Strength: training to failure – yes or no?

There are many points of contention when it comes to the optimal way to develop muscle strength and growth (hypertrophy). From free weights versus machines, to single-joint versus multi-joint lifts, to varying repetitions speeds and ranges, the debate rages. However, no topic is more controversial than whether individuals should or should not train to failure (the fatigue level at which it is physically impossible to perform another rep) in order to optimize progress.

Prior research

Training to failure is a topic that Sports Performance Bulletin has addressed before. A recent article examined a study that looked into this very question. The researchers in this study had subjects perform one of three different training protocols⁽¹⁾. The subjects either performed three sets to failure, four sets not to failure (with the extra set to match the total reps volume of the failure group), or three sets not to failure. After five and ten weeks, there were no differences between groups. They all were equally effective. In fact, a measure of power output was improved in the groups that did NOT train to failure, a finding absent in the group training to failure. Avoiding failure appeared therefore to be advantageous.

A second SPB article looked into this topic further. The researchers had the subjects perform one of two types of bench pressing sessions⁽²⁾. In one session, the subjects performed five sets to failure. In the other session, the subjects performed four sets with three repetitions in reserve – ie they stopped 3 reps short of failure. On the final set, they performed the set to failure. The results indicated that keeping reps in reserve allowed for the subjects to perform more total work across the five sets.  Staying away from failure allowed for more quality repetitions, exactly what an athlete wants!

Controversy

While these studies provide convincing evidence that training to failure isn’t required for optimal progress, evidence does exist to refute these claims. There appear to be specific circumstances where training to failure is potentially superior to avoiding failure.  A recent study had subjects perform one of four training protocols over the course of the 8-week study⁽³⁾. The participants performed sets with light loads or heavy loads, and they performed sets to failure or not to failure. 

The results showed that hypertrophy increased when using heavy loads regardless of training to failure, and when using light loads and training to failure. However, no changes were found when using light loads and avoiding failure. The conclusion is that if training with light loads, exercising to failure will likely be required to maximize improvement.

Moreover, there is evidence that maximal muscle activation is only achieved by training to failure in individuals with a significant strength training background⁽⁴⁾.  This is true regardless of whether the load is light or heavy. This effect is not seen in untrained individuals. As maximal muscle activation is thought to be necessary for maximal adaptation, occasionally training to failure may therefore be necessary for advanced trainees.

Definitive answers

With these somewhat conflicting and contradictory findings, it’s natural to ask which approach is best. A group of researchers recently sought to answer this question by performing a meta-analysis that examined all of the previous studies examining training to failure and pooling the data⁽⁵⁾. They also looked at studies that investigated the impact of velocity loss thresholds. This involves tracking the speed of the bar/dumbbell etc throughout the course of a set. As fatigue accumulates, the bar speed will slow. The more it slows, the closer one is to failure. Studies often implement various ‘velocity loss thresholds’ to determine when the set is complete. For instance, it the set stops when velocity drops by 40% versus 20%, that would indicate greater muscle fatigue and closer proximity to failure.

A total of 15 studies were included in the meta-analysis. After running the numbers, the researchers found there was only a trivial benefit of training to failure as compared to not training to failure. They concluded that there is no evidence to support the superiority of training to failure. They also concluded using higher velocity loss thresholds, indicating greater levels of fatigue, do not always lead to larger increases in muscle hypertrophy (see figure 1).

This led the researchers to conclude the relationship between proximity to failure and improvements in muscle mass may be non-linear. While initial increases in proximity to failure lead to large increases in hypertrophy, getting closer and closer to failure may not yield the same benefits. In other words, one needs to train hard enough and get close enough to failure to stimulate progress. However, training harder than that does not necessarily lead to improved hypertrophy gains.

Figure 1: Strength effects as failure point is approached