Now well into the knockout stages in the Round-of-16 is the FIFA World Cup 2014. Let us try to predict who will win each of the remaining matches. It is easy to fall into the trap to claim that Brazil or Germany or Argentina is the team to beat. This type of prediction is based on our ‘favouritism’, and not on actual evaluation of the strengths, weakness, or how well the team has played in the Finals.
Based on such favouritism, Brazil should have won the Brazil vs. Chile with ease. But what we saw was a ding-dong battle. So why did that happen? Has Chile been underrated? Or is our cockeyed favourite-playing making us see Brazil as more formidable than they are?
To predict the winners in the knockout stages (in the Round of 16), let us stitch up a formula in, what else, but MS Excel. In order to predict the winner of each match, we need to do a type of ‘adjustment’, which is called normalization. This is essential if we have to compare the performance of Brazil or Germany with that of Belgium, Greece or Algeria.
If that is so confusing, let us think back to India’s favourite sports - Cricket. A Sachin Tendulkar century against Bangladesh in India, one against Australia in India, and another against Australia in Australia, are not rated the same. With Cricket, we inherently understand the relative strengths, and instinctively adjust or ‘normalise’ our valuation. The one against Bangladesh is rated low, while the ones against the ozzies are treated as high class.
Something similar has to be done with the 16 teams, taking into account their performance against opponents in their own group, and also within the 32 teams overall. Then we can make some predictions. And it is an eye-opener. The Netherlands top the index, the Swiss and USA bring up the rear, with Brazil floating in the middle.
The method to calculate such normalization is well documented. The Americans with their obsession with baseball and college (American) football have almost nailed the rating system. The page at [http://football.kislanko.com/normalizing.html] is a good start, but also Google for ‘Ratings Power Index’.
Mind you, we have not used the exact formula, but mangled it, cannibalizing only the basic idea. That is take the Goal Difference, the FIFA Ratings, the opponents FIFA Ratings, and multiply and divide as given in there.
Here is the spreadsheet which has been correct true in the 4 matches of FIFA World Cup 2014, Round of 16 till yesterday. By this Index, France, Germany, Argentina, and Belgium should get through.
Just a point though. This formula does not take into account the home conditions for Brazil (and South American countries in general), the climate or heat adversity for Europeans, and the brilliance of individual skills which can be inspired in a match.
Cheers and Ole!
Based on such favouritism, Brazil should have won the Brazil vs. Chile with ease. But what we saw was a ding-dong battle. So why did that happen? Has Chile been underrated? Or is our cockeyed favourite-playing making us see Brazil as more formidable than they are?
To predict the winners in the knockout stages (in the Round of 16), let us stitch up a formula in, what else, but MS Excel. In order to predict the winner of each match, we need to do a type of ‘adjustment’, which is called normalization. This is essential if we have to compare the performance of Brazil or Germany with that of Belgium, Greece or Algeria.
If that is so confusing, let us think back to India’s favourite sports - Cricket. A Sachin Tendulkar century against Bangladesh in India, one against Australia in India, and another against Australia in Australia, are not rated the same. With Cricket, we inherently understand the relative strengths, and instinctively adjust or ‘normalise’ our valuation. The one against Bangladesh is rated low, while the ones against the ozzies are treated as high class.
Something similar has to be done with the 16 teams, taking into account their performance against opponents in their own group, and also within the 32 teams overall. Then we can make some predictions. And it is an eye-opener. The Netherlands top the index, the Swiss and USA bring up the rear, with Brazil floating in the middle.
The method to calculate such normalization is well documented. The Americans with their obsession with baseball and college (American) football have almost nailed the rating system. The page at [http://football.kislanko.com/normalizing.html] is a good start, but also Google for ‘Ratings Power Index’.
Mind you, we have not used the exact formula, but mangled it, cannibalizing only the basic idea. That is take the Goal Difference, the FIFA Ratings, the opponents FIFA Ratings, and multiply and divide as given in there.
Here is the spreadsheet which has been correct true in the 4 matches of FIFA World Cup 2014, Round of 16 till yesterday. By this Index, France, Germany, Argentina, and Belgium should get through.
Just a point though. This formula does not take into account the home conditions for Brazil (and South American countries in general), the climate or heat adversity for Europeans, and the brilliance of individual skills which can be inspired in a match.
Cheers and Ole!
No comments:
Post a Comment