Predicting Results.

With the almighty ASHES starting from today or Wimbledon "the hallowed ground of sports" as they call it, is in its final stages,Lets begin this discussion with couple of questions :
"Who will win the Ashes this time ? or "Who will be crowned as champion of Wimbledon?".

These are the classic examples of of predicting the results with the help of analytics. Many people can intuitively say that Mighty Aussies will thrash the Englishmen this time. Or, Novak with his current form will he champion undoubtedly,. If they think so,then they have not done their research. These intuitions may or may not be right. This is when data science comes into picture. 

The results or outcomes predicted using data science may not match the exact outcome all the time, but the probability of getting the predicted result is always high. After all it is all about the "probability". isn't it ?

A data scientist will start his process by gathering data related to the event occurring. Now, for Ashes & Wimbledon, we have a large data to handle. But,the data scientist should be smart enough to pick appropriate data. Lets see what questions we as a data scientist should ask before looking out for the data.

For ASHES :

Where will all the matches be held ?

What is the record of home team and visiting team on those venues?

what was the result When two teams met last time on that venue?

What was the result in the previous Ashes series ?
Who all players are included in the squad and what are their records when playing against each other?

...

You can form as many questions as you want to extract all the meaningful data that will help you predicting the outcome.

After you extract the data then it is all about the statistical methods to predict the outcome of an event. 

I will discuss these methods as we go along. But, for now, I hope you have got hands on with what data scientists do. Framing questions to extract data is very important part in predicting the outcome of an event. 

After reading this example, if you really feel that data science is for you then get going. You will be the one predicting all the results.

Similarly, when a data scientist works for an organisation he has to extract all the data related to the business problem. Believe me the business problem is more interesting than the example listed here.