What proportion of vaccine trials are successful usually? Could the probability of efficacy be estimated and used to judge if the study goes ahead? There needs

What proportion of vaccine trials are successful usually? Could the probability of efficacy be estimated and used to judge if the study goes ahead? There needs to be a reasonable amount of risk taken in research, otherwise we will find it impossible to make progress.