Assuming there is a high intent applied to every rep, the acute fatigue will build from the preceding reps leading to slower velocities the further we continue to take a single set. Eventually reaching the point of failure
But how much velocity loss should you be training too?
This guide should give you a rough estimate of how much fatigue you might want to be accumulating - or not accumulating - across a set.
An introduction to velocity loss
Before we worry about the different velocity cut offs you might want to train with, we need to understand how velocity loss, or percentage fatigue is calculated.
As we complete more and more reps in a set, our ability to maintain the same bar speed reduces. We fatigue. This can be given a subjective value such as in RPE training models but with velocity tracking, it's possible to precisely calculate the amount of fatigue we experienced across a single set, relative to our maximum potential. The most common approach with this is to take the difference between your first rep and your last rep, expressed as a percentage drop in fatigue.
Another method (and my preferred approach) is to take the fastest rep of the set, no matter when it occur and calculate the difference from this rep to your last rep. Quite often both methods give the exact same result, when the first rep is the fastest, but it can be quite common to mis groove the first rep, or be a little slow off the ground on say a deadlift so by waiting for the fastest rep, we get a better indication of your maximum velocity for that given set. This also helps prevent artificial fatigue numbers on sets with a low intent.
It's important to note that any drop in velocity assumes and high intent on every rep, without these don't bother with intra-set fatigue tracking.

Applying velocity stop points
Now that we know how velocity loss (% fatigue) is calculated, it is possible to use this velocity loss to manipulate our training, either actively chasing more fatigue and therefore more volume, or reducing the amount of fatigue allowed in order to reduce training volume. These are called stop points or cut-offs and are a brilliant way to to autoregulate the number of reps that an athlete will do, and control proximity to failure.
For example:
"Let's lift 155kg but stop once you reach 18-20% fatigue"
"For this drop set, you need to keep lifting at maximum effort until your velocity drops at least 35%"
So what is the ideal percentage fatigue that you should work with for your training goals?
40% velocity loss - 9.5+ RPE
Grinding into this much of a velocity drop is tough going.
For many exercises with at a reasonable amount of weight, a 40% drop in velocity is often associated with technical failure for many exercises (1).
This type of training is suboptimal for developing strength and power and typically leads to high fatigue accumulation.
We explore the research here but the key takeaway is:
Avoid training up to or beyond 40% fatigue on a regular basis

30% velocity loss - 8-9 RPE
This is a high volume training prescription.
30% is a good number to work around for when hypertrophy is a target, or you are in a high volume accumulation block of training such as an off-season/GPP phase. 30% also works great to control rep output on accessory or secondary movements. This works both ways, motivating more casual lifters to squeeze our more reps than they might self select and holding back hard workers from grinding out harder than might be neccessary.

20% velocity loss - 6.5-8 RPE
20% has a magical aura to it.
Probably due to the amount of research that has used 20% as the "low volume" group in studies, a 20% cut-off has become one of the most common training cut off points. And for great reason, it works.
In fact two great studies (2), (3), looked at 40% velocity loss compared to 20% over a block of training.
The first study (2) measured an exhaustive list of adaptive factors and showed that the 20% group had a significant advantage.
All while doing 40% less total reps across the eight weeks.

Some of the differences after only eight weeks are huge, especially given the 20% fatigue group did 40% less work.
- +7kg on 1RM (19 vs 12kg)
- +3cm on vertical leap (4cm vs 1cm)
- +11% increase in TIIA fibre ratio (the 40% group actually shifted down in TIIA composition)
The second study (3), focused directly on upper body power in a a group of elite kayakers over 12 weeks of bench press and row training.
Again, the result was night and day, with the 40% fatigue group doing almost twice as much work, but basically making zero improvement over the training block.

Taken collectively, it seems restricting velocity losses to only 20% is great cut-off for hitting the gains sweet spot.
Hard enough to create adaptations and force growth, but not so hard that you can't walk up the stairs tomorrow.
Even if they don't use velocity based training many powerlifting and S&C coaches have independently and intuitively worked out that this is the sweet spot for much of your hard training. Multiple work sets at or around an RPE of 8.
This is intuitively smart training, and it takes many forms:
- 8 reps at a 10RM in a volume phase
- A crisp triple with a 5RM load
- Multiple singles with your 3RM

15% velocity loss - 6-7 RPE
We might be splitting hairs now with this level of granularity, but there is some value for the big picture.

F. Pareja-Blanco (4) compared 30% and 15% velocity cut offs in soccer players for their in gym training. both groups saw similar strength gains over the six weeks, while the 15% group had a greater jump improvements.
When the findings from these three studies (2-4) are combined the following rough guidelines could be extrapolated:
- 40% in-set fatigue is the point where the proximity to failure begins to have a negative effect on strength and power adaptations. Avoid training this hard too often
- 15-30% is a optimal zone for regular strength training. This wide range gives plenty of room for volume adjustment over training blocks.
- 15-20% might be more optimal for power adaptation and maintenance of jump performance in season.
10% velocity loss - 5-7 RPE
10% is a strict cut-off, and if used incorrectly or too frequently this low of a volume may have negative effects on your progress. I mean you still have to do some work!
10% can have value when used intermittently in your training. I see two unique, high value applications; power training and deloading.

Fatigue and power training
Power training is about quality.
If you want to jump higher, you need to practice jumping higher, that means intent and output must be pretty maximal. No one developed elite level bounce without some pretty high power exercises.
Fatigue - both acute and chronic - is the enemy of this high quality power development, so multiple short sets are significantly more valuable than longer grinding sets.
One of the best strategies to increase power output, without compromising session density is to use cluster sets.
Tufano and his team highlighted this effect beautifully in 2016 (5):

Controlling fatigue when tapering or deloading
The second way to utilise a 10% cut-off is during a taper.
In tapering the goal should be to maintain as much intensity as possible while restricting volume, creating the space for supercompensation and peaking to occur.
A 10-14 day reduction in volume seems to be the gold standard (6), and there aren't many easier ways to implement this than by tightening up the in set fatigue.
Keep hitting the same weights but instead of working in the 20-30% range tighten up to 10%.
- RPE 8 becomes 6.5
- Sets of 5-6 become 2-3
Try not to mess with many other variables, keep rest periods and session lengths the same, these sessions should leave you feeling primed and fresh, not beaten and tired.
The big takeaways when using velocity loss
- Train hard, but also train smart, more sets are better than squeezing every drop out of every rep
- Grinding reps aren't as valuable as they are exhausting.
- RPE 8 is golden. Leave two in the tank
- Power is incredibly sensitive to fatigue, when in doubt do less, but do it well.
And one final note, velocity loss has limited application with heavy strength sets of three reps or less, the fatigue can jump up quite quickly on these sets, you might go from the second rep at a comfortable 15% fatigue and then the third spikes wight up to 30% or higher. In these situations use fatigue as a fall back, but for the lifter in the middle of that set subjective RPE systems will be more practical than the velocity data after the rep is already finished.
Record velocity loss with MetricVBT
I am part of the team that is building MetricVBT a brand new velocity based training technology that is now available. Metric is a computer vision iPhone app that tracks velocity, fatigue and range of motion in the gym. It is available for free on the App Store, you can download it from the Metric website at this link.
References and resources
- Jovanovic M, and Flanagan EP. 2014, Researched applications of velocity based strength training.
- F. Pareja-Blanco et al, 2016, Effects of velocity loss during resistance training on athletic performance, strength gains, and muscle adaptations
- Izquierdo-Gabarren et al. 2010, Concurrent endurance and strength training not to failure optimizes performance gains.
- F. Pareja-Blanco et al, 2016, Effects of Velocity Loss During Resistance Training on Performance in Professional Soccer Players
- Tufano, JJ, et al. 2016, Maintenance of Velocity and Power With Cluster Sets During High-Volume Back Squats
- Laurent Bosquet, 2007, Effects of Tapering on Performance:A Meta-Analysis
- Weakley, J, et al, 2019. The Effects of 10%, 20%, and 30% Velocity Loss Thresholds on Kinetic, Kinematic, and Repetition Characteristics During the Barbell Back Squat.
- Rodriguez Rosell D, 2018, Relationship Between Velocity Loss and Repetitions in Reserve in the Bench Press and Back Squat Exercises