CW Survivor WEEK TWO - W. Younis is off the cricket pitch!

[Paid The Umpire]

Week One Vote:
W. Younis (PAK) is out, bowled for 10th.

The challenge for Week 2:
The player with the lowest bowling average in Test Cricket, will get imunity:

Winner:
M. Marshall (WI)

The Vote:
Anyone from this list..
R. Hadlee (NZ)
Wasim Akram (PAK)
C. Ambrose (WI)
G. McGrath (AUS)
DK. Lillee (AUS)
A. Donald (SA)
J. Garner (WI)
M. Holding (WI):baddevil:

[Simon]

i already know who i want to win, so im just voting players off until the end. So i voted for Wasim.

[marc71178]

I'm going to vote Marshall, just for fun!

[Rik]

Sorry Dennis you've got my vote again!

[royGilchrist]

Posted by Paid the Umpire earlier...
R. Hadlee (NZ) - 95.08
Wasim Akram (PAK) - 99.88
C. Ambrose (WI) - 94.80
G. McGrath (AUS) - 94.86
M. Marshall (WI) - 90.11
W. Younis (PAK) - 93.10
DK. Lillee (AUS) - 98.94
A. Donald (SA) - 92.94
J. Garner (WI) - 92.46
M. Holding (WI) - 97.87

Paid the umpire, I think you have got the RPO formula wrong. I re-calculated assuming RPO means runs per 100 balls and here are my results, also please note that you had two glaring ommissions, S Pollock and Imran Khan who stack up very well against the rest. Also I used only fast bowlers with atleast 50 tests, there might be more ommissions (anyone?).

1 marshall 112.50
2 garner 113.08
3 s pollock 113.16
4 ambrose 114.02
5 g mcgrath 115.54
6 donald 116.60
7 hadlee 117.01
8 imran k 118.79
9 w younis 120.20
10 m holding 121.12
11 w akram 121.49
12 lillee 121.93

Now personally I always am very cautious about this kind of an index. First of all one problem that I see with this right away is the ranking of Garner and Waqar. I think Garner is over ranked and Waqar under ranked. Waqar has the strike rate of freaking 43!! about 3.5 points better than the next closest Marshall, there is no way he can be on number 9.

To make my point I did an analysis of two arbitary bowlers with fictitious bowling figures for an innings:

arbitrary 1: overs=20, runs=90, wkts=5
arbitrary 2: overs=20, runs=40, wkts=2

I think a cricket analyst can easily say that bowler 1 did much better, (it could only be different in special circumstances when for those few times in tests that runs are at a premium and wkts dont matter at all, third innings and giving the 4th batting team a huge lead, and the bowling team trying their best to curtail runs for example). But the index values are as follows:

arbitary #1 117.00
arbitrary #2 113.33

This should prove that the method of looking at s/r, ave. and rpo with equal weights is flawed.

Now I took the liberty of making another index, whihc only considers s/r and ave., equal weights:

w younis 66.41
marshall 67.71
donald 69.28
garner 71.84
haddlee 73.16
g mcgrath 73.43
m holding 74.61
s pollock 74.91
ambrose 75.56
lillee 75.94
imran k 76.35
w akram 78.27

But as you guys say that economy rate does have an impact and it cannot be ignored totally, and I totally agree. So I made a third index, in which I included all three s/r, ave. and rpo but gave 3 times more weight to s/r and ave than rpo:

marshall 35.42
w younis 36.14
donald 36.45
garner 36.68
g mcgrath 37.49
s pollock 37.57
haddlee 37.62
ambrose 37.88
m holding 38.62
imran k 38.79
lillee 39.12
w akram 39.72

I think this is the best off all the three.

Comments??

[luckyeddie]

That's really interesting

I'd love to see the actual formula you have used (email or U2U) so I can analyse it against (if you don't mind):

a) some other players and/or performances
b) the stats other 'informed' cricket analysts use.

You could be on to something - I've always thought that some of the other ratings systems left a lot to be desired.

Perhaps we could publish our 'own' lists as a means of identifying ourselves as a somewhat more-informed site?

[royGilchrist]

Here u go LE, I thought it will be appropriate to give it here so eveyone could see it, and hopefully improve upon it, or maybe find any mistakes.

So for example, a bowler bowls 10 overs, gives 50 runs and takes 5 wickets. I had given 3 different indices above, I will show details of how to calculate all three.

#1 Paid the umpire came up with this, it includes all three avg., s/r, and RPO. BTW, just to remind u, RPO means runs per 100 balls.

avg = runs/wkt = 50/5 = 10
s/r = balls/wkt = (10*6)/5 = 12
rpo = runs/100balls = (50*100)/(10*6) = 83.3

So his overall average becomes: 10+12+83.3 = 105.3

#2 This is the initial method I came up with, which disregards RPO, and gives equal weight to s/r and average:

10+12 = 22

# For this one I took all three but gave only one third weight to RPO (as I feel its not as important in test matches):

((3/7)*10) + ((3/7)*12) + ((1/7)*83.3) = 21.33

I persoanlly think that the third one is on the right track, but I still think that average is more important than strike rate in test matches, so that should be give a higher weight. So maybe something like weights of 5, 4, 1 to avg., s/r and RPO respectively. I am not sure what will be the best way to determine these weights. If someone has any ideas they are most welcome, I will try to come up with some justification of my own.

[Paid The Umpire]

Sorry guys but my system is fine :O

It works by thinking thus:

A great bowlers average is 25
A great bowlers Strike Rate is 50
A great bowlers Runs per 100 balls is 50

By halving RP100B you get a total of 100.

Therefore any fast bowler below 100 is a sexy biatch and a great player.

[Tim]

Therefore any fast bowler below 100 is a sexy biatch and a great player

lol, that has to be one of the great quotes in the history of cricketweb.

[Gotchya]

This is surprising McGrath with 14 votes :wow:

roy : I think that third list justifies a lot of things, ofcourse you cant account for everything but thats a fair way of measure.

[Anil]

Originally posted by Gotchya
This is surprising McGrath with 14 votes :wow:

McGrath is a great bowler, but he cannot in any way be called the greatest. So, ideally, if not this time, he should be voted out in the coming weeks.

[Gotchya]

Yes, but thats two modern era bowlers out of the list. Whats the implication ?

[Anil]

Originally posted by Gotchya
Yes, but thats two modern era bowlers out of the list. Whats the implication ?

I don't think there is any negative bias towards modern bowlers as such. My only question is why did PTU provide immunity to Marshall just because he has the lowest average? If that is the case, why doesn't he go ahead and declare that in his opinion, Marshall is the winner? If stats provide the only true pointer to the greatness of a fast bowler, just rank them according to their averages and be done with it. There are so many other varying factors which made them great. I had listed some of them in a previous post. Their attitude, the grounds and the weather conditions under which they played, the bowling attacks they were part of, the batting strengths of their teams, the fielding that supported them, the injuries they carried..... Of course, a lot of subjectivity comes into the ranking if all these are taken into account, but in a way, that is what makes such a study fascinating.

For example, Dennis Lillee is considered as the greatest fast bowler by a lot of knowledgable cricket critics and fellow fast bowlers for several good reasons. If we rank him according to pure stats, he doesn't stand a chance in this poll. When Lillee was asked who was the greatest fast bowler he had ever seen, he pointed to Andy Roberts and he is not even in this poll because his stats don't let him into this elite company. The Pak bowlers Akram and Younis toiled for long on featherbed tracks of the subcontinent and still came up with exceptional averages, shouldn't that be taken into account while deciding their greatness?

[royGilchrist]

I agree with aal the stuff u said Anil, but I do not agree with ur earlier comment about McGrath.

Why can he not be the greatest? He has stellar stats, has helped Aus dominet like few teams have done before, and has done exceptionally well in tough conditions like India. Im not saying he is the best ever, but he surely is in the hunt like many others.

[Rik]

Sexy Biach...depends on your gender or sexuality...:saint: