Cricketweb ODI Rankings

[marc71178]

It's simply because this rating system isn't cumulative, in a way it's similar to the way the ICC cut out series after a period of time, leading to a once a year reshuffle.

 

[Pratters]

It's simply because this rating system isn't cumulative, in a way it's similar to the way the ICC cut out series after a period of time, leading to a once a year reshuffle.

Well this rating system has been shown to be pretty dumb for the moment.

CW should take its name away from it and unsticky the thread as far as I am concerned.

This rating system is NOT COOL.

 

[Neil Pickup]

All it means is that India are getting better over time.

 

[Pratters]

All it means is that India are getting better over time.

The more I see it, the more I am convinced they have been getting worse over time.

When the ratings consider a series so long ago, it clearly isnt right when one day cricket is pretty frequent now.

 

[Neil Pickup]

It's not our fault that India and Pakistan refused to play for so long, undermining the system.

I agree that it's deceptive, but is this worse than the ICC system and the way that at an arbitrary cut-off point, the whole thing changes?

 

[Pratters]

It's not our fault that India and Pakistan refused to play for so long, undermining the system.

I agree that it's deceptive, but is this worse than the ICC system and the way that at an arbitrary cut-off point, the whole thing changes?

There should be some thing like which ensures series which were played in the past redcues in importance by geometric progressions.

Wouldnt actually be difficult to devise such a ranking.

One which isnt reflective of the true ranks of teams isnt good enough.

 

[Neil Pickup]

That would nonethleless still cause a large scale ranking change overnight at the arbitrary point whereby you class a series as "old".

We've said from day one that this system works so long as teams play one another on a regular basis, it then captures a genuine moving-average flow on an approximate four-year basis. It is not within the scope of the system to live with exterior interference.

 

[Pratters]

In tests long periods can work. In one dayers it wont. India was a quality one day side in 2003 reaching the ODI final, then was average in 2004, has gone down pretty much after losing in the asia cup and in the one day series.

I will try and help with a better system with you because I dont like CW associating itself with a system which isnt reflective of the current ranks.

Wild shifts because of cut off points isnt acceptable. But that is not a reason you want to put to show the rankings are less reflective.

Maybe the significance of past matches can be reduced over time in geometirc promortions. The Black Scholes formula uses it to value futures and options.

Some thing better than both the alternatives you can think of is surely there.

 

[Duncan]

India 936 (+33)
Pakistan 876 (-31)


What a joke... Pakistan thrashed India and they lose points? I'm sorry but I don't see the logic in this. Sure, they didn't win 6-0 or 5-1 like they probably should have, but 4-2 is still a convincing win, atleast they should have stayed at the same points... not lose points. The only team that should lose points for having 4-2 win is Australia or teams playing against Zimbabwe/Bangladesh.

BTW... On ICC's ranking... Pakistan went up a point to 110... the highest it's been since December 2003. India went down to 97, the lowest it's been since the point system started in November 2002.

 

[psxpro]

I agree its stupid how pakistan went down but duncan i disagree with you too, why should Even australia go down for winning?

 

[Duncan]

Because they are insanely and unhumanly good.

 

[Bazza]

Guys as you know this system works by allocating the 'proportion' of a series each team wins. In this case it was 2/3 Pakistan and 1/3 India. This replaced a series where Pakistan won 1-0 so took all of the series, hence India improved.

The alternative way of doing it is to simply give Pakistan more points because they won more games, but this clearly favours teams who play more ODIs and so is not worth considering.

I have stated several times in the past that the reason we use one series against each other team is to avoid a situation where a team plays weaker sides regularly and the best sides not very often and getting an inflated rating as a result. As Neil pointed out, I am not responsible for scheduling fixtures or politicans deciding one team shouldn't play another team.

 

[Pratters]

I have stated several times in the past that the reason we use one series against each other team is to avoid a situation where a team plays weaker sides regularly and the best sides not very often and getting an inflated rating as a result.

So for that compensation you keep a series so many years old? doesnt that make the system poor.

Accept the fact that there is a flaw and work needed to be done for improving it rather than defending the system.

A similar situation arised when the ICC awarded the World Championship of Test Cricket to South Africa for a while. And they went on defending a flawed system.

 

[Pratters]

Looks like a lot of number crunching to me, but the final answer does seem fairly accurate.

No it doesnt.

 

[Pratters]

Erm, the ratings are based on actual results, so he cannot change the figures, and so it looks like results suggest India aren't as good as you think.

In the past 12 months did they not get hammered 5-2 by NZ and also lose to the Windies 4-3? Hardly the stuff of a number 2 side.

Yet when they lose 4-2 to Pakistan, India improves in the ranking.

 

[marc71178]

Yes, because they have improved their performance on the previous effort in the rating series.

Is it really that hard to understand?

 

[FaaipDeOiad]

Yes, because they have improved their performance on the previous effort in the rating series.

Is it really that hard to understand?

It's easy to understand, but that doesn't make it a good way to go about it. As I said earlier, the best way to do it is to work out a predicted result (say 3.5 to 2.5 or something) based on the relative rankings of the sides and adjust their scores based on how they compared to the predicted result. Teams with very high rankings need to win by large margins to maintain their high ranking reflecting their overall dominance, and teams with close rankings to their opponants will go up for a win and down for a loss. Clearly Pakistan deserved to gain points for beating a closely ranked side like India 4-2. The fact that not only they didn't but India in fact moved above them shows that this system if flawed.

 

[Zinzan]

It's not our fault that India and Pakistan refused to play for so long, undermining the system.

I agree that it's deceptive, but is this worse than the ICC system and the way that at an arbitrary cut-off point, the whole thing changes?

Yes, clearly it is worse. At least the cut-off point keeps the latest form of the sides in mind. Sure its not perfect, but anyone with half a brain would acknowledge the ICC ratings of....

1 Australia
2 Sri lanka
3 New Zealand
4 Pakistan
5 South Africa
6 West Indies
7 England
8 India

are a much more accurate indication of oneday form in the last 2 years than......

1 Australia
2 England
3 South Africa
4 New Zealand
5 Sri Lanka
6 India
7 Pakistan
8 West Indies

The ICC ratings would possibly cause the odd argument for or against a side here and there as opposed to the cric-web ratings where apart from Australia at no. 1 it appears the countries have just been listed down randomly ...nothing more.

I'm sure there is a sound mathematical theory behind these ratings, but c'mon fellas, in reality anyone who's followed oneday cricket for the last 2 years can clearly see its a joke.

If the argument in response is that cricket-web-ratings take into consideration a much longer period than 2 years, then I ask, whats the point? Surely any results over 2-3 years old are irrelevant to form.

 

[Pratters]

Yes, because they have improved their performance on the previous effort in the rating series.

Is it really that hard to understand?

umm.. what I said about the ICC trying to show the South Africans were world test champs a few years back.

 

[Zinzan]

Yes, because they have improved their performance on the previous effort in the rating series.

Is it really that hard to understand?

No need to attempt to be condescending marc, everyone understands, they just think its wrong.