Genetic evidence that relative synaptic efficacy biases the outcome of synaptic competition