In this example, network node A routes packets to node B in order to reach node C. The links between the nodes are distinct point-to-point links.
According to the split-horizon rule, node A does not advertise its route for C (namely A to B to C) back to B. On the surface, this seems redundant since B will never route via node A because the route costs more than the direct route from B to C. However, if the link between B and C goes down, and B had received a route from A, B could end up using that route via A. A would send the packet right back to B, creating a loop. With the split-horizon rule in place, this particular loop scenario cannot happen, improving convergence time in complex,highly-redundant environments.
There are scenarios where the split-horizon rule may fail. Suppose 6 nodes(A,B,C,D,E,F) are each connected to every other node. The lowest cost routes will establish over time. Now if one of the nodes fails. Split-horizon only protects each node from where their current path is received from. But the same information received from the other nodes will be transmitted as well.
If A connects to E via B and C. And if D connects to E via A->B->C. Suppose B-C fail. A will not advertise its route to E to B directly due to the split-horizon rule. However, as each node is connected to each other. D will advertise the route to E to B (D->A->B->C-E). The split horizon rule stops D from advertising the route to A. Not to B.
Such scenarios and multiple links exist in practical networks. And it is the network administrator who needs to work out such intricacies.