Cricinfo Best Test 11 from last 25 years

[TheJediBrah]

Batting Average and RPI are two different measures that have different meanings and that can both be used to evaluate a batsman's performances. I don't think you can say which is the superior one. They're just different. Use them both if that makes you feel better. In some cases both will be relevant and in some cases it won't.

I like the batting average because I feel it is fairer on batsmen than RPI - a batsman with a sequence of low not out scores will see his RPI drop significantly for no fault of his own - he can't control when he bats, and if he didn't get out and scored some runs he arguably did his job. Conversely, if his average goes up he can atleast take some credit for that, in that he did score those runs that are now added to it.

But yea, to be precise use both, and use things like Strike Rate and Relative Strike Rate and % of Teams's Score and anything else really. More stats the better.

RPI could have some relevance if you're measuring the value of a player's performance to the team, but that's an entirely different argument, which relies heavily on other factors like batting position, match situation, rest of the team's performance etc.

 

[Bolo]

This simply means that if all the above games continued, first batter would have got more runs for the team than second one. Therefore the first one is better. That's what the averages show.

To quote what uppercut said once: Scoring runs increases average, getting out reduces average. Staying not out does nothing

For average, yes. But your argument is predicted on the fact that average can:
A: be projected and
B: is the only meaningful metric

The given example is a bit ridiculous, and I don't think it can be extended to assume the first guy would maintain the standards. Take a real ridiculous comparison. Mark Boucher has a test bowling average of 6. He's never failed in bowling. But he has only 1 wicket. Are you comfortable projecting his average from one wicket taken as you are for bat A, who has only gotten out once? Furthermore, don't you feel like Muralitharans 800 wickets are a greater accomplishment even though they have been more expensive, and that his 6 wickets a match are better than mark bouchers 0.01.

I'm not disagreeing with your second paragraph, and I don't think we should discount batting average as a result of not outs in real world examples (if anything the opposite is true as JB is saying), but it is not the only measure that counts- both of these ridiculous examples show this.

 

[TheJediBrah]

For average, yes. But your argument is predicted on the fact that average can:
A: be projected and
B: is the only meaningful metric

The given example is a bit ridiculous, and I don't think it can be extended to assume the first guy would maintain the standards. Take a real ridiculous comparison. Mark Boucher has a test bowling average of 6. He's never failed in bowling. But he has only 1 wicket. Are you comfortable projecting his average from one wicket taken as you are for bat A, who has only gotten out once? Furthermore, don't you feel like Muralitharans 800 wickets are a greater accomplishment even though they have been more expensive, and that his 6 wickets a match are better than mark bouchers 0.01.

I'm not disagreeing with your second paragraph, and I don't think we should discount batting average as a result of not outs in real world examples (if anything the opposite is true as JB is saying), but it is not the only measure that counts- both of these ridiculous examples show this.

Not relevant at all, we know how sample sizes work. Same as the Phil Tufnell example. It's an outlier as a result of a small sample size. It has no bearing on the veracity of using average as a metric in general. You just need to use common sense.

 

[Bolo]

Not relevant at all, we know how sample sizes work. Same as the Phil Tufnell example. It's an outlier as a result of a small sample size. It has no bearing on the veracity of using average as a metric in general. You just need to use common sense.

We are discussing an example of a bat who has only been dismissed once. Im using an example of how small sample size is not appropriate. Either you can extrapolate for both or neither. Pick one (the latter obviously), but apply consistently.

 

[Migara]

Yeah the idea that a given average is somehow worth less if a player has a lot of not outs is nonsensical.

50, 50*, 50, 50* has played just as well as someone who makes 100 + 100

I would say marginally better because the batsman got his eye in every time he got in rather than 100,0,100,0 where batsman was dismissed twice for low scores.

 

[TheJediBrah]

We are discussing an example of a bat who has only been dismissed once. Im using an example of how small sample size is not appropriate. Either you can extrapolate for both or neither. Pick one (the latter obviously), but apply consistently.

Not sure what you're getting at. The hypothetical example used was just used because it was easier to demonstrate than a large sample size. No one was saying that Batsman A is better than Batsman B because of a sample size of 4 innings. It's a hypothetically condensed representation of a larger sample.

I would say marginally better because the batsman got his eye in every time he got in rather than 100,0,100,0 where batsman was dismissed twice for low scores.

Nice to agree with you for a change

 

[Bolo]

Not sure what you're getting at. The hypothetical example used was just used because it was easier to demonstrate than a large sample size. No one was saying that Batsman A is better than Batsman B because of a sample size of 4 innings. It's a hypothetically condensed representation of a larger sample.

I used a ridiculous example to demonstrate why you can't extrapolate on another ridiculous example. In theory the batsman averaging 200 should manage more than 100 RPI given time to complete his innings. Like the Boucher example it should not be assumed he would actually do so given the sample size. Actual achievements count.

 

[TheJediBrah]

I used a ridiculous example to demonstrate why you can't extrapolate on another ridiculous example. In theory the batsman averaging 200 should manage more than 100 RPI given time to complete his innings. Like the Boucher example it should not be assumed he would actually do so given the sample size. Actual achievements count.

Yes, as I said it was a condensed example used for ease of understanding. It wasn't supposed to be taken as a literal translation of a player that has only batted 4 times.

 

[Bolo]

Yes, as I said it was a condensed example used for ease of understanding. It wasn't supposed to be taken as a literal translation of a player that has only batted 4 times.

And I'm demonstrating why it is inappropriate to extrapolate on extreme outliers, which for some reason you seem to have a problem with despite agreeing that it shouldn't be done.

 

[Engle]

Ok, while everyone's dealing with hypothetical’s, let me illustrate with a real life example.

We had a batsman in our club, who made it his MO to remain Not Out.
At the end of the season, he won a prize from the League for having the highest batting average. The League authorities never saw him bat.

But most everyone in our club knew he was a selfish batting bastard. He was not the best batsman in our club, nor did he do much to help the club's batting performance.

 

[cnerd123]

Ok, while everyone's dealing with hypothetical’s, let me illustrate with a real life example.

We had a batsman in our club, who made it his MO to remain Not Out.
At the end of the season, he won a prize from the League for having the highest batting average. The League authorities never saw him bat.

But most everyone in our club knew he was a selfish batting bastard. He was not the best batsman in our club, nor did he do much to help the club's batting performance.

Geoffrey Boycott plays for your club?

 

[TheJediBrah]

And I'm demonstrating why it is inappropriate to extrapolate on extreme outliers, which for some reason you seem to have a problem with despite agreeing that it shouldn't be done.

No one was extrapolating. I specifically said that you can't extrapolate extreme examples.

 

[ankitj]

For average, yes. But your argument is predicted on the fact that average can:
A: be projected and
B: is the only meaningful metric

The given example is a bit ridiculous, and I don't think it can be extended to assume the first guy would maintain the standards. Take a real ridiculous comparison. Mark Boucher has a test bowling average of 6. He's never failed in bowling. But he has only 1 wicket. Are you comfortable projecting his average from one wicket taken as you are for bat A, who has only gotten out once? Furthermore, don't you feel like Muralitharans 800 wickets are a greater accomplishment even though they have been more expensive, and that his 6 wickets a match are better than mark bouchers 0.01.

It was being used as a simplified example. Of course sample size is too small, Mr. Einstein.

 

[Mr Miyagi]

Mark Boucher has a test bowling average of 6. He's never failed in bowling.

1 Hobbes
2 Gavaskar
3 Bradman
4 Lara (it is actually difficult to start omitting Steve Smith already tbh)
5 Sangakarra +
6 Sobers
7 AbDV+
8 Khan
9 Hadlee
10 Marshall Boucher
11 Murali

Sanga and ABdV have a time share ensuring both are fresh.

 

[Bolo]

It was being used as a simplified example. Of course sample size is too small, Mr. Einstein.

It was being used as a simplified example. Until you attempted to extrapolate from it.

This simply means that if all the above games continued, first batter would have got more runs for the team than second one

 

[trundler]

1 Hobbes
2 Gavaskar
3 Bradman
4 Lara (it is actually difficult to start omitting Steve Smith already tbh)
5 Sangakarra +
6 Sobers
7 AbDV+
8 Khan
9 Hadlee
10 Marshall Boucher
11 Murali

Sanga and ABdV have a time share ensuring both are fresh.

Not enough Buttler

 

[TheJediBrah]

It was being used as a simplified example. Until you attempted to extrapolate from it.

omg fine we can spell it out, how about instead of this:

What about;

50, 50*, 50*, 50*: ave 200

or 100, 100, 100, 100 ave : 100?

we go with this as the example:

What about;

50, 50*, 50*, 50*, 50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,50, 50*, 50*, 50*,

: ave 200

or 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100,

ave : 100?

happy?

 

[Bolo]

omg fine we can spell it out, how about instead of this:




we go with this as the example:



happy?

This is not what his original statement means. Read it again.

 

[ankitj]

Dude stop.

 

[Bolo]

Dude stop.

Agreed. This conversation sucks.