The goal of this project is to create statistical data that can help the bookie make the right decisions.
All the data is taken from http://tennis-data.co.uk/data.php. We have all the tennis matches played in all tournaments (Grand Slams, Masters, Master 1000, ATP500, ATP250, International, International Gold, and Masters Cup) since January 2005 until November 2022.
There is a lot of information in this dataset. In this project, we will focus only on the analytical study of the betting market.
It can help the bookie make the right decisions before placing a bet on a tennis match.Let's take a look at the dataset.
There are many variables in our dataset, but what really matters to us are the columns that contain the odds set by the bookmaker.Let's take an example:If we bet $10 on a player with odds of 1.30:
- If the player wins the match, you will go back home with your $10, plus a $3 profit.
- If the player loses the match, you will go back home with $0.
Before we start the project, you need to understand the following:
- The bookmaker sets the odds based on statistical studies that enable them to predict the winner of the match.
- The player with the lowest odds has a higher probability of winning the match.
For example, consider two players with odds: player_1: 1.286 and player_2: 3.25. This means that player_1 has a probability of (1/1.286)*100 = 77.8% to win the match, while player_2 has a probability of (1/3.25)*100 = 30.8% to win the match.
We will quickly notice that the sum of the probabilities is higher than 100%, which means that the bookmaker keeps a profit margin for themselves. Below, we will display a chart representing the accurate prediction of the winner by the bookmaker over time.
We can conclude that in recent years the bookmaker has started to make more and more mistakes in his prediction of the winner, as the percentage has started to fall below 70%.
Let's take a look to the data distribution in our dataset
We can see that ATP250 has the highest number in terms of the number of matches played during these years.
• player ranking's impact on bookmaker decisions
Let's see how much the bookmaker is affected by the player's ranking when he puts odds.
We can conclude that the bookmaker is not entirely affected by the player's ranking before placing the odds, Where 80.66% of his predictions are directly related to the player's ranking .
Let's now see the ratio of winner prediction by the bookmaker when his prediction is related to the player's ranking
'Matching' means that the bookmaker expects that the player who has the highest ranking will win the match while 'Not Matching' means the opposite.
This means that if we placed a bet on a player with the smallest odds and this player has a higher ranking, then we have a 72.4% chance of winning the bet.
While if we placed a bet on a player with the smallest odds and this player has a lower ranking, then we have a 61.7% chance of winning the bet.
And this time, we will show you the success rates of the bookmaker in different series such as Grand Slam, ATP500, and International Gold.
The table below summarizes the performance of the bookmaker on all series
We can easily notice that the bookmaker have the highest performance on Grand Slam series.