Cricinfo Best Test 11 from last 25 years

[Bolo]

I guess Gilchrist was the first one to actually do it well. Flower was a great bat but not really a keeper, Stewart was an ok keeper and solid bat.

Gilchrist though was an excellent keeper and outstanding batsman. Averaged 50+ almost his whole career, 60+ for the first few years too.

For sure. It doesn't make him a pioneer though, it just means he's the most talented keeper bat. Pioneering something is a shift in outlook, not doing it better than others.

 

[Mr Miyagi]

Gilly was a bit Botham like iirc - his first half of his career is far superior with the bat to his second half.

 

[Mr Miyagi]

There are several earlier examples.
For instance, Bob Taylor vs knott, two great keepers.
Or the selection of either Engineer or Kunderan as the keeper for India, based on better current batting form.
Etc.

Long ago, on a tour to India, England played two wicket keepers in the eleven: Knott, who kept and a Roger Tolchard who was supposed to play spin well (Taylor was not even picked as the reserve keeper). I watched one of those games; this Tolchard's idea of playing spin was to run at every ball; needless to say, he was not particularly successful.

Wasn't Engineer just a tremendous cricketer though?

England do have a bit of history of doing it to increase batting strength, though - I agree. Hence the claims start with Ames.

 

[Borges]

Wasn't Engineer just a tremendous cricketer though?

Yes. That's why Kunderan flies under the radar for most people.

 

[TheJediBrah]

For sure. It doesn't make him a pioneer though, it just means he's the most talented keeper bat. Pioneering something is a shift in outlook, not doing it better than others.

I think that's the thing though, because he was so good at it, he made it more conspicuous, more of a necessity. If someone "pioneers" something but isn't that great at it and no one follows suit is that really pioneering? Debatable I guess.

Gilly was a bit Botham like iirc - his first half of his career is far superior with the bat to his second half.

Yeah I think he lost his eye a bit, rather than being "worked out" that seems to be a common opinion.

 

[Mr Miyagi]

I think that's the thing though, because he was so good at it, he made it more conspicuous, more of a necessity. If someone "pioneers" something but isn't that great at it and no one follows suit is that really pioneering? Debatable I guess.



Yeah I think he lost his eye a bit, rather than being "worked out" that seems to be a common opinion.

Yeah - by the time Flintoff did him in 2005 - he had already been on the decline if you look at his annual stats - it is just when it became obvious to everyone he was no longer the batsman he was previously.

I do like Gilly's candour as a commentator about his career though. He is fairly open about these things. Even how he thought Freddie had a plan only to discover he didn't like the deep footholes over the wicket, and then got him out and kept doing it.

 

[TheJediBrah]

Gilly's a great commentator, and a great guy too. So humble and down to earth.

 

[Starfighter]

No it's not. It happened with him because it is a ridiculously small sample size. Give him 100 innings and he wouldn't stay not out in 90% of his innings. His average would normalise around where it would belong, say 4-5, regardless of not outs.

The fact that he wouldn't stay not out in such a high proportion of innings is why his average would reduce towards, say, 5-odd. You say it yourself. It's a fundamental feature of the way averages are calculated.

 

[TheJediBrah]

The fact that he wouldn't stay not out in such a high proportion of innings is why his average would reduce towards, say, 5-odd. You say it yourself. It's a fundamental feature of the way averages are calculated.

Yes exactly. His high average is a result of his small sample size more so than just because of the not outs. The same as if he batted once in his career and made 15. It's an anomaly.

 

[Bolo]

I think that's the thing though, because he was so good at it, he made it more conspicuous, more of a necessity. If someone "pioneers" something but isn't that great at it and no one follows suit is that really pioneering? Debatable I guess.



Yeah I think he lost his eye a bit, rather than being "worked out" that seems to be a common opinion.

The conspicuous argument is what I was arguing a couple of posts ago before I'd considered both flower and Stewart. The idea still holds some weight, but with these two in mind it is more that the conspicuous nature of Gilly served to entrench an existing trend rather than beating a new path.

It's only pioneering if others follow suit. It's possible to pioneer without being great at something, just rare- why follow a failing plan.

Gilchrist may not have had the same standards in the second half of his career, but he was still excellent with the bat considering he was a wk.

 

[Mr Miyagi]

The conspicuous argument is what I was arguing a couple of posts ago before I'd considered both flower and Stewart. The idea still holds some weight, but with these two in mind it is more that the conspicuous nature of Gilly served to entrench an existing trend rather than beating a new path.

It's only pioneering if others follow suit. It's possible to pioneer without being great at something, just rare- why follow a failing plan.

Hey Paddy was great for 1 World Cup.

 

[Starfighter]

Yes exactly. His high average is a result of his small sample size more so than just because of the not outs. The same as if he batted once in his career and made 15. It's an anomaly.

Tufnell is not anomalous but merely the extreme end of a distribution. The effect would be same if he played 10 or 100 innings, as long as the proportion of not outs is maintained. The sample size effect is that the small sample has a higher likelihood of such a skewed distribution. Not outs vary inversely with batting average for the same runs per innings regardless of innings played, just more innings will reduce the likelihood of skewed not out proportions. I think you'd have to go into some rather complex probabilistic calculations with real data to see whether the boosting effect is less or greater than the foregoing effect, things like the probability of a batsmen having reached a certain score going on to make x more runs etc etc. But it's silly to dispute basic maths.

 

[TheJediBrah]

Tufnell is not anomalous but merely the extreme end of a distribution. The effect would be same if he played 10 or 100 innings, as long as the proportion of not outs is maintained. The sample size effect is that the small sample has a higher likelihood of such a skewed distribution. Not outs vary inversely with batting average for the same runs per innings regardless of innings played, just more innings will reduce the likelihood of skewed not out proportions. I think you'd have to go into some rather complex probabilistic calculations with real data to see whether the boosting effect is less or greater than the foregoing effect, things like the probability of a batsmen having reached a certain score going on to make x more runs etc etc. But it's silly to dispute basic maths.

That's the point though. It wouldn't be. The proportion of not outs is anomalous, and a result of the small sample size.

Of course not outs vary inversely with batting average for the same runs per innings regardless of innings played, but that's not really relevant. It's not a comparison between the same number of runs and different numbers of not outs; in this context if you reduced the not outs then the number of runs per innings would have increased.

 

[Starfighter]

That's the point though. It wouldn't be. The proportion of not outs is anomalous, and a result of the small sample size.

Of course not outs vary inversely with batting average for the same runs per innings regardless of innings played, but that's not really relevant. It's not a comparison between the same number of runs and different numbers of not outs; in this context if you reduced the not outs then the number of runs per innings would have increased.

That's funny, I mentioned this on the previous page and you ignored it while claiming that the raising effect isn't real. I asked:

is the effect not-outs raising your average smaller or bigger than the effect of the runs foregone when not out?

Which quite clearly includes both effects. But since you had to claim the the rating effect 'isn't real' I had to show, and you admit, that it in fact is. So we are left with the above question. It really isn't that hard.

 

[trundler]

How to account for not-outs more accurately when assessing batsmen - ESPNcricinfo

 

[Bolo]

Tufnell is not anomalous but merely the extreme end of a distribution. The effect would be same if he played 10 or 100 innings, as long as the proportion of not outs is maintained. The sample size effect is that the small sample has a higher likelihood of such a skewed distribution. Not outs vary inversely with batting average for the same runs per innings regardless of innings played, just more innings will reduce the likelihood of skewed not out proportions. I think you'd have to go into some rather complex probabilistic calculations with real data to see whether the boosting effect is less or greater than the foregoing effect, things like the probability of a batsmen having reached a certain score going on to make x more runs etc etc. But it's silly to dispute basic maths.

Complex to model accurately, as well as virtually impossible given small sample sizes on individual score groupings for individual batsmen.

Easy to work out in general terms for most batsmen in tests. The higher the score you are sitting on, the more likely you are to score more before being dismissed. You are more likely to convert a 50 into a hundred than a zero to a 50.

Not outs actually decrease batting average compared to playing a full innings in most instances.

Not really worth accounting for though. Rpi count for more than this disparity, and there's a reason average has survived so long as a metric, and it is because it is an extremely good one.

 

[Migara]

Anyway, still haven't figured out the 2 fast bowlers to go with Hadlee and Tayfield in the Surrounded by Imbeciles XI.

Javagal Srinath would be one

 

[Migara]

There's that same fallacy again. Not outs don't help your average. It doesn't help to have your innings cut short when you're in and seeing the ball well and having to start your next innings from scratch. I don't know why people think this.

Depends on whether you are 4* or 104*
or

 

[Migara]

1 Atherton
2 Warner
3 Bancroft
4 Dravid/S Smith
5 Faf
6 Chandimal +
7 Afridi
8 Imran
9 Akram
10 Younis
11 Pringle

12 Trescothick

Manoj Prabhakar?

 

[Bolo]

Depends on whether you are 4* or 104*
or

It hold true whether you are on 104 or 4.

HowSTAT! Test Cricket - Century Averages