If Sir Donald Bradman was born an Indian

[MCcricket]

If Bradman was Indian then Sachin would have been termed an average players by our fans lol

[zen]

It is difficult to be an outlier now because of this: vs this: With the population (and thus competition) rising, how will I ever be an outlier? :hitler:

[The Outsider]

Just a rejoinder to my previous post. Bradman is around a 1 in 200,000 outlier based on all batting averages. Freakonomics has done at least part of the plot and it looks close to a normal distribution by eye balling things : Given this, the hypothesis of 1.5 billion can easily be tested. If that was the correct data sample, we would have seen hundreds if not thousands Bradmans coming out by now, most of them from India. Do we see such a thing? No.

[zen]

Rett: With the population (and thus competition) rising, how will I ever be an outlier? :hitler: Santa: A star always shines Rett: But the competition is like a big black opaque cloud that doesn't let my light through. I need open space to shine .... and the competition is rising (more clouds) :(( Santa: The strong wave breaks the barrier Rett: The increasing population and competition is like a big human dam. It is so powerful that it will break the wave Santa: The mitchell bird's sweet song is always heard Rett: The competition makes so much noise that even I can't hear my own voice :sniffle: Santa: The sun always gives the warmth Rett: But ..... :cry_smile:

[zen]

To show that how is it easy for 'the cream of batting' to avg 50 even today and why you cannot consider someone avg in 50-60 as an outlier (like Bradman), I have an example below: Sachin Tendulkar from 2004 t0 2006 played 24 tests w/ an impressive avg of 52. Some folks thought he was out of form during that period but with numbers like these (see below), how could he have been out of form in that period: Year - Avg 2004 - 93 2005 - 44 2006 - 27 From the above one can say that he only had one bad year, that's it! But is the truth? Let's dig further: In this period, he had 37 innings w/ 6 not outs, to get to 1626 runs Avg: 52 RPT: 68 RPI: 44 Of the 6 not outs, he had the following scores: * against Aus - 248* (when Ind scored 700+) * against Pak - 194* (Sehwag scored a triple) * against BD - 248* That's 683 not out runs of those 1626 runs coming from 3 innings! This also implies that he had just 943 runs in the other 34 innings Does this show something negative about SRT? No, it doesn't. But what it shows that if a 'quality' batsman, who is untouchable in his team, keeps playing it is not hard for him to avg in 50s (even when he is not suppose to be in best of form). Avg in 50-60 doesn't make you an outlier even though you have played some 180 tests (9 tests per year in a 20 years career) and sometimes playing more helps to adjust your numbers :winky: .... As they say form is temperory but class is permanent! *today, i have some time to kill :P *

[pensionplan]

*today, i have some time to kill :P *

Paging the wolf pack...Paging the wolf pack...

[ravishingravi]

To show that how is it easy for 'the cream of batting' to avg 50 even today and why you cannot consider someone avg in 50-60 as an outlier (like Bradman), I have an example below: Sachin Tendulkar from 2004 t0 2006 played 24 tests w/ an impressive avg of 52. Some folks thought he was out of form during that period but with numbers like these (see below), how could he have been out of form in that period: Year - Avg 2004 - 93 2005 - 44 2006 - 27 From the above one can say that he only had one bad year, that's it! But is the truth? Let's dig further: In this period, he had 37 innings w/ 6 not outs, to get to 1626 runs Avg: 52 RPT: 68 RPI: 44 Of the 6 not outs, he had the following scores: * against Aus - 248* (when Ind scored 700+) * against Pak - 194* (Sehwag scored a triple) * against BD - 248* That's 683 not out runs of those 1626 runs coming from 3 innings! This also implies that he had just 943 runs in the other 34 innings Does this show something negative about SRT? No, it doesn't. But what it shows that if a 'quality' batsman, who is untouchable in his team, keeps playing it is not hard for him to avg in 50s (even when he is not suppose to be in best of form). Avg in 50-60 doesn't make you an outlier even though you have played some 180 tests (9 tests per year in a 20 years career) and sometimes playing more helps to adjust your numbers :winky: .... As they say form is temperory but class is permanent! *today, i have some time to kill :P *

:two_thumbs_up:

[Raghav_12]

Just a rejoinder to my previous post. Bradman is around a 1 in 200,000 outlier based on all batting averages. Freakonomics has done at least part of the plot and it looks close to a normal distribution by eye balling things : Given this, the hypothesis of 1.5 billion can easily be tested. If that was the correct data sample, we would have seen hundreds if not thousands Bradmans coming out by now, most of them from India. Do we see such a thing? No.

This graph is absolutely out of context. I had talked about of PDF of Bradman's time and PDF of current time. You have put all batsman in same graph and started pro-claiming that they follow normal distribution. And this looks like normal distrubution curve? No. It doesn't if you don't include if and but's of conditions and assumptions. You can take opinion of other junta of forum. I believe many people on this forum would know normal distribution curve very well. This graph has been drawn from 87 data points, inlcuding batsmen from all eras. If you concentarte on Bradman era. You would find it difficult to get 87 international players, forget about 87 great batsmen

[The Outsider]

This graph is absolutely out of context. I had talked about of PDF of Bradman's time and PDF of current time. You have put all batsman in same graph and started pro-claiming that they follow normal distribution. And this looks like normal distrubution curve? No. It doesn't if you don't include if and but's of conditions and assumptions. You can take opinion of other junta of forum. I believe many people on this forum would know normal distribution curve very well.

It is more than you have produced for sure and I agree it's not the entire sample but if the drape does not remind you if a normal distribution, then nothing will. Notice how there are more batsmen in the 45 range and going progressively down till 60. What kind of a distribution does that remind you of? I don't need the junta's vote on something scientific.

This graph has been drawn from 87 data points, inlcuding batsmen from all eras. If you concentarte on Bradman era. You would find it difficult to get 87 international players, forget about 87 great batsmen

How is that relevant to my points in my previous two posts? I've clearly mentioned that the graph is drawn from all batsmen in my previous posts, so what are you trying to show in your supposed rejoinder?

[The Outsider]

BTW, Raghav I noticed that you just chose to respond to the rejoinder of my post and the main reply. Can you please let us know the reasons for it: a.) You will respond to the main post later? b.) You will not respond to the main post but try to mudsling the argument according to what you understood in the rejoinder? c.) You don't care about the posts as long as you can try to peddle agendas like the human species have genetically evolved in the last 50 years?

[Raghav_12]

I am using a method to identify statistical outliers, called DFFITS. You can read more about it to see how the 1/sqrt(n) term comes about in it. It's a commonly used and powerful technique to identify statistical outliers. How I calculated the numbers? DFFITS says a data point can be identified as a statistical outlier depending on the value of some constant, call it K divided by sqrt(n). The point does not have to be an outlier, but the above can be used to identify how far away it is from the sample. Say, there were 20 international batsmen at Bradman's time then his average of 100 will be equal to some constant divided by sqrt(n) : K/sqrt(20) = 100 For a data point to be at the same level in a sample of 50 international batsmen to the extent Bradman was in 20, say it's value has to be x. Then x = K/sqrt(50) = 100*sqrt(2/5) which is roughly 65. The only assumption which goes in here is that the underlying samples have similar distribution. You can plot a histogram of number of batsmen and their averages and check that - I am pretty sure they'll be similar enough. For DFFITS to apply, they don't have to be identical distributions.

I am not an on expert DFFITS so can't comment on the way you have applied this method here. This way your own method proves that statistically speaking Ramiz Raza was in fact a better bat than Hammond. You have any chance of your application of this method being taken seriously if you agree with that statement. Your this method falls flat on many other account: 1. You didn't answer properly what would have Bradman's average if there were 15 test team playing today with 75 internation batsmen. You yourself admitted that you can not consider exact sample size of current players because a lot many cricket is being played nowadays. This is exactly something which I have been arguing - you can not compare Bradman with current batsman using merely statistical parameters as too little cricket was played in that time. 2. You allocated India 3 data points when in India cricket was played by Kings, Princesses, their servants and cooks, Soldiers only. And you allocated only 5 points when cricket is played by each Ramu, Kallu, Bahadur and Munna at each street corner.

15 test teams? Anyhow, the number of test teams is not relevant here. What's relevant is the number of international class batsmen - you can keep a clear cut off for them at some reasonable average like 30-35 or so and as long as you do it the same way for both samples it doesn't matter. There would probably be some adjustments which would need to be done to weigh different cut offs for the two samples because a lot more cricket is played today.

Thats where you admitting that comparison on statistical parameters is not possible.

My turn to call something absurd? The point I made was not about the difficulty of getting into Harvard, but about the performance once the students are in Harvard. If you want to do a statistical study of students' performance while in Harvard (batsmen playing test cricket) will you include everyone who took their SAT (batsmen playing club cricket). To make the analogy clearer. Let's say Harvard admitted 1500 students 40 years back and someone scored a 95 on some relative scale, which let's say qualifies him for a special award. Today if Harvard admits 2000 students what should the score be to be an outlier of the same magnitude so that Harvard can identify if it should hand out that special award? This is a very well defined statistical problem, which can be solved. Want to give it a try?

No, you are out of turn :-). Credit that a special award winner, for being a top performer at Harvard gets, is not because of he was best performer in a crowd of 2000. But he/she deserves credit because he topped in a class of very high quality 2000 student (creme de la creme) who came through very tough competition. If overall quality of class ( equal to sample size), which eventually come because of high level of competion, is immaterial than a Harvard top performer should not get more recognition than the topper of Bundelkhand University, Jhansi.

So, a cricket crazy country like India are monkeys that they have not been able to produce even as many great cricketers as the tiny island of Barbados? Well, in that case that's a self defeating argument because then the vast Indian population has made little impact on increasing the competitiveness in cricket. It's not just Barbados or West Indies take Australia if you want - they have a population the size of maybe the greater Delhi area or a bit more? New Zealand has a population the size of Lucknow, maybe? South Africa, Delhi and Bombay put together? On the flip side, Bangladesh has the third largest population among cricketing countries.

You are nit-picking here. My point was there are more factors which determine quality, sample size alone can not guarantee a better quality. Compare the difference in number of great players that Barbados and Bombay produced respectively before 80's and after 80's. For sake of comparsons with Bradman's era, Barbados also emerged after Bradman's time. If emergence of India, Pakistan, WI, SL, SA had not increased the competitiveness in world cricket than England and Australia put together would have won 10 WC's till now, but in fact they won 4 only. In fact, if only Australia and England played world cup, then Aus would have won some 8-9 of them and by sport journalist would have anointed that as the most comprehensive dominance of one team over rest of the world that ever happened in field of any sport.

I am not bothering about the Bradman or Tendulkar argument anymore. My point is that argument that things have become so competitive that it's not possible to dominate peers anymore is wrong on two counts: 1. It is quantifiable. I've given a rough way to quantify it and you can check up the paper I was referring to for a peer reviewed and complete way. 2. Sportsmen like Jahangir Khan and Michael Jordan have shown that it is possible to dominate your peers by a very large degree even in the modern era.

Let me know the paper that you talked about. Jahangir Khan hardly qualifies as Squash in not a very popular sport anyways. By this example, you indirectly suggesting that if sport is not very popular then you can have outliers. I don't follow Basketball, so can't comment on Jordan thing.

[Raghav_12]

BTW, Raghav I noticed that you just chose to respond to the rejoinder of my post and the main reply. Can you please let us know the reasons for it: a.) You will respond to the main post later? b.) You will not respond to the main post but try to mudsling the argument according to what you understood in the rejoinder? c.) You don't care about the posts as long as you can try to peddle agendas like the human species have genetically evolved in the last 50 years?

I guess you got the answer. :haha: I was late because I do have some life outside this forum as well. And through this post of yours you have proved that you not interested in genuine argument but more interested in upmanship, proving how good you are, and these are somethings that you have been accused of many times on this forum.

[The Outsider]

I guess you got the answer. :haha: I was late because I do have some life outside this forum as well. And through this post of yours you have proved that you not interested in genuine argument but more about upmanship, that you have been accused many times on this forum.

You are wrong. I was genuinely amused that you chose to have the time to respond to a rejoinder and not the main post, but I'll get back on the garbage you have posted above - not knowing DFFITS and talking with vigor about statistical outliers! :wall:

[Raghav_12]

It is more than you have produced for sure and I agree it's not the entire sample but if the drape does not remind you if a normal distribution' date=' then nothing will. [b']Notice how there are more batsmen in the 45 range and going progressively down till 60. What kind of a distribution does that remind you of? I don't need the junta's vote on something scientific. How is that relevant to my points in my previous two posts? I've clearly mentioned that the graph is drawn from all batsmen in my previous posts, so what are you trying to show in your supposed rejoinder?

It can remind me of any distribution be it poisson, binomial or decaying exponential. In most of the distribution when you move away from median, frequencies keep going down progressively. Last, I know normal distrubution should have following qualities: 1, It should have two tails. 1. There should be a mean some where between in the graph. Mean and Median should be same. Let me know how many of these you see in this graph of yours.

[Raghav_12]

You are wrong. I was genuinely amused that you chose to have the time to respond to a rejoinder and not the main post' date=' but I'll get back on the garbage you have posted above - not knowing DFFITS and talking with vigor about statistical outliers! :wall:[/quote'] I didn't know that you have a rule at ICF that you should respond to posts strictly in the order that were posted. I responded to your rejoinder first as I have always been told, attack very simple problems, which you can get out of your way very quickly, first. I didn't know about DFFITS but I know how normal distribution curve looks. About my knowledge on stats, I know where I stand. If you think I have posted garbage, why you should bother to reply back . I myself would not bother.

[Guest BossBhai]

-- Removed on request of the user --

[CSK Fan]

To show that how is it easy for 'the cream of batting' to avg 50 even today and why you cannot consider someone avg in 50-60 as an outlier (like Bradman), I have an example below: Sachin Tendulkar from 2004 t0 2006 played 24 tests w/ an impressive avg of 52. Some folks thought he was out of form during that period but with numbers like these (see below), how could he have been out of form in that period: Year - Avg 2004 - 93 2005 - 44 2006 - 27 From the above one can say that he only had one bad year, that's it! But is the truth? Let's dig further: In this period, he had 37 innings w/ 6 not outs, to get to 1626 runs Avg: 52 RPT: 68 RPI: 44 Of the 6 not outs, he had the following scores: * against Aus - 248* (when Ind scored 700+) * against Pak - 194* (Sehwag scored a triple) * against BD - 248* That's 683 not out runs of those 1626 runs coming from 3 innings! This also implies that he had just 943 runs in the other 34 innings Does this show something negative about SRT? No, it doesn't. But what it shows that if a 'quality' batsman, who is untouchable in his team, keeps playing it is not hard for him to avg in 50s (even when he is not suppose to be in best of form). Avg in 50-60 doesn't make you an outlier even though you have played some 180 tests (9 tests per year in a 20 years career) and sometimes playing more helps to adjust your numbers :winky: .... As they say form is temperory but class is permanent! *today, i have some time to kill :P *

Dude almost all cricket fans acknowledge these 3 years as the worst of Sachin's career so they are very well aware of the fact. What your analysis also shows is how easy its to average higher if you only play against a weaker opposition and you yourself ackonwledge that. If Sachin had not played against stronger opposition in the same period, his average would have sky-rocketed despite his being so out of form. So if you play on the same pitch, same bowling attack and same opposition thorughout your career, you can easily average in the 90s because the normal curve which brought down Sachin's average to 50 wouldnt be there

[zen]

Dude almost all cricket fans acknowledge these 3 years as the worst of Sachin's career so they are very well aware of the fact. What your analysis also shows is how easy its to average higher if you only play against a weaker opposition and you yourself ackonwledge that. If Sachin had not played against stronger opposition in the same period' date=' his average would have sky-rocketed despite his being so out of form.[/b'] So if you play on the same pitch, same bowling attack and same opposition thorughout your career, you can easily average in the 90s because the normal curve which brought down Sachin's average to 50 wouldnt be there

If I am not wrong, you posted the similar stats on some other thread to show Tendulkar was not out of form or struggling (something like that) .... so good to know that you are acknowledging that in fact he was! Weaker opposition is a relative term though. It is in relation to the stronger team of your time. And it also depends upon players on how they can motivate themselves when they are playing against a relatively weaker opposition. If you see Gavaskar's record, you will find he avgs highly against the strongest team of his time than against some of the other teams. If you take Ranji trophy, where teams are relatively weaker, I don't see Tendulkar having an exceptional record You are making the classic mistake of assuming that if had more teams were like BD, someone like Ten would avg more, when he could even avg less like he does in Ranjis. One of the reasons, he is able to avg more against BD is that he is able to compare them with Australia for example and "probably" set up goals for himself like he can bump up his avg .... Relatively speaking someone can even say that if someone like Bradman had played more against Ind and SA, he would easily avg 200 :winky: And, of the three not out innings, I listed, 2 were played Aus and Pak so ....

[The Outsider]

I am not an on expert DFFITS so can't comment on the way you have applied this method here. This way your own method proves that statistically speaking Ramiz Raza was in fact a better bat than Hammond. You have any chance of your application of this method being taken seriously if you agree with that statement. Your this method falls flat on many other account:

Think you've answered your own question there - the method does not choose who is better only identifies to what level a sample is a statistical outlier. Applying to Hammond and Rameez Raja, the only answer you will get is neither are statistical outliers.

1. You didn't answer properly what would have Bradman's average if there were 15 test team playing today with 75 internation batsmen. You yourself admitted that you can not consider exact sample size of current players because a lot many cricket is being played nowadays. This is exactly something which I have been arguing - you can not compare Bradman with current batsman using merely statistical parameters as too little cricket was played in that time.

I am not even bothered or talking about what Bradman's average would have been in the current discussion. Identifying the degree to which a sample is a statistical outlier is not an exercise in prediction.

2. You allocated India 3 data points when in India cricket was played by Kings, Princesses, their servants and cooks, Soldiers only. And you allocated only 5 points when cricket is played by each Ramu, Kallu, Bahadur and Munna at each street corner.

The 3 and 5 data points you are referring to apply to international level batsmen from India playing test cricket at that point in time.

Thats where you admitting that comparison on statistical parameters is not possible.

If such a simple problem could not be solved with statistics, then the field should be junked. I am saying that one has to be careful and scientific when applying the concepts.

No, you are out of turn :-). Credit that a special award winner, for being a top performer at Harvard gets, is not because of he was best performer in a crowd of 2000. But he/she deserves credit because he topped in a class of very high quality 2000 student (creme de la creme) who came through very tough competition. If overall quality of class ( equal to sample size), which eventually come because of high level of competion, is immaterial than a Harvard top performer should not get more recognition than the topper of Bundelkhand University, Jhansi.

I gave you a sample undergraduate stats problem to solve in my previous post. Once you work on it and come up with a solution, you would have answered your above question.

You are nit-picking here. My point was there are more factors which determine quality, sample size alone can not guarantee a better quality. Compare the difference in number of great players that Barbados and Bombay produced respectively before 80's and after 80's. For sake of comparsons with Bradman's era, Barbados also emerged after Bradman's time. If emergence of India, Pakistan, WI, SL, SA had not increased the competitiveness in world cricket than England and Australia put together would have won 10 WC's till now, but in fact they won 4 only. In fact, if only Australia and England played world cup, then Aus would have won some 8-9 of them and by sport journalist would have anointed that as the most comprehensive dominance of one team over rest of the world that ever happened in field of any sport.

Nit picking? Do a simple exercise - plot the number of quality cricketers produced by each test country over the last 30-40 years (the time when cricket has become competitive according to you and some others) and check if you see any positive correlation. If you can't make out from the plot, do a simple regression and get back to me with the correlation coefficient.

Let me know the paper that you talked about. Jahangir Khan hardly qualifies as Squash in not a very popular sport anyways. By this example, you indirectly suggesting that if sport is not very popular then you can have outliers. I don't follow Basketball, so can't comment on Jordan thing.

I'll try to search for the link. You can also try - I've posted the link before on ICF. Squash doesn't qualify and you don't follow Basketball? So, now sport outliers will have to be identified based on which sport Raghav follows or considers worthy of qualification. Just a small point about squash - Tendulkar won't be able to last on a squash court for more than 30 minutes. If you want some scientific justification for it, it is one of the most calorie intensive sport and one hour will normally burn 1000 calories.

[The Outsider]

There are 75 International class batsmen today going by avg > 40 with a min of 20 Tests. Same criteria in the 20s and 30s results in 18 batsmen which is close to the 20 batsmen figure for Bradmans time that you assumed in your calculation. DGB http://stats.espncricinfo.com/ci/engine/stats/index.html?class=1;filter=advanced;orderby=batting_average;qualmin1=20;qualval1=matches;size=25;spanmax2=14+Aug+1948;spanmin2=30+Nov+1928;spanval2=span;template=results;type=batting SRT http://stats.espncricinfo.com/ci/engine/stats/index.html?class=1;filter=advanced;orderby=batting_average;page=3;qualmin1=20;qualval1=matches;size=25;spanmin2=15+Nov+1989;spanval2=span;template=results;type=batting taking a round figure of 20 instead of 18 .... this gives us a equivalent modern day batting avg of = 99.94*sqrt(20/75) = 51.6 SRT (and many others) has done much better than this over a career that is nearly 4 times as long than Bradmans even if you want to pretend that the bowling and fielding are of the same standard today and ODI cricket is inconsequential. cue : let the sophistry and insults begin

Here is a suggestion - enroll in a Stats 101 course in your nearest community college before coming here and spouting garbage. They'll teach you how to select samples. I'll give you some hint for free though - the sample would be the number of international quality batsmen playing at a particular point in time, not the total number as you've given in your links. Try to think about, why that would be the case. I'll give you another hint for free - since the number of matches being played in the modern era is much higher the turnover would be higher which would bias your sample selection using the method you used.