AI Horse Racing Tips

About Our Tips

We make our tips using well established machine learning, or artificial intelligence, techniques. The concept is to develop a model representing a pattern in the data of past results that can be applied to future races. The tips produced on this site are the on-going product of over six years analysis and investigation.

It would of course be impossible to predict the winner of every race – horses are not robots, but we believe a well-developed machine learning technique can pick winners as well as, if not better than, a professional tipster. One approach is to use an Artificial Neural Network.

To successfully profit from betting on horse races we need to combine our selections with a Staking Plan. Intuitively we would look to stake more on a lower priced than a long odds horse. For more information take a look at our Staking Plan advice.

Artificial Neural Networks

How artificial neural networks work

The building block of a neural network is an artificial neuron. It can be thought of as a simplified model of a biological neuron in the brain. The concept is straight-forward – inputs to the artificial neuron are combined and processed to produce its output
Artificial Neuron
The processing power of artificial neurons comes in combining them to form a network- an Artificial Neural Network (ANN). The output of one neuron is the input to the next. In a fully connected network each node in each layer provides inputs to all the nodes in the next.
Artificial Neuron

Our Progress

Every day our computers number crunch through hundreds of variables and test the results with different Neural Networks. The process is constantly being refined and improved with aim of developing a profitable system when followed over a long period.

We publish the results of all our tips the following day. Take a look at our results page for our complete tipping history. Our tips are independently verified at racing-index.com.