• Welcome to the Cricket Web forums, one of the biggest forums in the world dedicated to cricket.

    You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join the Cricket Web community today!

    If you have any problems with the registration process or your account login, please contact us.

**Official** Paktel Cup (Pakistan, Sri Lanka, Zimbabwe) Thread

marc71178

Eyes not spreadsheets
Richard said:
The economy-rate is hardly suggestive of an ODI-class bowler - given that you can expect to be about 0.1-an-over more expensive in international than domestic cricket.
0.1?

Based on what?

Another arbitrary number no doubt.
 

Yasa

Cricket Spectator
well pak won and erm..younis khan kept wicket well and scored 70 odd and got motm...wat a crap player eh ! :blink:

afridi won it with a six, malik scored 80..
 

Richard

Cricket Web Staff Member
marc71178 said:
0.1?

Based on what?

Another arbitrary number no doubt.
An approximation.
Based on the fact that the domestic level is a level below the international, and no-one can be expected to do quite as well at international as domestic level.
 

Richard

Cricket Web Staff Member
Yasa said:
well pak won and erm..younis khan kept wicket well and scored 70 odd and got motm...wat a crap player eh ! :blink:

afridi won it with a six, malik scored 80..
And Prosper bowled well yet again.
I really like the look of this fellow - he looks almost impossible to get away on a slow\turning (or both) pitch.
 

marc71178

Eyes not spreadsheets
Richard said:
An approximation.
Based on the fact that the domestic level is a level below the international, and no-one can be expected to do quite as well at international as domestic level.
I agree that, but I'd say it's a lot more than 0.1 an over (which equates to about 2 or 3%) - I'd say there's a bigger gap than that!
 

Neil Pickup

Request Your Custom Title Now!
0.1 an over is the equivalent of 1 run per spell of ten... not statistically significant unless you have a very long International career.
 

Richard

Cricket Web Staff Member
It mightn't seem much but it's about the difference if you take a random sample!
A random sample (this is the key) of players who have been successes at both levels.
Not fair to use someone who's got a domestic rate of 4.8-an-over and a ODI rate of 5.7-an-over!
Nor is it fair to use someone who's played 5 ODIs!
 

Neil Pickup

Request Your Custom Title Now!
Define "ODI Success" first, as I'm not bothering if you're going to claim they're not successes...
 

Richard

Cricket Web Staff Member
Very sensible, I was going to suggest something similar. Trouble is, me and marc have a different definition.
Certainly I don't think either of us would dispute that ER under 4.3-an-over is a success. Nor can we use playing regularly as a denominator (Henry Olonga, with his ER of 6-an-over all-but, played for ages).
Personally I've always said that over 4.5-an-over is too expensive (unless you're taking plenty of wickets at an average of under 27), but that doesn't automatically mean that someone going at 4.45-an-over is a success. Indeed, mostly I'd argue that he wasn't.
So shall we say 4.4-an-over? Anything under that, having played, say, 20 ODIs, is a qualified success? Marc should be on in a minute, so we can get his take on it.
 

Richard

Cricket Web Staff Member
I'm no good with strike-rates - I'd say an average of under 27, but if you let me make the guidelines for ERs, you can do it with SRs.
 

Neil Pickup

Request Your Custom Title Now!
Right, limiting the survey to bowlers who played ODIs in 2004, the participants are: Harbhajan Singh, Glenn McGrath, Michael Kasprowicz, Ashley Giles, Muttiah Muralitharan, Dilhara Fernando, Chaminda Vaas, Nuwan Zoysa, Shaun Pollock, Jacob Oram, Jason Gillespie, Ian Bradshaw, Brad Williams, Heath Streak, Andrew Flintoff, Shoaib Malik, Daniel Vettori, Martin Suji,
Ray Price, Chris Harris, Shoaib Akhtar, Darren Gough, Brad Hogg, Shabbir Ahmed, Chris Gayle, Corey Collymore, Makhaya Ntini, Anil Kumble, Abdul Razzaq, Brett Lee, Mohammad Sami, Ajit Agarkar and James Anderson.

Their average List A economy rate is 4.28 and ODI, 4.35. We also tellingly get a Pearson Coefficient of 0.89 and R-squared value of 79%, which suggests that there's a strong correlation between the two, namely that 79% of the variance of the ODI stats is caused by the variance of the List A stats, which is easily significant (but I'd be concerned if good List A economy didn't mean good ODI economy).

However, I wasn't convinced that there was enough data to suggest that the stats "ODI" and "List A" were even significantly different. Reducing them to a T-distribution of 64 degrees of freedom with common variances (having conducted an F-Test to ascertain this case - look all this up if you want to understand the steps), I calculated a T-value of 0.81, which isn't large enough (needs to be 1.66) to prove, statistically that there's a significant difference.

So basically, we're all wrong.
 

Richard

Cricket Web Staff Member
And another good innings from Shoaib Malik at the top of the order.
Of course, it's in the subcontinent so it doesn't prove anything we don't already know, but still - better to play it than not play it.
 

Scaly piscine

Cricket Web: All-Time Legend
Samuel_Vimes said:
So basically, you need more data, is another way to put it.
That's pretty much always the case with statistical analysis - can never prove anything 100% with it, or in this case way less than 100%.

Neil Pickup said:
I calculated a T-value of 0.81, which isn't large enough (needs to be 1.66) to prove, statistically that there's a significant difference.
I'd say there are too many other factors involved to get anything statistically out of the sample you used. In England for example all the national league games are 45 overs (earlier they were 40 overs if my memory serves), so you're gonna get slightly more expensive figures with slightly better bowling averages. If you can get like-for-like figures (ideally where the List A stats are from games in mostly the same country as those ODI stats with both being 50 over matches) for recent previous years you might get something if you can get a big enough sample.

Anyway surprised how easily Pakistan won today.
 

Top