***Official*** Australia in India 2012/13

[silentstriker]

Thinking about it another way, from here on out, Dhoni is equally as likely to lose more tosses (thereby making his record even worse) as he is to win more tosses (and coming down to 50-50).

 

[ganeshran]

Yes really! Do not Monte Carlo fallcy CW again! It should seriously be a bannable offence.

At any moment in time, the chance of winning the toss is 50%. If you've only won 10% of tosses it doesn't make you more likely to win the next toss than someone who has won 90% of tosses.

When did I claim that throw of a coin was not a discrete event? The probability of a single throw is always 50%.

However the more tosses Dhoni participates in, the closer his win percentage would be to 50%. A throw of 5 unbiased coins may bring 4 heads and 1 tails. 10 coins can also lead to 8 heads and 2 tails.

But as the number of tosses increase, the probability of the total percentage of heads will reach closer to 50%. This is simple probability theory.

 

[Furball]

Where is Gingerfurball?

Ed's biggest booster

Ed of the 30 average, who deserves a place over Watson apparently

 

[ganeshran]

No, it does not. As Prince EWS mentioned.

Each coin toss is an independent event - there is no bias towards going one way or another. If you start counting with an initial situation of (for example): 30 heads and 0 tails, you will always have a bias of +30 heads because the rest of the coin tosses are 'expected' to be 50-50.

This is simply not true. If you get 30 heads in a row, then the probability of getting a head on the the 31st toss is also 50%. Its probability does not increase because of the preceeding 30 heads in a row. This is not what I am contesting.

What is wrong is your assertion that Dhoni will always have a 30 heads bias - that is not true. According to the law of large numbers in probability theory. As the number of trials increase, the total heads will converge to the expected value i.e. 50%. So if Dhoni ends up at 500 tosses even after a 30 heads bias, the final result is likely to be close to 50%. If you keep increasing N, then the final percentage will keep getting closer to the expected value.

 

[ganeshran]

 

[Prince EWS]

This is simply not true. If you get 30 heads in a row, then the probability of getting a head on the the 31st toss is also 50%. Its probability does not increase because of the preceeding 30 heads in a row. This is not what I am contesting.

What is wrong is your assertion that Dhoni will always have a 30 heads bias - that is not true. According to the law of large numbers in probability theory. As the number of trials increase, the total heads will converge to the expected value i.e. 50%. So if Dhoni ends up at 500 tosses even after a 30 heads bias, the final result is likely to be close to 50%. If you keep increasing N, then the final percentage will keep getting closer to the expected value.

Yeah but that's only because that bias of 30 becomes really insignificant when represented as part of a percentage if you've had 1,000,000,000 tosses. It's still there.

 

[ganeshran]

Yeah but that's only because that bias of 30 becomes really insignificant when represented as part of a percentage if you've had 1,000,000,000 tosses. It's still there.

Yup, That is what the law states. The impact of initial skewed ratios go down as you keep increasing total trials.

Just because Dhoni's win percentage is only 37% today. As his career progresses and his total tosses increases, his win percentage will inch closer to 50%, even though the probability of a single win is always 50%.

 

[Furball]

Yup, That is what the law states. The impact of initial skewed ratios go down as you keep increasing total trials.

Just because Dhoni's win percentage is only 37% today. As his career progresses and his total tosses increases, his win percentage will inch closer to 50%, even though the probability of a single win is always 50%.

Not if he keeps losing tosses.

 

[centurymaker]

As his career progresses and his total tosses increases, his win percentage will inch closer to 50%, even though the probability of a single win is always 50%.

Not "will" inch closer.
You can expect it to inch closer but there's no guarantee because each toss is separate from all other tosses.

 

[silentstriker]

Yup, That is what the law states. The impact of initial skewed ratios go down as you keep increasing total trials.

Just because Dhoni's win percentage is only 37% today. As his career progresses and his total tosses increases, his win percentage will inch closer to 50%, even though the probability of a single win is always 50%.

It is expected to inch closer as you go towards a high number of coin tosses. However, even after a million coin tosses, you can still expect a person who started with more wins/loses to have that slight difference compared to a person who started completely evenly.

Let's say Dhoni is +20. Clarke might be +0. If they play another 50 matches, you'd expect (50 isnt large enough but lets say it is) them both to have an even number of heads and tails. At the end of that, Dhoni's percentage would get close to 50% but his tally would still remain +20.

 

[BeeGee]

You're all tossers.

 

[MyLord]

You're all tossers.

Is that why they are nerding out over ... tosses?

 

[ganeshran]

It is expected to inch closer as you go towards a high number of coin tosses. However, even after a million coin tosses, you can still expect a person who started with more wins/loses to have that slight difference compared to a person who started completely evenly.

Let's say Dhoni is +20. Clarke might be +0. If they play another 50 matches, you'd expect (50 isnt large enough but lets say it is) them both to have an even number of heads and tails. At the end of that, Dhoni's percentage would get close to 50% but his tally would still remain +20.

What you are saying is Dhoni wins the first 20 tosses and they win exactly 50% of the remaining matches making their final tally 45-25. Whilst not mathematically impossible, such a scenario is highly improbable.

Assuming 50 is not large enough, lets take it close to a million, there is just as much probability that there are likely to be many more such patches where the success ratio goes entirely in favour of one or the other player. In the end it wont matter who had the better start (the first 20 tosses in a row) and it just as likely for Dhoni to have the slightly higher number as it is for Clarke to have the higher number, even though the difference in the number would be so miniscule that it would represent only a tiny fraction of the total.

This is why gamblers always lose money on casino games involving chance. the probability of winning at any game is skewed in the casino's favour. If a large enough number of gamblers play for a long enough time, the odds will be satisfied and the casino makes money.

 

[Burgey]

You're all tossers.

Good from Barry Gibb.

 

[social]

Why is it that some of Australia's lower order batsmen have more sound defensive techniques than our top order?

Take Warner, Hughes and Smith as examples

These guys have obvious ball-striking gifts but Starc, Pattinson, and Siddle generally play better defensive shots than they could dream of doing

Seems to me that the selctors identify talented batsmen and just allow them to do their thing whilst the lesser lights do the hard work

 

[silentstriker]

What you are saying is Dhoni wins the first 20 tosses and they win exactly 50% of the remaining matches making their final tally 45-25. Whilst not mathematically impossible, such a scenario is highly improbable.

Actually, it is the most expected scenario if you start from the baseline, of +20 for Dhoni.

Assuming 50 is not large enough, lets take it close to a million, there is just as much probability that there are likely to be many more such patches where the success ratio goes entirely in favour of one or the other player. In the end it wont matter who had the better start (the first 20 tosses in a row) and it just as likely for Dhoni to have the slightly higher number as it is for Clarke to have the higher number, even though the difference in the number would be so miniscule that it would represent only a tiny fraction of the total.

No, the bolded is not true. The difference would be miniscule but statistically Dhoni would be more likely to have a higher number - slight as it may be. You can't go back in time and undo the coin tosses you already have. Each additional one is exactly 50% - there is no undoing of past events.

 

[Furball]

Why is it that some of Australia's lower order batsmen have more sound defensive techniques than our top order?

Take Warner, Hughes and Smith as examples

These guys have obvious ball-striking gifts but Starc, Pattinson, and Siddle generally play better defensive shots than they could dream of doing

Seems to me that the selctors identify talented batsmen and just allow them to do their thing whilst the lesser lights do the hard work

Same thing with some of England's bats. I reckon James Anderson is up there in terms of his footwork against spin. He's usually very decisively forward or back, and showed up his colleagues in the UAE.

 

[ganeshran]

Actually, it is the most expected scenario if you start from the baseline, of +20 for Dhoni.



No, the bolded is not true. The difference would be miniscule but statistically Dhoni would be more likely to have a higher number - slight as it may be. You can't go back in time and undo the coin tosses you already have. Each additional one is exactly 50% - there is no undoing of past events.

the point is that whatever the difference is, it would not be statistically significant. I think we can both agree on the fact that for a large enough N, the final percentage of toss successes would be very close to 50%.

If Dhoni goes on to captain India in a million test matches , it is a fair enough assumption to say that his final tally of toss wins would be very close to half a million, even though it might be very very slightly lesser than 50% because of his poor present record

 

[Burgey]

Why is it that some of Australia's lower order batsmen have more sound defensive techniques than our top order?

Take Warner, Hughes and Smith as examples

These guys have obvious ball-striking gifts but Starc, Pattinson, and Siddle generally play better defensive shots than they could dream of doing

Seems to me that the selctors identify talented batsmen and just allow them to do their thing whilst the lesser lights do the hard work

I have my own theory on this, and it's climatic.

The current generation have grown up playing their junior cricket in an era of a prolonged drought. Flat pitches and little if any swing for the bowlers. Theirs is a generation which could pick the line and hit through it without the ball really doing much by way of sideways. The past few years the conditions have changed, but I don't think their games have adjusted.

 

[benchmark00]

Dodgy Indian joint across from the cinema? That place is the best.

Mall.