The Ashes 2010/11
- Thread starter strung out
- Start date
DrRingDing
'anti-human wanker'
Like shit off a shovel. I wouldn't want to face it.
littlebabyjesus
one of Maxwell's demons
Yeah, I have to say that I'd have loved to have seen Tait tearing in in the Ashes.
mattie
missing in inaction
A footballing analogy - Arsene Wenger insists his scouts find out about a footballer's home life and personality before he thinks of signing them. If he thinks he could bring them to Arsenal and improve them, he will. Their record might not be that terrifying, but that's not important. They might have 'failed' for all sorts of reasons, reasons which can be addressed. Conversely their records might exceed their abilities (preferential treatment, well-suited to a given system etc).
If you can quantify that in something worthy of statistical analysis, you're a better man than me.
I'm not dismissing stats, but I worry that some people are seeing judgement as subservient to statistics (I'm not assuming anyone's position on this thread, btw). In my mind and in my experience it should be the other way round. Statistics can provide a useful insight and pointer as to where to begin enquiry, but in that specific order. Stats will identify good batman (to follow the example), but - with the possible exception of Hughes and maybe Smith - all of the batsman in the frame for selection are good batsmen and have good records. What separates them?
An approach which would have meant Alistair Cook watching the ashes on TV. I have no doubt that the selectors look at all these stats and they are certainly a decent tool, but they have shown repeatedly that they do not over rely on stats.
Gut instinct continues to play a key role in selection, which of course can backfire, like the aussies leaving out clearly their best spinner (best of a bad bunch) because of some sort of vague feeling that he hasnt got the bottle.
remember seeing him at Trent Bridge in 2005. He was laughable. Completely erratic.
Comments I remember hearing at various points through the Ashes suggest he can't hold it together for more than a couple of overs at a time. This being the case, yeah, I'd have loved to have seen him playing against England
littlebabyjesus
one of Maxwell's demons
Thing is, how do you define a good batsman?
It is by the number of runs he scores.
Mark Ramprakash had the best technique of any Englishman of his generation. But he was a poor test batsman, not for any technical reason, simply because he had a psychological block over scoring more than 40. To go back to Kabbes's point, ramps was very consistent, but never made the big scores, and so he was a failure.
It is by the number of runs he scores.
Mark Ramprakash had the best technique of any Englishman of his generation. But he was a poor test batsman, not for any technical reason, simply because he had a psychological block over scoring more than 40. To go back to Kabbes's point, ramps was very consistent, but never made the big scores, and so he was a failure.
I would, test of nerve, test of technique plus he'd give you loads of freebies. Its the dull line and length stuff I don't want to face.
I think you're hugely underestimating what can be done with statistics. To be fair, the incredibly simple things being spoken about on this thread don't help -- it's like dismissing epic poetry because you don't like limericks.
Think of it this way -- every factor you are using is, by definition, a criterion that you are using some function to evaluate. Even if this evaluation is subconscious, you must be able to order two alternatives to be able to use it to make a judgement.
If you can codify that evaluation, if you can make the evaluation function explicit, then you can use the same data that you are using (i.e. what is in front of you) in a more systematic fashion and produce an objective valuation.
It's not easy, but (a) it's been done for harder tasks than this; and (b) cricket is prime for the task because it is in any case so inherently statistical in nature. Unlike, for example, football.
This is a wider digression, of course. I'm not suggesting that it is what is happening when we look at KP's average score.
Think of it this way -- every factor you are using is, by definition, a criterion that you are using some function to evaluate. Even if this evaluation is subconscious, you must be able to order two alternatives to be able to use it to make a judgement.
If you can codify that evaluation, if you can make the evaluation function explicit, then you can use the same data that you are using (i.e. what is in front of you) in a more systematic fashion and produce an objective valuation.
It's not easy, but (a) it's been done for harder tasks than this; and (b) cricket is prime for the task because it is in any case so inherently statistical in nature. Unlike, for example, football.
This is a wider digression, of course. I'm not suggesting that it is what is happening when we look at KP's average score.
Trescothick's average when he was first picked for England was nothing special. The sort of average loads of county openers have. Clearly Fletcher and at the time he definitely was pulling the selectorial strings) saw something in him that wasn't there in other candidates.
Well yeah, but you don't need stats to tell you that he wasnt up to it (when his chance came). Thing is if a young Ramprakash was selected to play for England now, with this set up, with this changing room mentality I think he would succead.
I was at Cape Town in 2003 watching Shoaib Akhtar bowl against England. Terrifyingly fast but easily left and if you caught one it flew - remember Nick Knight top edging a six over the keeper's head. Neutrals were wetting themselves over him hitting 100mph but England fans were intensely relaxed abbout it because it simply wasn't threatening. Pace isn't everything
"What the selectors saw" is not some kind of soothsaying, though. They are using "data" (i.e. what they can see of a player) and some kind of subconscious algorithm in order to evaluate and order the players.
There is no inherent reason why these data and algorithms should not be made explicit. It's a hard task (and when a mathematician says "hard", he doesn't mean "tricky") but it could theoretically be done.
There is no inherent reason why these data and algorithms should not be made explicit. It's a hard task (and when a mathematician says "hard", he doesn't mean "tricky") but it could theoretically be done.
mattie
missing in inaction
That's my point!
Ramps, and indeed Hick, did not have the necessary 'ability' (synthesise that how you want) to make Test batsmen of a quality that their county stats indicate. The selectors were constantly seduced by said county figures.
From reading a few biographies I'd argue that many such disappointments at Test level were as related to the poisonous changing room atmosphere (itself a function of shitty selection policy, constantly reacting 'the stats') than any inherent lack of ability. The stats should have indicated that the problems lay elsewhere - and every effort should have been made to eradicate those problems.
Even ignoring the kind of advanced statistics I am talking about, this kind of thing is easily addressed by concentrating on test performances rather than county performances.
mattie
missing in inaction
The human is the greatest inference engine going. Any analysis can only answer the questions it's constructed to answer.
We might one day mimic human comprehension artificially, but do not hold your breath.
DrRingDing
'anti-human wanker'
To be fair, I'm a bowler and thus shy of the quick stuff.
And I think you're letting your love of stats and the fact that cricket lends it self so well to them, to obscure the fact that it is only one tool available to the selectors. A useful tool, but also fallable like every other tool.
Cook would have been left at home, Bresnan wouldnt have played.
mattie
missing in inaction
After he's failed? Brilliant!
I think even the most numerically challenged person could figure that Hick struggled at Test level.
To be fair, not even the most brilliant selector with all his intuition could have known Ramprakash would fail before playing him in a test or two.
mattie
missing in inaction
And stats would say he's a failure after a Test or two?
Then why did the selectors play them? What criteria did they use to pick them? And what is the inherent reason why those exact same criteria couldn't be built into an algorithm?
Did the selectors, with all their intuition, only play him for a test or two?
The point is that the same data is being used either way. Statistically or intuitively, you can only get an idea for how somebody performs in tests by playing him in tests.
mattie
missing in inaction
I doubt Flower could articulate them, let alone quantify them.
The problem I am addressing isn't even near the same order of magnitude of difficulty of mimicking human comprehension.
Anyway, human beings are demonstrably terrible at intuiting probabilistic events and correlations. There are any number of experiments that show this very fact. That's precisely why it's useful to gather decent, bias-free data. The whole of science is built around recognising this!
That's why it's hard.
Hard doesn't mean "impossible" though, although it might mean "beyond our ability". I don't think it is beyond our ability in this specific instance, however.
littlebabyjesus
one of Maxwell's demons
Flat track bully who couldn't handle the short stuff. A different problem from Ramprakash. Hick actually had a poor technique, it was just that only the very best bowlers could expose it.
strung out
💩 🤣 🍆 💦 🧐 👻 🐝 🐈⬛
in the 90s mike smith was consistently one of, if not the best english bowler in the country, yet he only played one test. clearly the selectors didn't like the look of him, despite his outstanding figures. i'm still trying to figure out what it was that the panel couldn't see in him, because the stats suggest he should have been a top bowler for england in england.
mattie
missing in inaction
As I've said before, the selectors were going by stats - which is why they dropped people left, right and centre and made the atmosphere so horrendous, because they had absolutely no intuition who they thought their best players were!
This is exactly what people like Fletcher and Flower do - they make a decision based upon the best evidence, be that stats, their judgement, the opinion of the players' county coaches, whatever.
