Training hard is a necessity if you want to maximize results from your training program. This is often misconstrued as the need to train to failure on every set of every movement. While this may lead to some results when an athlete first begins training, is this the optimal way to train when we’re trying to maximize throwing velocity?

Adaptations Needed for High Velocity Throwing

Let’s first discuss the adaptations that are necessary to help improve an athlete’s throwing velocity. We can divide these adaptations into two categories: neurological and physiological.

  • Coordination: Basically, this means that muscle force is applied at the right time, in the right direction, in the right sequence, etc. This tends to be specific to the mode of training chosen (e.g. grinding reps vs high velocity sets)
  • Motor Unit Recruitment: Low threshold motor units are recruited first due to the size principle. These are comprised mainly of type I slow twitch muscle fibers which are resistant to fatigue, but are not capable of very high outputs. High threshold motor units are comprised mainly of type II or fast twitch muscle fibers. These are only called upon when the CNS determines that their assistance is required. This could be when heavy loads are being lifted, when light loads are being lifted in close proximity to failure, or when maximal velocity is intended. Repeated exposure to maximal effort activities can help make more high threshold motor units available that were not available before.
  • Rate Coding: Rate coding is the frequency at which motor units discharge action potentials to activate motor units. Increasing the rate at which motor units are activated increases the potential for overall force output and force output in shorter time frames. High-velocity movements have been shown to increase rate coding to a greater degree than heavy loading (Van Cutsem et al., 1998).
  • Activation Level of Antagonist Muscles: Antagonists are the muscles that oppose the motion you’re trying to create (e.g. the triceps are the antagonist to the biceps as they extend the elbow while the biceps create elbow flexion). Antagonist coactivation is important for joint stability, but if antagonist activation is too great it will limit the net torque produced by the agonist.
  • Fascicle Length: Longer fascicle length is important as it is associated with faster contraction velocities.
  • Fiber Type Shifts: Within the fast twitch or type II muscle fibers there are two important subcategories: type IIa and type IIx. Type IIx fibers have the fastest shortening velocities (~5-6 fiber lengths/second), while type IIa are still very fast (~3-4 fiber lengths/second), they are a bit slower than IIx (Beardsley, 2021). But, both types are significantly faster than type I fibers (0.5-1.0 fiber lengths/second) (Beardsley, 2021). Depending on the type of training performed, fibers can shift their type from IIa to IIx, and vice versa. The more fatigue that is present in a training program, the more likely a shift is to occur from type IIx to type IIa.
  • Stiffness and Connective Tissue Adaptations: In order to increase the amount of force that is transferred from a muscular contraction and increase the amount of elastic energy that can be stored and released, tendon stiffness often must be improved.
  • Muscle Pennation Angle: The greater the pennation angle, the fewer sarcomeres in series, but the greater the number in parallel. This means that a greater, or more obtuse, pennation angle means that the muscle is capable of generating higher forces, but lower velocities.

For a more in-depth discussion about these specific adaptations check out this article where I dive deeply into this topic.

What Happens When You Train to Failure

Training to failure, or close to the proximity of failure, has long been popular in the bodybuilding and “gym bro” world as it is assumed that this maximizes hypertrophy. While training to failure does have implications related to hypertrophy, we need to dig a bit deeper than this when training for increasing velocity. The need for reps close to failure is more important when it comes to training with light loads. This is because one of the main factors when it comes to hypertrophy is mechanical loading. This means how significantly individual muscle fibers are activated and utilized during a set, as well as, what capacity these fibers have for growth. When a light load is lifted, low threshold motor units, which tend to be primarily composed of type I (slow twitch) muscle fibers, are activated. This is because it does not take much force to move these loads. The problem with this approach is that in order to maximize muscle growth high threshold motor units are important as these are primarily composed of type II (fast twitch) muscle fibers which have a greater capacity for hypertrophy. When a heavy load is lifted on the other hand (generally greater than 85% of 1RM), high threshold motor units are immediately activated. This helps reduce the need for training to failure as early in the set we are getting a lot of mechanical loading.

Implications of Training to Failure

Training to failure leads to some results that are important to understand when evaluating its fit in a program.


Designing a training plan ultimately comes down to balancing the fatigue that comes from a specific training session or series of sessions, with the ability to recover and adapt to the training. When training to failure, fatigue is increased considerably along with soreness, and this fatigue tends to be longer lasting even when volume is equated (Fonseca et al., 2020). This has important implications for the other training that will take place over the course of the next few days and beyond. When it comes to training athletes the most important piece of their training program is sport-specific practice. In the case of pitchers this means a relatively high volume of throwing and throwing-related drills. If the quality of these sessions is impacted negatively by what an athlete does in the gym, then it’s obvious that the weight room side of the training plan must be adjusted. In order to maximize the effect of sport-specific training, maximum quality must be achieved. This means extreme focus on every rep and low fatigue so that the output during high-intensity sessions can be maximized in order to drive maximal adaptation.

This can have even more serious implications in-season. Due to the frequency of games, in order for performance to be maintained throughout the season various fitness qualities may need to be touched-up in minimal retention doses. This is even more critical for relievers that pitch more frequently and may need to microdose training to focus on specific fitness qualities. In order for this approach to work, fatigue cannot be carried after the session. The fitness quality should be touched with a minimal number of high-intensity sets that do not approach failure. The closer the athlete gets to failure, the more negative impact these training sessions are likely to have on on-field performance.


One of the positives of training to failure is that it does tend to result in more hypertrophy (if you’re training with light or moderate loads), and greater muscle cross sectional area (CSA) (i.e. hypertrophy) has been shown to be associated with greater strength.

However, there are also some negative effects of hypertrophy. Tissue inertia is an important factor to consider. Since the muscle is larger, it takes longer for the muscle to shorten and create movement at the joint (Gunther et al., 2012). This means that in order to create the same joint angular velocity as before, the muscle contraction would need to be even faster since there is more to shorten. This can have obvious negative effects on a time-constrained movement, such as pitching, where contraction velocity is extremely important.


Stiffness, in terms of the ability to resist deformation is an important quality for connective tissue and has implications for force transfer and movement velocity. Heavy resistance training and high-velocity movements have been shown to increase stiffness (Bohm et al., 2015), but high repetition training with light loads has not shown a similar positive effect.

Muscle Fiber Type Shifts

As discussed above, if the goal is to maximize movement velocity then an athlete must maintain the highest possible number of type IIx muscle fibers. The more fatigue that occurs during a set, the greater shift there is from type IIx to type IIa fibers. This is definitely a negative for pitchers and is made worse by training to failure.

What To Do Instead

So, what should you do instead of training to failure? There are a few options that can improve your results.

Sets with a Specific Velocity Loss

The first option is the most optimal as the feedback is most objective. Training with a specific velocity loss involves planning a movement with a specific adaptation in mind. Just as you would use percentages of one-repetition maximums (1RM) when planning a session, you can do the same with movement velocity. For example, when training speed-strength you would use a load in the 40-60% range. When training speed-strength and utilizing velocity based training, you would use a load that you can move in the 1.0-1.3m/s range (Mann, 2016). In this scenario you would train with the highest possible load for the velocity desired, and when a predetermined velocity drop-off is hit, either the set, the movement, or the session is terminated. Here’s how a speed-strength session may look:

Band Resisted Safety Bar Squats @1.3 m/s with a 5% velocity drop-off:

  • As many sets of 3 as possible with a hard stop at 10 sets if a drop-off does not occur
  • If any rep drops below 1.24 m/s this movement is terminated

In this scenario, training volume is being autoregulated which keeps intensity high and allows an athlete to maximally develop a specific quality as we’re able to give them exactly as much as they need without adding detrimental fatigue.

Reps in Reserve

The second option is still a great option if you do not have access to velocity tracking tools. With reps in reserve athletes are still able to autoregulate training based on how they’re feeling on a given day, and the specific adaptation is still able to be trained without fatigue being added.

Here’s how a high force session using reps in reserve (RIR) might look:

Trap Bar Deadlift

5×3 @3RIR

The athlete will select a load that can be lifted 6 times (their 6RM on this day), but each set will only be three reps. This allows athletes to handle very heavy loads (~85% in this case) while minimizing excess fatigue. This also allows athletes to base the load on how they’re feeling on a given day, rather than rigidly sticking to pre-planned percentages which may be too heavy or light on a given day.

Cluster Sets

The third option is my personal favorite for maximizing hypertrophy gains without sacrificing as many type IIx fibers as typical hypertrophy training. This option breaks up higher rep sets into smaller clusters allowing movement velocity to stay high, heavier loads to be handled, and more type IIx fibers to be preserved.

A cluster set aimed at higher load hypertrophy may look like this:

Front Squat

4×4, 4 @80%

In this case an athlete would perform 4 front squats, rest 15-30 seconds, then perform the next 4 reps. This will still be a relatively challenging set, but compared to a set of 8 straight reps at 80%, which is about an 8RM, the movement velocity will be higher. Cluster sets can be split up in many different ways and the above layout is just one option.


When training to increase throwing velocity we should have one of the following broad goals: move light loads very fast or move heavy loads with maximal intent. On the light load side of things, we’re trying to move weights very fast in order to obtain specific high velocity adaptations which require low fatigue. If you’re training to failure or close to failure with this method you are no longer achieving that goal. If you’re training to improve maximal strength then you’re likely using heavy loads at or above 85% of 1RM, which means that motor unit recruitment is already maximal. Training to failure in this case does not increase motor unit recruitment any further. While no training method is inherently “good” or “bad” training to failure is a poor choice for athletes trying to maximize throwing velocity.