cricrate: new cricket ratings website

[Prince EWS]

Good point, but that effect should be relatively minor. The Bangladesh bowlers (aside from Shakib) would be rated pretty low to start with that 1-2 tests of good performances (against inferior opposition, so those performances would be discounted as well) won't make much of a difference going into a third test. Interesting to think about though.

I tended to find this had a much bigger affect across the board than you'd guess. It works the other way too -- my system kept under-rating McGrath's and Warne's opposition (and then by extension, the quality of McGrath's and Warne's performances) because Australia kept hammering whoever they played and nerfing their team stats for that period. It's hard to find a balance; on one hand you don't want to cast too wide a time net over the performance of a team or you'll be evaluating a different side altogether with mass change in personnel, but on the other hand you don't want to cast too narrow a net or you get the effect I just spoke about. It's why BowlerA+BowlerB+BowlerC+BowlerD is so much more accurate a measure of the standard of an opposition attack when determining how good a batting innings was than BowlingAttackX, when done properly. It's difficult to do that properly though because some bowlers play so few matches in their careers.

 

[Cabinet96]

It's a bit weird about AbdV. Great batsman, great to watch, but not as memorable as many other greats, or even Amla for that matter.

I went through a period of thinking this, but I think really it's just as a result of South Africa not really having any major rivals and not playing long series. I mean, I thought his efforts at Centurion last year, when pretty much every English and South African batsman had looked like Johnson would get them out every over and he batted serenely until getting out with the tail both innings, was pretty memorable. His biggest issue IMO is that he really struggled in the crux of a few big series, such as the India one in 10/11 and the tours to England and Australia in 2012, which created the downhill skier tag, but he's had enough solo knocks to have destroyed that by now. People remember those series much more than a one off series in Sri Lanka, the UAE, NZ or whatever, and the same goes for the at times average sides they've played at home in recent years.

He just doesn't get the same amount of opportunities as many others. He debuted over 10 years ago and has played every SA test since. Yet he's only played two more tests in England and Australia than Virat Kohli, who started three and a half years ago for example. Michael Clarke debuted around the same time and has played 30 tests against England and 22 v India, compared to 16 and 13 respectively for AB, etc. So of course he's likely to have more iconic moments we remember.

 

[Spark]

Personally I never really got on board with the downhill skier tag per se but I definitely thought his stats flattered him in 2011-2012, especially because I was noticing he was coming in at 3/lots and absolutely murdering attacks and ending up 70* off 50 or something. So it wasn't so much a case that he wasn't necessarily any better or worse under pressure than comparable batsmen, just that he was putting up incredible-looking numbers without really putting up the sorts of innings and distinct performances that Amla, Smith, Kallis etc were putting up. The last series down in Aus epitomised this - he did not a whole lot right until the very end where he carted around an attack that was frankly already consigned to an absolute flogging, and as a result of that innings his series figures looked fine whereas up til then they looked fairly ****. Ironically, the 30 he scored in Adelaide was a far, far more significant innings in terms of winning the team the series.

He's rectified all this to some extent in 2013-4, particularly with the way he batted against us and his innings in the Steyn-bopping-Shami-for-six chase (and to be honest I always thought he would once he got the opportunities), but I definitely felt that he was a rung below Clarke, Amla, Sanga etc etc for that reason, because they were playing seriously memorable, series-defining innings. The Amla comparison is particularly telling because to an extent it nullifies the "he doesn't play enough" point - Amla has played just as many Tests in that same period and pulled off some ****ing amazing knocks and series-winning/defining performances along the way (vs us in 2011, vs England the next year in particular). And I know this is getting off-track, but this is why I was so surprised that so many people would take ABdV over Amla in that other thread; it's very clearly the other way around for me.

TL;DR be careful when you rate batsmen based on raw stats, because samplesizelol and quality > quantity.

 

[fredfertang]

What's the point when we already have DoG's ranking.

I'm all for as many people as possible doing their own rankings on the infinite number of monkeys with keyboards principle that eventually someone will come up with a list I approve of

 

[viriya]

I tended to find this had a much bigger affect across the board than you'd guess. It works the other way too -- my system kept under-rating McGrath's and Warne's opposition (and then by extension, the quality of McGrath's and Warne's performances) because Australia kept hammering whoever they played and nerfing their team stats for that period. It's hard to find a balance; on one hand you don't want to cast too wide a time net over the performance of a team or you'll be evaluating a different side altogether with mass change in personnel, but on the other hand you don't want to cast too narrow a net or you get the effect I just spoke about. It's why BowlerA+BowlerB+BowlerC+BowlerD is so much more accurate a measure of the standard of an opposition attack when determining how good a batting innings was than BowlingAttackX, when done properly. It's difficult to do that properly though because some bowlers play so few matches in their careers.

I'm not sure you follow.. every player's current ratings involves his whole career, not just a limited recent time period (which would make the issue you point out much more significant). In my setup for McGrath/Warne's case, their 5th Ashes test opposition stats might have dropped because of a 5-0 whitewash, but not as significantly as you might think since Graham Thorpe's current rating would stay decent since his overall career is still decent. If Thorpe was in the top 10 at the start of the series he might've dropped to the top 20 at the end after a poor series say.

On evaluating bowling attacks, mine is based on each bowler's current rating weighted to how much the bowler actually bowled in the innings. The best attacks would have 3-4 very good bowlers.

 

[weldone]

I tended to find this had a much bigger affect across the board than you'd guess. It works the other way too -- my system kept under-rating McGrath's and Warne's opposition (and then by extension, the quality of McGrath's and Warne's performances) because Australia kept hammering whoever they played and nerfing their team stats for that period. It's hard to find a balance; on one hand you don't want to cast too wide a time net over the performance of a team or you'll be evaluating a different side altogether with mass change in personnel, but on the other hand you don't want to cast too narrow a net or you get the effect I just spoke about. It's why BowlerA+BowlerB+BowlerC+BowlerD is so much more accurate a measure of the standard of an opposition attack when determining how good a batting innings was than BowlingAttackX, when done properly. It's difficult to do that properly though because some bowlers play so few matches in their careers.

Maybe that somewhat offsets the team support part? (e.g. McGrath and Warne were helped due to having excellent catchers & ground-fielders, and excellent support bowlers who could keep the pressure on)

 

[viriya]

Maybe that somewhat offsets the team support part? (e.g. McGrath and Warne were helped due to having excellent catchers & ground-fielders, and excellent support bowlers who could keep the pressure on)

Another factor I've been thinking of implementing is "scoreboard pressure".. it really has an impact when a batting team has to bat after chasing leather for 2 days.

 

[Prince EWS]

I'm not sure you follow.. every player's current ratings involves his whole career, not just a limited recent time period (which would make the issue you point out much more significant). In my setup for McGrath/Warne's case, their 5th Ashes test opposition stats might have dropped because of a 5-0 whitewash, but not as significantly as you might think since Graham Thorpe's current rating would stay decent since his overall career is still decent. If Thorpe was in the top 10 at the start of the series he might've dropped to the top 20 at the end after a poor series say.

On evaluating bowling attacks, mine is based on each bowler's current rating weighted to how much the bowler actually bowled in the innings. The best attacks would have 3-4 very good bowlers.

Yeah, the effect should be negligible if you do it like that. That's good.

What I was saying is that.. lets say McGrath played a series against England in England in 2001 and Australia hammered them. If you determine the quality of his opposition by looking at how England did overall in 2000-2002 period with more emphasis put on their home games, then fact that McGrath had a good team around him would count against him as those overall England stats would be disproportionately affected by Australia's dominance. If Australia consistently dominated batting lineups then this would affect his end rating quite a bit, as the system would consistently think the teams he was playing against were worse than they were, simply because their stats in periods he played them would disproportionately include matches against McGrath's team.

The way you've done it should make that effect negligible though, and that's good.

As an aside my system ranks Masakadza at #117 based on "current form" / "current ability" / whatever you want to call current rankings. It takes domestic cricket, tour games, A-team games etc into account for that too though which probably puts him more in his rightful place. There are a bunch of players ahead of him who haven't played Tests at all.

 

[weldone]

What I was saying is that.. lets say McGrath played a series against England in England in 2001 and Australia hammered them. If you determine the quality of his opposition by looking at how England did overall in 2000-2002 period with more emphasis put on their home games....

should do 1999-2003 anyways imo

As an aside my system ranks Masakadza at #117 based on "current form" / "current ability" / whatever you want to call current rankings. It takes domestic cricket, tour games, A-team games etc into account for that too though which probably puts him more in his rightful place. There are a bunch of players ahead of him who haven't played Tests at all.

Just out of interest - where are Rohit Sharma, Badrinath and Manoj Tiwary now?

 

[Prince EWS]

should do 1999-2003 anyways imo

Yes indeed. That's why I said that if you do it like that you have to find a balance. Make it cover too wide a period and you risk including irrelevant games after mass personnel changes; make it cover too narrow a period and you give too much weighting to the series in question which cannibalises the process. Viriya has rated the standard of the opposition on a player-by-player basis though which, aside from being harder, I do think is probably better on balance (it certainly makes the issue we're discussing negligible). It does have its own issues when it comes to opposition players who had really short careers though, and I suppose with good players at the start of their careers too.

Just out of interest - where are Rohit Sharma, Badrinath and Manoj Tiwary now?

I don't want to reveal too much because the secrecy of the actual ratings is vital for a few things outside of CW, but they're all close together in the 55-65 range; Rohit is still marginally rated the best of the lot.

 

[Cabinet96]

I don't want to reveal too much because the secrecy of the actual ratings is vital for a few things outside of CW, but they're all close together in the 55-65 range; Rohit is still marginally rated the best of the lot.

This is fine as long as it's not only for a hypothetical CPL.

 

[weldone]

...they're all close together in the 55-65 range; Rohit is still marginally rated the best of the lot.

It sounds right.

 

[smash84]

Shakib Al Hasan greater than Imran Khan and Keith Miller.

I think you might be slightly over rating Shakib with your criteria

 

[viriya]

Shakib Al Hasan greater than Imran Khan and Keith Miller.

I think you might be slightly over rating Shakib with your criteria

The case of Imran being rated relatively low was gone through earlier in this thread. It's a function of all-round match ratings being a multiplication of batting and bowling performances in the match, and Imran being more bowling focused early in his career and more batting focused later on. Yes, he had a few series where he dominated with both, but for most of his career he did very well only in one discipline. That said, the career ratings gives a bonus for his great batting and bowling averages.

While Shakib might be slightly overrated, I do think he deserves to be in the top 10. After just 37 tests he has already made 100 runs + 5 wkts in a match (cumulative in both innings) 5 times. The most it's ever been done is 7 times by Sobers and Botham which he is almost certainly going go past easily.

 

[weldone]

While Shakib might be slightly overrated, I do think he deserves to be in the top 10. After just 37 tests he has already made 100 runs + 5 wkts in a match (cumulative in both innings) 5 times. The most it's ever been done is 7 times by Sobers and Botham which he is almost certainly going go past easily.

The sitting ducks and cannon fodders of Zimbabwe will agree.

 

[smash84]

The case of Imran being rated relatively low was gone through earlier in this thread. It's a function of all-round match ratings being a multiplication of batting and bowling performances in the match, and Imran being more bowling focused early in his career and more batting focused later on. Yes, he had a few series where he dominated with both, but for most of his career he did very well only in one discipline. That said, the career ratings gives a bonus for his great batting and bowling averages.

While Shakib might be slightly overrated, I do think he deserves to be in the top 10. After just 37 tests he has already made 100 runs + 5 wkts in a match (cumulative in both innings) 5 times. The most it's ever been done is 7 times by Sobers and Botham which he is almost certainly going go past easily.

Yeah, I know that was dealt with earlier in the thread but its very difficult to take accept the ratings when Imran's bowling was >> than Shakib's batting and bowling combined. I haven't seen Keith Miller play but I have definitely seen Shakib play and while he is good he doesn't merit a place in the top 10.

 

[viriya]

The sitting ducks and cannon fodders of Zimbabwe will agree.

Unfair to say it's cos of Zim. 2 vs WI, 1 vs Pak, 1 vs SL.. only 1 vs Zim and that was the recent series.

 

[viriya]

Yeah, I know that was dealt with earlier in the thread but its very difficult to take accept the ratings when Imran's bowling was >> than Shakib's batting and bowling combined. I haven't seen Keith Miller play but I have definitely seen Shakib play and while he is good he doesn't merit a place in the top 10.

It's really about how you define an all-rounder.. If it's about performing with both bat and ball in a match, then I don't think you can say Shakib hasn't been a better all-rounder than Imran in the 37 matches he's played so far. Imran of course is an ATG bowler, and imo has the best overall all-round record, without actually performing as much as an all-rounder through his career. So depending on how you define the concept of all-rounder, Imran is either overrated or underrated.

 

[Riggins]

I don't want to reveal too much because the secrecy of the actual ratings is vital for a few things outside of CW, but they're all close together in the 55-65 range; Rohit is still marginally rated the best of the lot.

This is fine as long as it's not only for a hypothetical CPL.

Nah look, I'm gonna need a proper explanation of this.

 

[Dan]

Yeah, the effect should be negligible if you do it like that. That's good.

What I was saying is that.. lets say McGrath played a series against England in England in 2001 and Australia hammered them. If you determine the quality of his opposition by looking at how England did overall in 2000-2002 period with more emphasis put on their home games, then fact that McGrath had a good team around him would count against him as those overall England stats would be disproportionately affected by Australia's dominance. If Australia consistently dominated batting lineups then this would affect his end rating quite a bit, as the system would consistently think the teams he was playing against were worse than they were, simply because their stats in periods he played them would disproportionately include matches against McGrath's team.

The way you've done it should make that effect negligible though, and that's good.

As an aside my system ranks Masakadza at #117 based on "current form" / "current ability" / whatever you want to call current rankings. It takes domestic cricket, tour games, A-team games etc into account for that too though which probably puts him more in his rightful place. There are a bunch of players ahead of him who haven't played Tests at all.

Yeah, the old cricketing circular logic argument -- do well enough to make the opposition look dog****, and you devalue your own performances.