Why cant there be another Bradman again??

[Guest BossBhai]

-- Removed on request of the user --

[zen]

Appears as if "Tendulkar is the greatest" brigade is licking their wounds :giggle: And it's always fun to read "mainpulators" get beaten by their own game and then as usual restort to whining. They lost not because they didn't have the "DATA" but because someone didn't buy their data and on top of that that someone used smileys, what Blasphemy :hysterical: So now they will take their revenge on the one who used the smileys :hehe: in short, dumb DATA and DRAMA don't work :winky:

[goose]

Standard deviation. However, it's not immediately clear how useful a measure it is in cricket. Firstly, to even go to a second level of measure, batsmen should have comparable averages. There is no advantage in having a very consistent batsman who averages 20 over an inconsistent one who averages 50. Secondly, it's not immediately clear whether more consistency around the average is actually good or not. For example, two players can have the same average and scores like : 1. 40, 50, 45, 55, 50, 50 to average 50 2. 120, 10, 30, 110, 30, 0 to average 50. Player 1 is much more consistent than player 2. Who would you have in the team? There are arguments to be made for both. Personally I would have player 1 in ODIs and player 2 in Tests.

Granted it would be more meaningful to compare batsman with similar averages. Player 2 leaves himself exposed to the idea that his high scores come under favorable circumstances e.g. on certain wickets or against certain attacks. He needs a start. Player 1 finds a way to take the circumstances out of the equation - he overcomes. Some guys do an especially good job at being Player 2 (Ponting, Lara). Few can be Player 1 (Tendulkar).

[thongale]

Maybe this extract can help him understand - Bowler Challenge Index (BCI) The Bowler Challenge Index evaluates the survival of these batsmen against the best bowlers of their time. BCI is calculated as the ratio of the Peak ICC Ratings of the Top 10 Bowlers the batsman faced to the Dismissals per Match that these bowlers were able to inflict on the batsman. Only bowlers who have played more than five Tests against the batsman are chosen. Therefore, it is measuring the quality of bowlers faced by the batsman during his career and his survival against these bowlers. The theory that the bowlers of the 90’s were the best that Test cricket ever witnessed is confirmed here. Notice that the legends of the 90’s Inzamam, Lara, and Sachin along with Rahul Dravid are way ahead of the rest. Obviously when you face bowlers of the caliber of Wasim Akram, Waqar Younis, Curtly Ambrose, Courtney Walsh, Allan Donald, Shaun Pollock, Brett Lee, Glenn McGrath, Shane Warne, Anil Kumble, and Muthiah Muralitharan at their very best, you would top this parameter. The high value for Javed Miandad is due to the fact that he was rarely dismissed by the top bowlers of his time. Whereas Sunil Gavaskar although faced a lot of the top bowlers at their very best, he was dismissed by them often and hence scores less compared to the others. Also notice the very low numbers for the English greats. This is primarily because of the relative strength of the bowlers during their time. It would be fair to say that Inzamam, Lara, and Sachin faced the toughest bowlers in Test cricket history while these bowlers were at their peaks. Although the West Indies Pace Quartet, Indian Spin Quartet, the Aussie speed demons, and the 4 Great All-Rounders operated during the 70s and 80s their peaks did not coincide for long periods because of which not many batsmen of the time faced all groups at their very best for extended periods. Name : Bowler Rating Score : DpM : BCI = BRS / DpM Barrington : 1672.8 : 0.40 : 4215.46 Chappell : 1692.9 : 0.4 : 4232.25 Dravid : 2437.8 : 0.43 : 5723.53 Gavaskar : 2072.8 : 0.52 : 3962.71 Hammond : 943.3 : 0.46 : 2043.82 Hobbs : 1037.5 : 0.39 : 2658.59 Hutton : 1186.1 : 0.36 : 3261.78 Inzamam : 2474.3 : 0.41 : 5974.53 Kallis : 1866.7 : 0.43 : 4340.8 Lara : 2356.2 : 0.41 : 5690.45 Miandad : 1772.8 : 0.30 : 5875.57 Pawning : 1789.8 : 0.45 : 4000.73 Viv : 1967.3 : 0.40 : 4898.18 Sobers : 1305.8 : 0.34 : 3798.69 Sachin : 2437.8 : 0.43 : 5688.20 Bradman : 669.1 : 0.41 : 1639.3 Inzi has faced the toughest bowling ever, and the bowlers he, Sachin, Lara, and Dravid faced were almost 4 times (2400 v/s 600) that of Bradman. Compare Bradman to his peers...Hobbs : Bradman :: 1037 : 669 Hammond : Bradman :: 943 : 669 Now, Inzi : Bradman :: 2474 : 669 Sachin : Bradman : 2437 : 669 Lara : Bradman : 2356 : 669

That analysis is TOTALLY flawed when you consider PEAK values. Botham with a oeak value of 905 or some where is not even close to Walsh who has a lower peak value. Should have been the AVERAGE RATING during their careers.

[Nirvanam]

That analysis is TOTALLY flawed when you consider PEAK values. Botham with a oeak value of 905 or some where is not even close to Walsh who has a lower peak value. Should have been the AVERAGE RATING during their careers.

Thongale, pls see my post on the new thread about these ratings...that will give you a better idea of what the ratings are - http://www.indiancricketfans.com/showthread.php?t=246876

[thongale]

Thongale, pls see my post on the new thread about these ratings...that will give you a better idea of what the ratings are - http://www.indiancricketfans.com/showthread.php?t=246876

Still the fact stands that you have used PEAK ratings instead of AVERAGE one, and it's a massive mistake. Since you are harping on IC rankings so much, this will show that Bradman, during 40% of his career did better than the PEAK performance of the next best. That speaks volume about him. Then considering top 10 bowlers is another mistake. Bradman had only 3 teams to play with, so his top 10 bowlers will include second rate ones too compared to a modern player who has 8 oppositions to choose from. It should be averaged across similar proportion of players. Say first n X 2 players, where n is the number of oppositions

[dial_100]

Appears as if "Tendulkar is the greatest" brigade is licking their wounds :giggle:

Wrong. I will speak for myself and I am simply ignoring you and will always do. I just wanted to let you know that. You only worth that. Remember, the two you are arguing with are not the only people who are strong SRT followers. So see if you stop generalizing them, it will be good for you. Also extending this request to others group as well. Please stop personal abuses just because others have different opinions. If you want to continue discussing it then keep away from personal attacks. Someone has to mature here. I dont know which one of you two will take the first step.

[Raghav_12]

Wrong. I will speak for myself and I am simply ignoring you and will always do. I just wanted to let you know that. You only worth that. Remember, the two you are arguing with are not the only people who are strong SRT followers. So see if you stop generalizing them, it will be good for you. Also extending this request to others group as well. Please stop personal abuses just because others have different opinions. If you want to continue discussing it then keep away from personal attacks. Someone has to mature here. I dont know which one of you two will take the first step.

I think moderators have to intervene here. This thread after some good discussion turned in fighting match of personal abuses. Another thread was created and that did not last long. Yet another thread is going some way. Few posters are starting with personal abuse from their very first post in latest thread.

[Nirvanam]

Still the fact stands that you have used PEAK ratings instead of AVERAGE one, and it's a massive mistake. Since you are harping on IC rankings so much, this will show that Bradman, during 40% of his career did better than the PEAK performance of the next best. That speaks volume about him. Then considering top 10 bowlers is another mistake. Bradman had only 3 teams to play with, so his top 10 bowlers will include second rate ones too compared to a modern player who has 8 oppositions to choose from. It should be averaged across similar proportion of players. Say first n X 2 players, where n is the number of oppositions

Isn't that the whole point I am trying to make by presenting the ICC ratings of the bowlers? Your point about using the Average rating is definitely better, and I myself wanted to do it that way but the constraint of data collection methodology forced me to pick either the peak, or the bottom...I choose the peak. I can assure you it is only the difficulty of data collection process otherwise I'd have taken the average itself. A quick explanation of the data collection methodology for peak rating follows - 1. Go to Statsguru and query for all the bowlers the batsman played against. Copy-paste results to Excel. You will have on an average 100 bowlers per batsman. 2. Pick bowlers who played > 5 games against the batsmen 2. Next by a quick scan of the names figure out who were proper bowlers and separate them from the list. Records per batsman left = 50. Total records.. 50 * 16 = 800 3. Intuitively pick the bowlers out and bring it down to about 20 bowlers by verifying their ratings visually on Reliance Mobile ICC Ratings website. 4. For every bowler, go to his ICC ratings graph...move the mouse over the graph and the tooltip shows the date of the Test match...move your mouse/finger-pad in multiples of about a millimeter to traverse the graph such that you do not miss out any of the Tests. 5. You already know the dates you are looking for (the main downloaded data of batsman's innings by innings list also has the start date of the match). So you will have a minimum of 5 data points (usually about 10 - 12 datapoints per bowler i.e. 120 for top 10 bowlers) to look at on the graph. 6. Comprehend the ratings score and pick out the best rating for those 5 (or 12) matches when that bowler played that bastman. In case I have to use Average ratings then I have to make at least 11*10 = 110 more data entries in Excel per batsmen. For 16 batsmen, 16*110 = 17000 approx (which is 1100 times that of the peak rating data entries) There is no other way to do this...I requested them to give me data for at least a few bowlers but they didn't respond. The only way is to move your mouse over and traverse millimeter by millimeter over the contours of the graph on the ICC rankings website. Now, the logical part....I guess you have an idea of how the Ratings are calculated and typically how much change happens in rating points match over match. We know that the ratings above are all Peak ratings, so if we take the average over the 5 (or 12) Tests the bowler played against the batsman then it would necessarily be lesser than the peak. Question is how lesser? Make a guess...you could argue it could be 100 or it could be 1 or it could be anything in between. But just make an educated guess...I'll guess it will be about 25 points lesser. So use this logic and reduce the ratings of all ten bowlers by 1 point for Bradman (or don't even do that let's assume peak and average are same for Bradman's case) and 25 points for the rest. What'll be the new values? How much would you gain from them? Bradman obviously cannot go higher than 669 coz we are already using peak ratings for him. For the others reduce 25*10 / 10 = 25 points from theirs. So Inzi would be 855 - 25 = 830...alright if you wanna make it even more severe let's say Inzi's reduces by 50 points...his new numbers would be 855 - 50 = 805....what's the great difference? Does it make Bradman's bowlers look any better? hey Thongale, I'll request you to come to the other thread and discuss the stats there...please :-)

[zen]

The comical complaining is still going on. Appears as if people are frustrated at their struggle to get a foothold :giggle:

[thongale]

Now Allan Border played better bowlers than Tendulkar. So Border > Tendulkar?

[Nirvanam]

Now Allan Border played better bowlers than Tendulkar. So Border > Tendulkar?

Thongale' date= I request you to come to the other thread once again. I'll answer your question about Border > Tendulkar. - That is a Strawman logical fallacy. Why? Because the argument is not what you are assuming. You are assuming the argument is the following. The better the quality of the Top 10 bowlers, as per ICC Peak ratings of the bowlers when bowling to the said batsmen, the better is the batsman. Please note, that is not the argument. Just in case you'd also like to know the truth value of your assertion that Border > Tendulkar, I am extremely uncertain that it is true based on the parameter of interest in this argument (that of ICC Peak ratings of the bowlers when bowling to the said batsmen). Unless you go to the ICC Ratings website and traverse through the graphs, millimeter by millimeter, of the players you will not be in a position to make that assertion. The links you have provided are not about this parameter at all.

[zen]

Quality Index After years of extensive research (actually few minutes) and data analysis, yours truly (I am sure many would have done the same before me), finally managed to come up with the quality index, where by batting/bowling across the eras can be compared. So what did rett do? Basically, I calculated the batting and bowling averages of the top 6 runs getters and top 6 wicket takers for different periods, excluding Bradman's. Why top 6 and not 7 or 5 or not general avg? Well, top 6 because Sir Gubby Allen was at #6 and I wanted to include him. And usually when people talk abt top, the numbers are like top 5, 7 wonders, etc so I thought let's take 6. Another reason of taking top 6 is that for 1 JAn 1989 to todays period, the top 3 wkts taker are spinners and the next 3 are seamers so I thought that would be a good balance :P Why not general avg? Well, I figured that the top 6 would have bowled and batted a lot so that would give us a good idea about the standard of top quality. What's the conspiracy behind excluding Sir Don Bradman's numbers? No conspiracy there. He has been so far ahead that bringing him in to this may not account for general overall top quality. What did Rett find? 1 Jan 1928 to 1 Jan 1949: Avg batting avg of top 6 (excluding DB): 52.73 Similar bowling avg: 26.78 Index: 1.97 1 Jan 1989 to 1 Jan 2011 I know 2011 is still a few weeks away but this basically gives you the data till today. Smart-ass: Rett, So why not just put today's date? :--D Rett: Good Q, but instead of worrying about putting (and remembering) today's date, I usually put 1 Jan 2011. Smart-ass: So there is no conspiracy involved in this. Rett: absolutely none Avg batting avg of top 6: 54.62 Similar bowling avg: 24.54 Index: 2.23 What does this stupid analysis suggest? If the batsmen of 1928-49 were to bat today, their avg would be: 48 If the batsmen of 89-today were to bat in 28-49, their avg would be: 60 So what would be Bradman's avg now based on your stupid quality index? 100*0.08 = 80 Avg And what do someone like Tendulkar avg in Bradman's times (ofc based on your dumb analysis): 57*1.25= 71.25 Avg So, based on quality index: Now: Bradman 80, Tendulkar 57 Then: Bradman 100, Tendulkar 71 Smart-ass: No 57>100, so that can't be true :hitler: *Rett is awarded the Noble Prize for his contribution and is laughing all the way to the bank* *Smart-ass and his gang think that Rett pulled off the biggest con in the history of cricket data analysis. Mr. don't ridicule me is still working on manipulations to show 57>100* CASE CLOSED

[dial_100]

In my opinion, until I know what is the value of that 99 or 100, Even 50 could be better than 100. I dont give damn about peer performances. But when I see DG Bradman ...b Hollies and see the kind of bowler he was, I dont want to compare the averages of the batsmen from that era to the current. I am wondering that we dont take it as obvious. Just because peers of DGB were not so good to score more than avg of 50-60 then that does not make DGB's 99 worth as much as today's 99. PERIOD and yes CASE CLOSED.

[yoda]

In my opinion, until I know what is the value of that 99 or 100, Even 50 could be better than 100. I dont give damn about peer performances. But when I see DG Bradman ...b Hollies and see the kind of bowler he was, I dont want to compare the averages of the batsmen from that era to the current. I am wondering that we dont take it as obvious. Just because peers of DGB were not so good to score more than avg of 50-60 then that does not make DGB's 99 worth as much as today's 99. PERIOD and yes CASE CLOSED.

So how do you close a case without knowing answer to your main question? May be cause you weren't interested in finding out the answer in the first place?

[Raghav_12]

1 Jan 1928 to 1 Jan 1949: Avg batting avg of top 6 (excluding DB): 52.73 Similar bowling avg: 26.78 Index: 1.97

This analysis is flawed. In period mentioned by you Bradman was part of around 50% of the test played. And most of the time Bradman was top scorer in these matches. Excluding him means that you taking average of bottom 5 players (out of top 6) from the era mentioned by you. For apple to apple comparison, from other era also you should take batting average of top 6 after removing the runs of highest scorers (in other terms batting average of bottom five out of top 6)

[zen]

So how do you close a case without knowing answer to your main question? May be cause you weren't interested in finding out the answer in the first place?

Why bother responding to such hollow arguments :P

[zen]

This analysis is flawed. In period mentioned by you Bradman was part of around 50% of the test played. And most of the time Bradman was top scorer in these matches. Excluding him means that you taking average of bottom 5 players (out of top 6) from the era mentioned by you. For apple to apple comparison' date= from other era also you should take batting average of top 6 after removing the runs of highest scorers (in other terms batting average of bottom five out of top 6)

What's the conspiracy behind excluding Sir Don Bradman's numbers? No conspiracy there. He has been so far ahead that bringing him in to this may not account for general overall top quality.

Bradman's performance would in fact increase the quality of that era, where by pushing the quality index up of his era, which in turn would NOT favor the batsmen of 1989-2010 period! Simply put, the more quality players you take the quality index of that period would get better! By excluding Bradman, this analysis in fact doesn't favor Bradman, which is why (because of the handicap) he is only averaging 80 in 1989-2010 period :winky: PS appears as if you edited your post from adding Bradman's run to removing the top scorer of 89-10. Removing the top scorer of 1989-2010 period doesn't in fact affect the general quality of that era. By now you must have realized that having more quality players increases the quality index so having all top 6 from 89-10 helps that period!

Arre bhai, you once again seem to have missed the point probably because you don't have a clue of what I did. Now don't go about complaining that you are being ridiculed for having missed the point :giggle: First of all, I have clearly mentioned that Bradman's performance is not included because I am measuring the general overall quality of that era.

[Raghav_12]

Arre bhai, you once again seem to have missed the point probably because you don't have a clue of what I did. Now don't go about complaining that you are being ridiculed for having missed the point :giggle: First of all, I have clearly mentioned that Bradman's performance is not included because I am measuring the general overall quality of that era. Bradman's performance would in fact increase the quality of that era, where by pushing the quality index up of his era, which in turn would NOT favor the batsmen of 1989-2010 period! Simply put, the more quality players you take the quality index of that period would get better! By excluding Bradman, this analysis in fact doesn't favor Bradman, which is why (because of the handicap) he is only averaging 80 in 1989-2010 period :winky: PS appears as if you edited your post from adding Bradman's run to removing the top scorer of 89-10. Removing the top scorer of 1989-2010 period doesn't in fact affect the general quality of that era. By now you must have realized that having more quality players increases the quality index so having all top 6 from 89-10 helps that period!

My bad.. I missed that and I do admit that including Bradman's number in your analysis would rather make case stronger for Bradman. But I am not fully convinced with your quality index, particularly with the fact that how to use this quantitatively, though qualitatively you are getting results right and that is Quality of cricket at Bradman's time was inferior to today's time. Now I have a method on how to use these results quantitatively: Quality Index in Bradman's time was 1.73. Take Anti-Log of quality indexes (on the base of 10) Antilog (1.73) = 54 Antilog (2.23) = 170 Comparing those mean Quality of today's cricket is three times (170/54) more than cricket of Bradman's era. So any player who averages 54 today would have scored - 170 in Bradman's era. Vice-versa, if we transport player's from Bradman's era to current era, we'll have to divide their averages by same quotient (roughly about three). So as per this logic Bradman would have score at an average of 100/3 = 33.33. Now in either way it can be seen that 170 (Sachin's) > 100 (Bradman's) or 57 (Sachin's) > 33 (Bradman)

[zen]

My bad.. I missed that and I do admit that including Bradman's number in your analysis would rather make case stronger for Bradman. But I am not fully convinced with your quality index, particularly with the fact that how to use this quantitatively, though qualitatively you are getting results right and that is Quality of cricket at Bradman's time was inferior to today's time. Now I have a method on how to use these results quantitatively: Quality Index in Bradman's time was 1.73. Take Anti-Log of that (on the base of 10) Antilog (1.73) = 54 Antilog (2.23) = 170 Comparing that means Quality of today's cricket is three times (170/54) more than cricket of Bradman's era. So any player who averages 54 today would have scored - 170 in Bradman's era. Vice-versa, if we transport player's from Bradman's era to current era, we'll have to divide their averages by same quotient (roughly about three). So as per this logic Bradman would have score at an average of 100/3 = 33.33. Now in either way it can be seen that 170 (Sachin's) > 100 (Bradman's)

First, let me put your quote properly. In the last post you mixed it with mine's :P From what I read, based on "your" analysis: NOW: Bradman 33, Sachin 57 THEN: Bradman 100, Sachin 170 This is even worse than saying Rameez > Hammond :giggle: :icflove: PS Outsider would be shocked to find that he over estimated Hammond and tried to compare him with Rameez. he should have probably used someone who avg in 20s now for his comparision. :P