Enhanced cluster set training with velocity based training

Clusters are a great way to increase workout density without sacrificing the quality of your output. Here is how VBT can be used to program clusters.

One of the constant battles in training is the opposing forces in time management in the gym.

Rest longer to get more out of each working set to maintain high quality of strength or power output.

But also pack enough into a session without it becoming so jam packed that it just becomes a cardio circuit.

Time is almost always the biggest limitation for both athletes and the general training population so finding strategies to fit enough high quality training in your week is a constant battle,

The good news is there is a solution.

One of the best ways to minimise fatigue in your training without letting the rest periods balloon out is to use cluster sets.⁠⁠

Cluster sets reimagine strength and power training

Traditional programming guidelines suggest a few sets, with a few or many repetitions with long rest periods in-between.

  • 3x8 with 90 seconds rest
  • 4x6 with 3 minutes rest
  • 5x5 with 5 minutes rest

Clusters turns classic programming on its head. Instead of doing a few sets with long rests, cluster training takes those longer sets and rest periods chops them up into many submaximal sets with more frequent short rests between them.

Illustrative example of Cluster sets
An example cluster set approach vs a traditional 3x5 protocol.

Clusters enable higher quality

When they are programmed correctly, cluster sets are a significantly more efficient and effective way to train (1-2), allowing you to :⁠⁠

  • Lift the same weight⁠⁠s
  • Perform the same number of reps⁠⁠
  • With the same total rest ⁠⁠
  • With less accumulated fatigue
  • All while achieving a higher average velocity and power across the session!⁠
Tufano et als findings of velocity across cluster sets
Cluster set velocity compared to classic set rep schemes. Adapted from Tufano et al 2016 (1)

Tufano et al looked at proving these benefits in 2016 by comparing two different cluster protocols with a traditional 3x8 training session. Both the 4x3.4 and 3x6.2 cluster approaches maintainedsignificantly higher velocities across the same training load and total number of reps than the traditional set-rep protocol.

Both the four rep and two rep cluster methods saw only 1-5% velocity loss for the entire working series, compared to the ~24% fatigue seen in the traditional sets.

Tips for designing the perfect clusters

How to write clusters

Clusters break the standard sets-times-reps model, because each cluster is effectively a set within a set so you will need a new consistent nomenclature, especially if you are programming clusters online. I like the following:

3*(4*2) 30/120 rest

3*4*2(30/120) means would be three sets each with four clusters, each cluster has two reps, with 30 seconds rest between each cluster and 120 seconds rest between the sets.

This is by no means a hard and fast law, you might find a better way to explain your clusters, if so let me know!

Logistical factors to consider in cluster training

Set up and dismount time will factor in hugely, eating away at your energy and rest time more than you think. Deadlifts, jumps and rack cleans are gold, Barbell bench is good too. Back and front squats on the other hand are tricky and dumbbell bench can be a downright nightmare for cluster training.

As a rule,the trickier the set up, the less cluster-friendly an exercise will be. There is more risk every time you have to restart the movement (think walking the bar out of the squat rack) and more chance of a wonky preparation leading to suboptimal reps. Ideally avoid these exercises altogether, but if you have too stick with 4+ reps per cluster if there is a level of set up involved and don't include the set up time in the rest period.

Spend the time getting your training environment right before jumping in. Take a few seconds to get everything just right before you start that first cluster, the rests are shorter than you think and you need to be using that time to recover, not scamper off to find the chalk. Put your stopwatch/phone in an easy spot to hit stop and start and if you are going to need water mid cluster, keep it handy.

Always make sure you give your clusters a test run before programming it for an athlete. 10x2x1 with 20 seconds rest might look good on paper, but you won't know until you reach the 7th set, so give it a test run first if you are trying something brand new.

The point is to reduce fatigue, not increase it. Clusters are designed to increase the density of your training without a big hit to the quality of your output. Don't hesitate to increase the rest periods or play with the variables in real-time should you need it, breaking longer clusters into more shorter strings.

Monitoring your velocity and power is key to managing this fatigue and ensuring high quality training is maintained throughout the clusters. You could set a velocity stop limit or use a fatigue cut off for each string of clusters that would trigger a longer rest period if needed.

Be wary of going overboard on total volume. Even when programming clusters, total training volume and the amount of high intensity work needs to be considered. I really like Dan John's Rule of ten (working reps) as a starting point.

My favourite cluster protocols to try

There are plenty of ways to manipulate and program cluster sets, the only real limitation is your imagination just be sure to factor in the rest periods for your total session duration because 10 seconds here or there may not be an issue if you are only doing three sets, but when you are suddenly doing twelve clusters, those extra ten second rests will add up fast.

Here are a few examples compared to a classic 3x8 structure:

A table of example cluster set protocols
Some example cluster set training protocols

Learn VBT.
Dominate the gym.

Join the VBTcoach community, and unlock a stack of velocity training resources immediately.

Perfect for committed lifters and coaches alike.

Get the free resources