistics will never be able to show how a batsman bats only you can say is that if the player have average over 50 then he is a very good batsman.
 
Re: The new, improved batting average

It's an interesting exercise but the results produced virtually no difference. Stats will never tell the whole story so in my view they need to be as simple as possible.
 
Re: The new, improved batting average

An average should be the average of how many you score in total. Nothing more nothing less. Stats dont lie ;).
 
Re: The new, improved batting average

Timed Out said:
An average should be the average of how many you score in total. Nothing more nothing less. Stats dont lie ;).
Click to expand...

I agree, although that said, the 'not out' is a bit sticky. It's impossible to factor in how many runs a player might have got when he finishes not out due to factors beyond his control.
 
Re: The new, improved batting average

they said that lara scored more runs in less innings than tendulkar...... but laras average is still less than tendulkar after using the new method.....i cant understand how it can be better system.
 
Re: The new, improved batting average

I think I am missing something here, so you will all have to help me out.

I am not a cricketer [as if you hadn't already noticed :D] but I am a statistician.

I was under the impression that the batting average is a 'simple and convenient way of putting a number to a player’s ability with the bat'. Not the ability to stay in. Getting out is surely an aspect of this ability.

Assuming that both Lara and Tendulkar are both playing similar quality matches; both batting against similar quality bowlers, then the fact that Lara has fewer not outs than tendulkar reflects on his ability with the bat, so surely must be taken in to consideration! Remember, this is 'ability with the bat' and not any indication of the style or quality of batsman he is.

Now, as a statistician, I would pay little attention to batting averages in isolation anyway, although averages do not tell the whole story, statistics do:

In ten less tests, Lara had four more innings and scored 746 more runs than Tendulkar.
Or you could say that, in ten more tests, Tendulkar had four less innings and scored 746 less runs than Lara.

This tells me that Lara is a more attacking batsman; he takes more risks so, therefore, he is likely to suffer more 'outs'. He may consider this worth the risk, whereas Tendulkar plays it safer; he takes less risks so, therefore, will be out less but will score less runs.

Now, from a manager's point of view, I need to know the differences between my batsmen; I need to know their 'true' average and I need to know their strike rate. When you start messing around with the figures and introducing 'maybe(s)', what do I look at. It is almost like looking at a great landscape only to suddenly find all the colour green has gone!
 
Re: The new, improved batting average

It's interesting but of little value. You cant try to work out an average by adding on what ifs. Also, it doesn't take into account the match situation, did they get out when chasing or when batting for a draw?

I like the idea of average runs per game though.
 
Re: The new, improved batting average

I think statistics should be kept simple. Two players can have the same average but that doesn't mean they are equally good. The best way to determine a players' ability is to watch him bat.
 
Re: The new, improved batting average

swghayward79 said:
...it doesn't take into account the match situation, did they get out when chasing or when batting for a draw?
Click to expand...

I agree, this is very important, especially for the cricket enthusiast.

For the statistician, this would only be important if the sample was small, [say] 5 tests. In this instance the data population is large enough to show a significant pattern. However, if these were my players or players wishing to join the squad, I would need to see individual test statistics to ensure there were no outliers. For example, were there a couple of times Tendulkar was out for a low score or duck, affecting his total runs and if so, why? Or, did Lara score 800+ runs in a couple of innings against the local primary school!

With a population such as this, a statistical profile is easily composed, if you know what you are looking for, as Will says, stats don't lie; just make sure you do not lie for them :D. From this, together with other variables, a psychological profile is easily compiled.
 
Re: The new, improved batting average

I'd prefer to see them add percentage of boundary's in stats in the future. For example a bloke scores 10,000 career runs id like to know how many he's actually ran. It would be probably best used in 20/20 internationals and domestically. But it would show a players temperament more and somewhat their class.
 
You must log in or register to reply here.
Back
Top