Prediction Success Rate and Correlation

Some of you asked for it, so here it comes: a summary analysis of’s success rate.

Success Rate

The following table shows the last 14 full weeks, with counts and success rates for probabilities 70% and over, 75% and over and 80% and over respectively. This might sound a bit complicated but we’ll take a deeper look at it soon.

So, let’s dig into the numbers.

What is success rate?

Success rate is the rate of successful tips per all tips: Success rate = number of successful tips / number of all tips.

Let’s see some examples from the table.

  • In week 25, we predicted 308 tennis matches at or above 70% probability. 81.8% of those won.
  • In week 27, we predicted 136 matches at or above 75% probability and 86.8% of them were a winner.
  • In week 29, we predicted 56 matches at or above 80% probability and 92.9% of them won.


These numbers are good on their own, but an even better thing is correlation.

Correlation is one of the most useful statistics. Without going deep into details, it basically describes the degree of relationship between two variables.

In our case, the two variables are the predicted probability and the actual success rate. It can be best observed on ourĀ Accuracy page, but the following snippet from the table above shows similar things:

It’s easy to discover the correlation: the higher predicted probability results in better success rate:

  • 70% and over: 81.8% success rate
  • 75% and over: 86.4% success rate
  • 80% and over: 92.9% success rate

As said, more about accuracy and correlation can be found on ourĀ Accuracy page.

I hope it helped to understand how extremely important probabilities and correlation are. Simply put, if you don’t find correlation at any tipping service or tipster, think again.