Removing statistics against Bangladesh and Zimbawe

[Prince EWS]

Wouldn't use global averages as my weights, personally. Just manipulating the same poor measure to explain other poor measures.

Any better ideas for weights?

Not being a smartarse by the way; it's a genuine question.
It is a bit of a circular problem though because when you look at a team's bowling average over any period, they've played different amounts of games against certain teams than others which means you should really weight all that too. So basically, in order to figure out how good all the teams are, you need to first find out how good all the teams are.

 

[Top_Cat]

cbf. It really would take a massive amount of time and effort to generate a decent design weight for countries, especially if you want to take into account runs scored against teams as their standard changes. Nightmare.

 

[Prince EWS]

cbf. It really would take a massive amount of time and effort to generate a decent design weight for countries, especially if you want to take into account runs scored against teams as their standard changes. Nightmare.

Exactly. Which is why I just use team average over specific time periods.

 

[ankitj]

Bang defeated NZ 4-0 in the recent ODI series. Does that change anything?

 

[Prince EWS]

Bang defeated NZ 4-0 in the recent ODI series. Does that change anything?

Well, not when we're talking about Tests.

 

[Howe_zat]

I've already weighed in on this in a different thread, so I'll just quote myself. Essentially, my point is that obviously when you rate how valuable a performance is, you have to take everything into context, including the opposition.

However, stats are a way of measuring success, not performance, and so these kind of modifications or qualifiers are often brought in to make the statistic - a measure of success - closer to an objective measure of performance.

Removing Bangladesh and Zimbabwe from general statistics in order to do this, I think, is bollocks.

For context, I was debating against someone who believed that since a performance against Bangladesh isn't going to affect the outcome of the series, it should be discarded.

Well that's a very interesting concept, and most of the overall weakness in cricketing statistics that they often don't reflect the match situation.

However, getting a better reflection on how "important" your match performance can work in the opposite direction and removing weaker sides from the equation isn't the answer.

Firstly, there is the issue of being able to capitalize in the right situations as oppose to scoring when it is easy, which we count and which we don't.

Which innings was easier, for example - Jonathan Trott's hundred against Bangladesh, scoring against a poor attack but early in the game on an early-season English wicket while under personal pressure to deliver - or Matt Prior's hundred at Sydney, coming in with the aim to stamp a half-beaten and demoralised attack on a flat deck.

It's not as simple as it sounds, so why do we pick so easily?

Then there's the matter of what those playing for a weaker team might achieve. Take Shakib Al Hasan for example. In your post you used the slightly unfair qualifier "test-standard" to distiguish against his performances in ODIs against Zimbabwe or Tests against a depleted West Indies unit in 2009.

People who remove Bangladesh from averages might have removed that side too. But for Shakib, he was going on tour with the first reasonable chance he has ever had of winning. The runs and wickets he got there will mean a tremendous amount to him and indeed he had a huge effect on that series. So by your own standards those are the runs that count the most.

Conversely, suggesting that performances that don't affect the match situation don't count could suggest you'd remove Shakib's excellent bowling effort in South Africa. He never helped his side win, they never even came close. But you would have to admit it's unfair to take that series away from him personally.

It's not just Bangladesh of course - would you remove the stats of, say, Vaughan in Australia, just because his runs didn't effect the outcome of the series?

These are obviously delicate and trying to adjust the stats to reflect the competitiveness is clearly a lot more complicated than just removing Bangladesh from the equation. So we have no choice but to keep them in.

 

[silentstriker]

Yeah, I've posted my view on this before. Runs are only worth anything if they might help change the result of a cricket match in your favour, and since Bangladesh have so far lost every game they've played against a test-standard opposition, it's fair to say that the chance of a batsman's runs changing the result is small enough that it's better to discount it.

India didn't win a Test match for fifteen years overseas - from 1986 onwards. Should we take them (their overseas matches) out when comparing players?

 

[GIMH]

I think we should take India out of cricket altogether tbh

 

[silentstriker]

I think we should take India out of cricket altogether tbh

I like it. When the BCCI are gone, the barmy army will be the only annoying thing left in the game.

 

[ankitj]

Well, not when we're talking about Tests.

Haha. I meant ODIs alone. You mentioned a few posts back that defeating AUS and Ind, should be treated as outlier performances.

 

[silentstriker]

Well, that and Ricky Ponting, but he's on his way out anyway.

 

[GIMH]

I like it. When the BCCI are gone, the barmy army will be the only annoying thing left in the game.

Yep, I can see why a Test Cricket elitist would hate the group that provide larger attendances than anyone else worldwide

 

[Prince EWS]

Haha. I meant ODIs alone. You mentioned a few posts back that defeating AUS and Ind, should be treated as outlier performances.

Nah, I meant that Bangladesh themselves were outliers; not that those were outlier performances.

It doesn't really detract from my argument because, unlike Uppercut, my reasoning wasn't based around the fact that they never ever ever win games (and he was just referring to Tests anyway).

 

[silentstriker]

Yep, I can see why a Test Cricket elitist would hate the group that provide larger attendances than anyone else worldwide

They're more fun when they lose.

 

[Faisal1985]

Think this thread needs to be made as I would like to debate with those people who think that it is perfectly reasonable completely discount a players achievements simply because it was against a lesser team. Seriously dont understand it, it really is just a way of manipulating stats to help your argument isn't it? Because people use this argument the other way around and put down a player who does not score well against BD/Zim. Doesn't seem quite fair to me :/

awta...

a bell curve always has 2 tails. You can't just take one tail out of the question for analysis.

 

[vcs]

India didn't win a Test match for fifteen years overseas - from 1986 onwards. Should we take them (their overseas matches) out when comparing players?

Pretty sure they won in Sri Lanka in the '90s, but **** me, that still is a mind-boggling fact. Was the win in Port of Spain (I think) in 2002 our first win outside the subcontinent in 16 years?

 

[vcs]

Any better ideas for weights?

Not being a smartarse by the way; it's a genuine question.
It is a bit of a circular problem though because when you look at a team's bowling average over any period, they've played different amounts of games against certain teams than others which means you should really weight all that too. So basically, in order to figure out how good all the teams are, you need to first find out how good all the teams are.

Yeah, it seems like a chicken and egg scenario..

When you are using a team's bowling average over a period to weight a performance against them, do you make any special consideration for the actual bowlers who were bowling in that game or does the performance receive the same weight even if the main bowlers were injured/absent? India's tour of Australia in 2003-04 comes to mind here.

 

[bagapath]

Pretty sure they won in Sri Lanka in the '90s, but **** me, that still is a mind-boggling fact. Was the win in Port of Spain (I think) in 2002 our first win outside the subcontinent in 16 years?

there was one in zimbabwe in 2001.

u still cant take out india completely in the 16 year period; they beat australia 2-1, twice. and southafrica got the same treatment in '96. england was given a 3-0 whitewash. all these happened at home. but it is million times better than typical minnow behavior, which is basically getting walloped everywhere. spineless outside the country, yes. but, minnows? NOOO.

If you include the minnows in the current scenario, asif and irfan pathan become comparable bowlers. thats just not right.

 

[smash84]

this is a dream come true. always wanted to stay on ikki's side and fight the bad guys as a team; like richard burton and clint eastwood coming together in "where eagles dare" or t.cruise and v.rhames in the MI series. and here is the moment.... i totally completely agree with your words here, mate. now, i hope my support doesn't make you change your mind.

A sigh of relief from Bagapath. You should have nothing to fear though. I think you, Ikki, Migara, and (PEWS coming into this category of late) debate at a very good level. It is like the great all rounders from the 1980s.

Pretty sure they won in Sri Lanka in the '90s, but **** me, that still is a mind-boggling fact. Was the win in Port of Spain (I think) in 2002 our first win outside the subcontinent in 16 years?

WTF......is that true. India did not win a test abroad for 15 years/????

 

[Furball]

I've generally stopped doing this - at the end of the day, surely scoring piles of runs/taking stacks of wickets against weaker opposition is what makes strong teams strong and weak teams weak?

I do think there's merit in putting asterisks against player's names who have significantly bolstered their stats with good performances against Bangladesh and Zimbabwe - such as the Irfan Pathan example mentioned earlier - or removing both when comparing two players when one has played the two far more than the other (for example, comparing Warne and Murali) - but if you're looking at how player X has done in the last 3 years, then any runs he's scored against Bangladesh count as part of that. And for someone like AB de Villiers, should we really remove Bangladesh from the analysis when doing so actually improves his record?