Leaving out the minnows...

[zaremba]

Your theory... smacks of a contentedness for simplicity.

Rather the opposite. I'm saying that we should analyse every facet of a player's record rather than seeking some holy grail formula for producing a number which represents a player's true worth.

surely the deeper the analysis done, the better, regardless of whether it's an exact science or not.

Sorry if I didn't make myself clear, but this is exactly what I was saying.

 

[marc71178]

well... if it happens bangladesh's performance in the next few years would be watched with a bit more respect. it wont change the past, obviously.

It ought to change how people view the past, but there's so many people who are only interested in the numbers and stats rather than the actual game that they'll just continue to ignore what is becoming more and more obvious.

 

[Uppercut]

Reading through all this almost makes me want Bangladesh to beat England in the series next year, might finally get people to recognise that they have improved and are far more competitive now than even 12 months ago.

Hmm don't think so. If their abject humiliation at home to Sri Lanka stood out in anyone's memory as significantly as the game the three-wicket loss to New Zealand no one would be saying anything of the sort.

 

[Uppercut]

Now before Upper's jumps in his "that's all only hypothetical' argument', let me say, yes it is hypothetical, but still should be considered & discussed when making such a comparison. Just like it's only hypothetical (for rugby fans) to suggest Brian O'Driscoll would have probably scored more test tries from the same amount of games had he played for the All Blacks....doubt many rugby fans wouldn't argue too much with that theory hypothetical or not.

Yeah, but no one ever says that O'Driscoll would have been a better player had he played for the All Blacks. There's a good reason for this- he's a ****ing awesome player already. You have to have failed at something before people start making excuses for you.

The point isn't that Vettori wouldn't have been a better spinner had he been Indian, although we can't know for sure, the point is just that everyone has to face all kinds of difficulties in life. O'Driscoll, to use your example, played for a less-than-world-class team and had to deal with some serious injuries. He forged a successful career anyway. Now, I wouldn't call Vettori's career unsuccessful. But it's all relative. If he played the hand he was dealt and came out with plenty of wickets, no one would need to say these kinds of things.

Or to put it quite crudely, it's a bit like me saying "I would be a top-class cricketer, if I was much, much better at cricket". It's true, but it's irrelevant and IMO quite facetious.

 

[Athlai]

Yeah, but no one ever says that O'Driscoll would have been a better player had he played for the All Blacks. There's a good reason for this- he's a ****ing awesome player already. You have to have failed at something before people start making excuses for you.

The point isn't that Vettori wouldn't have been a better spinner had he been Indian, although we can't know for sure, the point is just that everyone has to face all kinds of difficulties in life. O'Driscoll, to use your example, played for a less-than-world-class team and had to deal with some serious injuries. He forged a successful career anyway. Now, I wouldn't call Vettori's career unsuccessful. But it's all relative. If he played the hand he was dealt and came out with plenty of wickets, no one would need to say these kinds of things.

Or to put it quite crudely, it's a bit like me saying "I would be a top-class cricketer, if I was much, much better at cricket". It's true, but it's irrelevant and IMO quite facetious.

Think he was implying that Vettori's figures would possibly improve not his ability.

 

[Uppercut]

Yeah, that he would have taken more wickets and been more successful?

It's certainly a possibility- but much more important is how successful he's been in real life.

 

[bagapath]

It's certainly a possibility- but much more important is how successful he's been in real life.

yeah. i hate ifs and buts. fantasies have no place in real analysis.

 

[Richard]

How about the Ashes series between 1977 and 1985 - should they be expunged from the record? The 78/79 sides, Australia in particular, were lamentably weak due to WSC

I'd be fully in favour of that (not Ashes, just all Australian series' between 1977/78 and the opening bit of 1979/80 and the one-and-a-half West Indian ones in the same timeframe) if it weren't for the fact that official Test status is vital to emphasise that Packer's games were unsanctioned and rebellious. Removing Test status from them might go a little way to legitimising WSC, which is a massive no-no.

Anyone attempting to use those matches to prove anything about any team involved is an ignoramus, IMO. One country's first XI vs another's second is of complete irrelevance.

I'm not sure what the 1977, 1981, 1982/83 and 1985 Ashes series have abnormal with them BTW?

 

[Richard]

There's no statistically defensible reason to exclude a player's record against 'minnow' teams. You might weight them differently if you wish but they're not outliers.

As you pointed-out a while back, there's no statistically defensible reason to even use stats in a cricket context because cricket is totally unscientific. Anything you do with cricket stats is going to be challengeable by a good statistician.

 

[Richard]

If you start taking out stats, you almost have to pick and choose matches, and that's too complicated.

For example, why would you remove this test from Vettori's figures? What would have been the impact on NZ cricket if he hadn't performed on this occasion? This was a pressure situation and when his team needed it, he delivered big-time. Likewise, why would you remove these centuries from Gilchrist and Ponting? Top, top knocks, the pair of them. Can you imagine if Australia had lost to Bangladesh? And yet these two world-class players saved the team a world of pain.

Yes, overall runs and wickets against Bangladesh come easier than against other teams. But it's not so cut and dry and just removing every performance. They just need to be taken into consideration.

The same can be said of any team. There should never be any picking-and-choosing; it's all or nothing. You cannot say "they performed like a Test team in that match so that should be a Test", because you can say that about a vast number of First-Class games.

Something often forgotten by a great many people at the current time is that Test cricket is not the only level of the game that exists. You should give some consideration to all cricket a player plays when assessing him, and just because something was a Test doesn't make it automatically more worthy than something that wasn't.

There have been some performances by some players in "Test"s against Bangladesh which are certainly worthy of huge praise; ditto there are some knocks in matches not recognised as Tests, rightly so, that are worthy of more praise than some knocks in matches which are quite rightly recognised as Tests.

The issue of giving credit to a player for a performance and the issue of whether a team deserves Test status is not the same thing. Too many people take the line of "it's a Test so it has more worth than something that's not a Test, by default".

Getting the best and most apt Test figures to apply is good; using Test figures as the be-all-and-end all is bad. The latter, thus, should not preclude the former.

 

[Richard]

Therefore, the answer to the question "should we exclude games v Bangladesh" is both yes and no. "Yes" because they have consistently been a sub-standard Test team and there are relatively cheap runs and wickets to be had against them. This can lead to an average player's Test figures being seriously flattered. "No" because Tests v Bangladesh are still Tests, and a (minor) facet of your statistical record is how good or bad you are at scoring easy runs and taking cheap wickets: in other words, how good a "minnow-basher" you are.

They're only Tests because ICC say they are. There are all sorts of ways to see whether somone is good at taking easy runs\wickets, and you can do that without looking at their record against Bangladesh.

As I say, the trouble with this comes from people who try to "perfect" Test stats, to make them what they want them to be - the be-all-and-end-all of judging a player. They aren't and cannot be.

 

[Uppercut]

As you pointed-out a while back, there's no statistically defensible reason to even use stats in a cricket context because cricket is totally unscientific. Anything you do with cricket stats is going to be challengeable by a good statistician.

Almost anything you do with any stats anywhere is going to be challenged tbf. A lot of the time you either leave a lot of variables uncontrolled or you create a scenario with no resemblance to real life whatsoever.

 

[Migara]

A Definition of minnows

 

[Zinzan]

Yeah, but no one ever says that O'Driscoll would have been a better player had he played for the All Blacks. There's a good reason for this- he's a ****ing awesome player already. You have to have failed at something before people start making excuses for you.

The point isn't that Vettori wouldn't have been a better spinner had he been Indian, although we can't know for sure, the point is just that everyone has to face all kinds of difficulties in life. O'Driscoll, to use your example, played for a less-than-world-class team and had to deal with some serious injuries. He forged a successful career anyway. Now, I wouldn't call Vettori's career unsuccessful. But it's all relative. If he played the hand he was dealt and came out with plenty of wickets, no one would need to say these kinds of things.

Or to put it quite crudely, it's a bit like me saying "I would be a top-class cricketer, if I was much, much better at cricket". It's true, but it's irrelevant and IMO quite facetious.

Can't believe you've missed my point on the O'Driscoll analogy, because it was a fairly simple one......
If rugby critic A (who only cares about stats, specifically how many tires backs score) said ..."bah bah bah, O'Driscoll's not that great a center, such & such from New Zealand is better because he's scored a higher number of tries per test". rugby critic B, could then make the reasonable argument that "this may be the case, however if O'Driscoll had played for the All Black's he's would have scored more tries a Test & hence had a better record - both in terms of tries scored & tests won.

The point is, most of us recognize that O'Driscoll is a very fine centre, despite the fact he doesn't play for a team conducive to him scoring a great no of tries or a team who have a remarkable winning record. ,, and accept that had he played for the All Blacks, he'd almost certainly have scored more tries & been involved in more wins.

Whilst it's not exactly apples v apples (rugby v cricket), when you look at Vettori's career record, you're not prepared to make any consideration for his bowling average. I still maintain he'd have a better test average (prob around 30) albeit, possibly not as many wickets, if he'd played the same no of tests for a team like India, one because their conditions are more conducive to spin & also because they are a stronger test team.

 

[Zinzan]

Yeah, that he would have taken more wickets and been more successful?

It's certainly a possibility- but much more important is how successful he's been in real life.

OK that's fine, then from a statistical standpoint, O'Driscoll's a rubbish centre by All Black standards judging by his ratio of tries per test & winning record, which is inferior to almost all recent All Black centres. This is based on real life with no context or perspective provided of course, because it's all about real life right?

 

[aussie]

It wasnt supposed to be a comparison between the two teams during the same period of time. My point was that the WI bowling attack from 2002-07 was probably as bad as the present day Bangladesh bowling attack (you could say last 1-2 years).

I fail to see how the likes of Mashrafe Mortaza, Enamul Haque Jr, Shakib Al Hasan etc are that much worse than Fidel Edwards, Adam Sanford, Tino Best, Vasbert Drakes, Darren Powell, or Dwayne Bravo. Not one of those WI bowlers were averaging anything lower than the high 30s with the exception of Collymore and to an extent Pedro Collins.

Not that I don't think performances against Bangladesh should be excluded, but if that is the case then excluding performances against present day Bangladesh is equivalent to excluding performances against the 2002-07 WI side.

Not really.

For any opposition bowling attack wickets againts the West Indies batting line-up given it was still solid during 2002-07, wouldn't be comparable to wickets againts any BANG batting line-up.

Plus even though the WI bowling attack was very poor between 2002-07. They where stll wayyy ahead of BANG imo.

- They managed to beat IND 2002.

- Lawson ran through AUS in 03

- Edwards troubled SRI & ENG in 03/04 respectively

- Beat SA in a test late 07

- Powell /Lawson troubled SRI when a second string team toured in 05/06

- Taylor @ Kingston 06

No BANG attack never came close to challenging a test batting line-up this much during that period expect of PAK 03 & vs AUS 06.

 

[Uppercut]

OK that's fine, then from a statistical standpoint, O'Driscoll's a rubbish centre by All Black standards judging by his ratio of tries per test & winning record, which is inferior to almost all recent All Black centres. This is based on real life with no context or perspective provided of course, because it's all about real life right?

Ack. They're incomparable. You can be an all-time great rugby player without ever scoring a try. You can't be an all-time great batsman without ever scoring a run. Cricket's unique in that a player's contribution can be quantified. Not completely, of course, but to some extent.

 

[NUFAN]

Ack. They're incomparable. You can be an all-time great rugby player without ever scoring a try. You can't be an all-time great batsman without ever scoring a run. Cricket's unique in that a player's contribution can be quantified. Not completely, of course, but to some extent.

Gotta say I'm loving this thread, good answer.

 

[Richard]

Almost anything you do with any stats anywhere is going to be challenged tbf. A lot of the time you either leave a lot of variables uncontrolled or you create a scenario with no resemblance to real life whatsoever.

Yup, that's basically what I was saying. No set of statistics will ever be unchallengeable, and in cricket statistical work is less exact than in so many places, for the reasons mentioned by McNamara (2009): "No two batsmen have ever walked out to bat under the exact same circumstances. Ever."

So it's all about getting the most useful stats you can. For me, Test-standard side is more worth being a separate classification than Test-playing-as-defined-by-ICC. I don't see that that latter category has any real merit at all for being something on which to assess players - apart obviously from the fact that by-and-large the latter is the former.

 

[Top_Cat]

As you pointed-out a while back, there's no statistically defensible reason to even use stats in a cricket context because cricket is totally unscientific. Anything you do with cricket stats is going to be challengeable by a good statistician.

Nah mate, that's not what I meant. Data relating to cricket is not a scientific dataset because it's not collected in a scientific way. Many studies are completed with non-scientific data because, as with cricket, it's not always possible to collect it scientificially. So you can use cricket/sports raw data but with appropriate caveats/context/limitations.