A Powerlifter’s Guide to Velocity Based Training Pt. 2: The Case for VBT

TL;DR

• This is a conceptual piece on how velocity based training fits into powerlifting. This is not a collection of tips and tricks. Those practical tips will be in part 3. 

• Velocity based training can give fairly comprehensive feedback in terms of intensity, fatigue, exertion, and readiness to train.

• Velocity based training will only work for you if auto-regulation works for you. Auto-regulation may not work for you and/or velocity based training may not work for you.

• Rather than buying into VBT first, it’s probably better to see if auto-regulatory methods like AMRAPs driving training maxes or the RPE scale works for you. It isn’t the same thing, but aspects of both tie back to VBT.

• There are gaps in which VBT is not appropriate. Where these gaps may exist, it may be appropriate to use different methods. These include full range tempo reps, reps with multiple concentric and eccentric points, and movements with a high balance component.

Update 2/28/2017: An eccentric VBT article came to my attention. It's not of consequence, but it is worth mentioning. Additions have been made with no changes to other content.

Update 8/17/2017:  I added some content since this article gets a good amount of hits

INTRODUCTION

Force-velocity curve borrowed from Training Principles
 for Power by Haff and Nimphius from this link
from the Strength and Conditioning Journal.

               Velocity based training (VBT) has been around for a long time. Despite that, we haven’t tried to generalize the implications of exercise velocity as many readily do now. Part of that issue is velocity is very specific in a lot of ways, so we can only speak of it in terms that are general in broad strokes but still specific enough for to the athletes we’re working with – and then tack on a bunch of caveats as appropriate for skill level, height, etc.

               VBT is an auto-regulatory style of training that attempts to change variables of training as appropriate for individual athletes according to their readiness to train. The underlying assumption is that force and velocity share a relationship and that different aims of training can be quantified according to ranges of training. I hate starting here since everyone does (I have in the past and hate myself for it), but I feel like I have to. This graph on the left shows what I mean.

               You could also superimpose the power curve over this, but I’d also be obliged to parrot the mantra that powerlifting has little to do with power. Back on target: if powerlifting is all about force output, then we can keep tabs on Force by essentially keeping track of velocity. There have been some generalized charts that every VBT device manufacturer tends to circulate (myself included) showing the F-V curve and defining arbitrary ranges as absolute strength zones, strength-speed zones, etc. These are generalized. There’s a world of difference of what’s bottom end velocity (closer to 1RM and strength) for pulls from the floor and a pull from above. If you go by the orthodoxy, your overhead pull 1RM might never be slow enough to be considered within the absolute strength zone. It’s an easy way to explain VBT, but not necessarily accurate. Here are the generalized training zones of velocity:

Training Quality	Velocity (m/s)	%1RM
Absolute Strength	<0.5	80-100%
Accelerative Strength	0.5 - 0.75	65-80%
Strength-Speed	0.75 - 1.0	45-65%
Speed-Strength	1 - 1.3	25-45%
Starting Strength	>1.3 m/s	0-25%

               I disagree with this chart simply because it doesn’t take into account the floor (minimum velocity threshold) or ceiling of different velocities and usually isn’t accompanied by an acknowledgment that individual experiences may vary. Every load-velocity profile I’ve generated for myself diverges from this chart in at least one training quality. Additionally, enough movements diverge from the category I concern myself with the most: absolute strength. We need to either accept training quality zones in broad guidelines or understand it’s different for everyone and every movement. Strength is specific. Velocity can be used to measure strength. Velocity is specific.

Notice that hypertrophy is not listed in this. Hypertrophy could be considered an ancillary effect to training in the mid to lower velocity/moderate to high intensity. Understanding that this table is a basic principle, we can move on though. We won't ignore hypertrophy as an essential training effect to progress absolute strength, but the fact that something doesn't fit neatly into this table doesn't matter.

               So let’s talk about the general ways we can use velocity, ignoring the table because the table fucking sucks. In the most general terms, a light load moves faster than a heavy load, showing intensity. The first rep in the tank moves faster than the last rep in the tank, showing exertion across a set. And drops in velocity from the first rep in a set across multiple sets at the same weight can show fatigue across a session. Additionally, losing form, tightness, and regaining balance (like getting out of position) tends to be reflected in velocity as a loss. That variance in velocity behavior is essentially variance in performance of the movement. There are caveats to many of these, but that’s generally what VBT does. It quantifies intensity, fatigue, exertion, and to some extent also proficiency. With a good load-velocity profile of the movements you plan to use, you can effectively gauge intensity, fatigue, and exertion - for the most part. Unlike 1RM and multi-rep maxes, this can be used across different time domains for feedback from immediate to trends in the broader mesocycle.

Velocity based training quantifies intensity, fatigue, exertion, and to some extent proficiency/movement quality.

TYPES OF FEEDBACK

               My first introduction to VBT was through a facility that didn’t use it for powerlifting. I’ve always been interested in it from a powerlifting perspective, but I won’t pretend like that’s how it’s commonly used or what it’s most optimal for. But it does pose some advantage for powerlifting since it’s a different type of feedback. John Grace has written and presented on athlete monitoring systems and he breaks it down well enough that a simple army guy can understand it (I can say that, am army guy) – hell, maybe even a Marine (sorry Cody). All of it comes back to motor development principles of feedback operating on two intersecting continuums: the internal and external, and the objective and subjective.

• AMRAP - As Many Reps As Possible. Can be used to finish an exercise. From this AMRAP you can calculate an estimated 1RM, which feeds forward into a training max that loading schemes are based off of. Usually planned in the weekly context.

•  RPE - Rating of Percieved Exertion. Popularized by Mike Tuscherer. A given set can be rated on a 1-10 scale in terms of difficulty and how many repetitions/additional load the lifter felt they could have completed had they gone until failure.

Something that is internal is a feedback variable that is of origin to the athlete, like their heart rate or their rating of perceived exertion. Something that is external is a feedback variable that is of origin outside of the athlete, like their movement velocity or a qualitative assessment after reviewing video (IE: “That squat looked like hot garbage”). Any type of feedback is going to be internal or external as well as objective or subjective. Objective feedback is tied to some universally agreed upon standard, like time, velocity, or some other metric common to all. Subjective feedback is tied to qualitative assessments, opinions, and personal inclinations.

Now that we’ve unnecessarily deconstructed that, here’s how that breaks down according to some examples of auto-regulation methods:

RPE	Internal, subjective
AMRAPs	External, objective
VBT	External, objective
Video review of the set	External, subjective
Rest interval is over after HR returns to X bpm	Internal, objective

               Anyone that’s interested in VBT should probably be interested in auto-regulation first. That said, VBT is only one way to auto-regulate training. If you’ve had success with auto-regulation training, then you might have success with VBT. For example, if you find that increasing your training max by performing an AMRAP at a given intensity both helps upregulate (due to improvement) and downregulate (due to fatigue) your training appropriately, then VBT might work for you. If you find that making judgment calls yourself about the difficulty of your workout, RPE or otherwise, guides training productively by increasing or decreasing training load – then VBT might work for you. If you find yourself diverting frequently from your training plan based on feel, then VBT might work for you. Chances are, it’s better for you to try at least one of those methods first before you try VBT. There is no monetary cost associated with performing an AMRAP or using the RPE system. But there is certainly a cost for adopting VBT. Under sell much?

Auto-regulation training isn’t anything new or fancy. Bro’s have been using auto-regulation for years based purely on how they felt the day of their planned training. The difference is most auto-regulation methods attempt to assign values and utilize quantitative measures to dial in just how much or what factors to regulate. 

               The common argument that people have against VBT is the idea that it will consume your entire training style – which is ridiculous. It’s another tool to improve your performance, not a crutch. It’s not essential, but neither are wrist wraps, and I don’t see people up in arms abo… oh wait, nevermind. You don’t need VBT anymore than you need to video record your top end sets for review – but it sure can be helpful. If you don’t want to use it – don’t. If you haven’t really used it, then what is your criticism worth? Using a device for a week doesn't qualify you to have a worthwhile opinion on the matter. Just sayin.

I forgot we’re in the age of the internet and everything is offensive, including what other people are offended about. I’m not going to attempt to convert anyone on this. This is absolute niche of the niche. We’re talking about one tiny segment of ideologues (VBT proponents) in a barbell sport that struggles to gain recognition as a sport (powerlifting). You could be either of those two and be resented by a good amount of people. If you enjoy resentment/being forever trivial, do both.

               Another thing we have to accept is that VBT is essentially snowflake training. It is based on the assumption that everyone responds differently and adapts differently to different types of stressors in different ways and it affects different factors like different intensities and different volumes and different differences. You can over-fit VBT to your training and individual responses, and it might be meaningful, but it might be a poor investment of time for little gratification. If it helps you make improvements, like improving your 1RMs, increasing your volume, or pulling you back from CNS nullification (I was told doing more than 1 set of 5 deadlifts would fry my CNS /s). The idea of daily estimated 1RM’s fluctuating wildly is probably overstated by VBT proponents. Usually, when I’ve observed that phenomenon, it’s was either obvious before conducting a submaximal test or such a small difference its consequence was minor. That said, if you could get a 5% improvement in your total, why wouldn't you take it? Some folks obsess about nutrition which could have equally small effect size, but for some reason auto-regulation is questionable? 

Where "/s" means sarcasm. Just want to make that clear. If you're trying to transition from StrongLifts to VBT you should probably do a powerlifting program first.

               Like almost every grand idea in powerlifting, you can essentially ask one question and deem relevance. Make gains? If yes=is good. If no=lol STFU. It’s not for everyone, and it might even cause more harm than good – but if it works for you, then it works. It’s not the first crazy, niche thing to come to powerlifting. Hopefully, it’s a step above keto.

FREQUENCY OF FEEDBACK AND PROCESS-ORIENTED TRAINING

               The frequency of feedback is another factor. A simple analogy here is using an AMRAP to adjust your training max. You’re likely going to do an AMRAP once a week to change your loading scheme. That single source of feedback will up or down regulate loading until you conduct another AMRAP. RPE, on the other hand, is readily available. You can access that feedback on the fly: in the middle of a set, after completing a set, after a session, etc. You can also use multiple time domains with that information and shape the training. VBT acts in the same way, depending on the device. Some devices allow you access that feedback within the set, some readily give you useful metrics after the set (percent velocity loss, average set mean velocity change vs previous set, trend analysis, etc), and some cloud services will allow you to view how one day’s performance stacks up against previous days’ performance.

               That frequency of feedback is important to how it shapes training. Normally when we think of training plans, we think of overly thought out planning. It immediately makes me think of this:

Borrowed from Bompa and Haff from Periodization:
Theory and Methodology of Training

               In the powerlifting perspective, this is primarily seen in percent-based training methods, assigning a percentage of 1RM as intensity and varying it with intensity to approach different aims within a training cycle leading up to a competition. I’m not going to rail against periodization in powerlifting though, I’m not Louis Simmons. I'm also not going to try to represent different methods of periodization in powerlifting, but I think a good place to refer anyone to on the subject would be Chad Wesley Smith's Scientific Principles of Strength Training.

Food for thought: a lot of periodization attempts to dissipate fatigue of a training cycle. If  auto-regulation training can control for fatigue, you could get better training economy or have actual factors to determine appropriate cycle length and better 'potentiate' one phase to the next.

               That said, if we use objective feedback to shape how our training plan evolves, our plan becomes more process oriented than it is plan oriented. As a coach of mine used to put it, “We write our plans in pencil.” We’re not completely winging it, but essentially we can complete the volume of work we’re capable of within a session or microcycle at the given intensity that is appropriate given our recovery to ensure we stay on track to accomplish the goals of our tentative plan. If our feedback is only once a week, we can only adjust once a week. If our feedback is more frequent, we can adjust more frequently. However, if we adjust too liberally or accept every variance as a cause for adjustment, we run the risk of over-fitting our training. We can’t afford to miss the forest for the trees, and that is a particular trapping for auto-regulatory training.

               RPE is great to use as an example here because Mike Tuscherer has already developed a thorough system to help recognize the potential trappings of over-relying on feedback. For example, his development of TRAC helps identify those trends through a mix of internal, external, objective, and subjective measures. More simply, looking at previous training logs could do the same thing. There a hundred ways to get broad pictures of how training is developing, VBT provides a window into how training is working fairly immediately. If you want to know if it's worthwhile, you should probably measure it by another metric, like baseline volume, volume by lift, the number of lifts, or one of the million other metrics places like MyStrengthBook track.

               Training is partly process-oriented, and you should strike the iron while it’s hot. Maybe not every time, and probably not always with intensity or volume, but if we fail to capitalize on training opportunities when they present themselves we’re doing just as much as a disservice to ourselves as we would be when training through our low points when we’ve objectively over-reached.

SOME PRACTICAL LIMITATIONS: The Case AGAINST VBT

               ISOLATIONS AND ENDURANCE WORK

From Supertraining by Siff and Verkhoshansky based on the
data from others listed at the bottom of the pic. This is better
than the F-V curve and you can make the argument that
hypertrophy is the product of strength-endurance

VBT helps give objective feedback on many things, but we should probably give a realistic expectation of where it isn’t best suit or absolutely struggles. One place VBT is not appropriate is isolation movements. Firstly, you might not observe the same velocity relationship in a bicep curl that you do a squat. Concerning the F-V relationship, Siff and Verkhoshansky posited the additional idea of strength-speed-endurance. We can think of strength as force, speed as velocity, and endurance (in powerlifting terms) as reps. In these isolation movements, we’re talking about supporting muscles that could have atypical velocity responses to what we see in a compound movement because their role is more endurance focused – an element we’re not incorporating. That’s an overly complicated reason why you shouldn’t implement VBT with isolations, but the simpler reason is that it’s wasted effort to apply VBT across the board. Adding VBT does add time, but it’s also not worth majoring in the minors here. Additionally, you’ll see weird results for sets of 15+ reps. And by weird I mean unactionable data.

A good rule of thumb to apply here is if there isn't a 1RM, 3 RM, and 5RM for that exercise, it's probably not fruitful to add VBT to it. I say that because for some people and/or some exercises, there are only 1RM's and 6RM's, with no weight inbetween. Velocity operates more reliably in situations when you need to gauge proximity to failure. The less granular points in proximity to failure, the more the relationship of velocity to exertion/intensity falls apart.

For compound movements for several reps, the mechanism that probably screws up the measurement is probably self-selected pacing. If you know 30 reps are coming, you likely won't exert maximum velocity - at least not throughout. This is something you might be able to mentally power through for hypertrophy work or high volume work, but in my experience can be avoided. VBT will give you feedback and you'll be able to see the points where you reset to take your air or reinforced form cues.

Adding VBT to isolation movements isn’t practical or logical. Maybe it’s a selling point of incorporating VBT for recreational lifters by manufacturers, but it doesn’t provide actionable data for powerlifting. Collecting data for the sake of building a data silo makes about as much sense as worrying about how your hair product is going to increase your lifts. But then again, ROM gains? Mullets= legal ROM cheat

TIME COMPONENT MOVEMENTS AND VARIABLE CONCENTRIC/ECCENTRIC REPS

Tempo work is an obvious shortcoming for VBT measurement. If you’re doing tempo movements with controlled descents, pauses, and ascents, there’s not much useful data there. If you control the time and the displacement is near the same every time, you’ve controlled velocity as well. Maybe you can identify inconsistencies, but likely no better than you could just by feel or video. Using VBT for tempo work is like using the snatch or clean and jerk for conditioning, it’s the wrong tool for the job.

Update: An article came out in November covering VBT eccentrics. The results did not overwhleming favorable using VBT eccentrics over VBT regulated training for strength. I'm not going to open the can of worms that is ecceentric training, nor am I going to muddy the water with VBT in eccentric training. 

Depending on what device you’re using, paused positions will be fine. For example, paused squats in the hole or bench paused at the chest. A high pause near a sticking point might not work for different devices, but it all depends on device limitations. Most VBT devices output the upward part of the movement in their metrics (mean velocity, peak velocity, power, etc). In the device’s eye, unless it knows how to measure this movement, it might record it as two separate lifts, but will most likely start recording from the first upward movement until the downward movement, skewing that velocity. As much as we want to believe that our pauses are still action, they are milimiters of movement up and down. The actual velocity we would want in this case would be the velocity of each upward movement, but I haven't seen an easily device that does that yet.

There is some evidence that mean propulsive velocity is less appropriate than mean conventric velocity in the bench throw. 

Some devices compensate for this by setting a threshold for reversal (the slight change in direction despite pausing) or velocity. Some devices just can’t detect well under certain velocities, so they terminate the measurement below a threshold (for example, 0.15 m/s) and count any additional movement as a whole other rep. Open Barbell filters out velocity below 0.02 m/s. These numbers and this unnecessarily technical explanation don't have any meaning, though. Just do the movement and see if it’s screwy. If screwy, don’t do-y (with VBT).

Special note on tether devices: There could be some additional limitations, but not many that I feel I impactful. For one, I'm not sure there's an optimal way to tether-track pull ups. Even though there is a way to attach a tether to a barbell lunge, keep in mind that the tether does not compensate for angle of pull (except the $2200 GymAware), so you're likely going to introduce so much anterio-posterior movement to make the measurement meaningless. For a device like Open Barbell, simply attempt to measure out what the vertical bar displacement should be vs the length as measured by the device. Since velocity is displacement by time, the velocity could be off. Regardless though, that movement requires balance, which I added at the bottom.

Two movement types that shouldn't be excluded are movements where the pause happens between a beginning eccentric phase and a concluding concentric phase as well as pauses that occur in the eccentric phase only. This is exactly what describes a competition bench. A study by the Spanish VBT proponents seems to say a pause between the eccentric and concentric increases the reliability of the measurement. An example of pausing during the eccentric phase only would be something like pausing at the high sticking point in the squat on the way down (not on the way up). The explanation for this is more simplistic, as devices generally only measure concentric velocity. The load might be reduced by the intensity of what you're able to accomplish, but the measurement is solid given the minute changes in height don't exceed the device's filtering threshold. I believe Open Barbell's threshold is 2.8 mm based on their wiki, but I suspect it might actually be higher (anecdotal, or I don't understand metric system units of length smaller than a meter).

Gabrielle Tucker has a IG post show what I mean with pauses, and also she's just an amazing lifter you need to be aware of. In this video she has two pauses. One coming directly off the floor, and one right before returning to the floor. In the VBT context, the one coming off the floor would skew things. However, the one returning to the floor would not skew the measurement. Gabbrielle doesn't use VBT but is a world record holder, proving the point that VBT is just a training method, not the only training method.

I'm sort of obliged to ask if pausing concentrically or eccentrically nets the same results from an adaptation standpoint. On one hand, the mechanism could simply be that isometric exercise increases strength in that range of motion, give or take 15 degrees on either side. On the other hand, maybe the important part is the rate of force production from a dead stop in the concentric phase. I'm more inclined to think the main benefit comes from isometric strength effects rather than the accelerative part of the movement, but I'll also admit this is not something I'm well versed in. You could still yield the positive effects of acclerating from a dead stop with things like pin squats in that weak range of motion, but I also understand flexibility is appreciated.

BUILT IN GRAPH OUTPUTS

I think people expect thorough outputs of data that aren’t realistic from VBT devices. A simple example is velocity-time graphs, acceleration-time graphs, displacement-time graphs, and displacement_x-diplacement_y graphs. Firstly, much of that information can be obtained in more qualitative ways and perhaps in more appropriate ways through video. If you want it in more quantitative ways (because you absolutely need to know what the bar displacement/speed is every 60^th of a second, etc), you can run it through some free video analysis software. I prefer to make more qualitative judgments through recording sets where appropriate (external, subjective – if you didn’t skim through to this point), but it might be helpful to get down to the nitty-gritty for top-end sets or near competition prep. Some VBT devices might offer this and others might be developing it for future updates, but don’t expect it right off the bat or think that’s the only utility for VBT. You could get that sort of overly-analytical information for free with equipment you have now, but it’s not going to be immediately actionable (nor does it need to be).

               Some might be lost on what I’m talking about. Here’s what displacement-time, velocity-time, and acceleration-time graphs would look like.

Displacement, velocity, and acceleration in the y component vs time. Data pulled from tracker.

All X axes are the time in seconds. The Y axes from top to bottom are displacement, velocity, and acceleration. All graphs are synced to time. These might be useful to have, but you’d have to have a decent intuition and understanding of what these sorts of graphs are supposed to look like for different movements. If you wanted to track bar path in the X and Y (fore/aft and up/down) quantitatively across time, you could do that if you had a triaxial graph, but it’s more helpful to watch the graph as the video plays. Here’s an idea of how messy that information can be:

An example of an ugly and indiscernible x-y displacement graph. Scale is important here. Notice the X axis is emphasized

               It’s probably going to be more helpful to watch the graph as it plots. This video (linked to watch in a new window) explains exactly what I mean. All this information was obtained through an iPhone shot at 60 frames per second and a free software called Tracker. 

Open Barbell Output
as of Feb 2017

This means this is very low cost. Maybe spring for a tripod with a level on it and a phone holder. Bryce Lewis from The Strength Athlete has a YouTube walkthrough of how he uses it, but it’s very, very intuitive for an entry level user given a remedial understanding of physics. VBT devices not having this may be a limitation, but it might not be worth consideration when purchasing a device. Getting a VBT device for this exact purpose is like buying a BMW because you like the cup holder. I understand holding beverages is important, but the main purpose of BMW’s is not their cup holding capacity. Consider how “little” information Open Barbell as of February 2017 presents. The data is not super sexy, but it has a high level of utility. I know VBT isn’t necessarily in the habit of presenting itself as KISS, but maybe we should just keep it simple, stupid. With the output in the picture, you can get 85% of the data you need to make actionable VBT decisions.

               FORCE AND POWER OUTPUT

Some systems give you measurements of force. Some don’t give it to you upfront, and you’ll have to export the data through a web portal that syncs all the metrics. How it determines force is somewhat amiss to me, but the easiest way would be accepting the user input of mass and multiply it by acceleration since F=ma. This isn’t “real” force though, it’s just force of the barbell that's mostly in the vertical direction, a rough proxy for ground reaction force that accounts for the lifter’s body mass and all resultant forces. It's sort of a truncated metric of force, viewing the system as the barbell and additional weight alone. Nonetheless, sometimes a proxy is good enough. At least I think. Who knows, it's proprietary. 

In the Westside Barbell scheme of things, you can train force output (and therefore strength) by increasing mass on the bar, or increasing acceleration (reminder: F=ma). If you work with a system that actually outputs force, you’ll see that you mostly increase Force by increasing mass and the discrepancy between light and fast weight force vs heavy and slow weight force tends to favor heavier loads. Conceptually it sounds good, but in practice it doesn’t work out so well if we accept force as measured by VBT devices. Mike Tuscherer has written about it on JTS (here and here). Jovanovic also makes a strong point that viewing it from the exact readouts of force output might not be the most appropriate way of looking at speed work. Chad Wesley Smith has more thoughts on the matter in YouTube format if you're extra lazy. The TL;DR version is you produce more force by just adding weight while moving as fast as possible - not by using lighter weight and trying to move it faster. At that same time, it's arguably whether the long term outcomes of training at submaximal weight as fast as possible is comparatively better than just training with heavyish weight. There’s a host of other reasons that people inside and outside of the PL world have cited why this sells well but delivers short, but I won’t belabor the point. You might want to have force as a metric, but you don’t really need it. Weight on the bar or velocity is a good enough proxy.

Force and Velocity by %1RM on a straight bar
deadlift and a hex bar deadlift. Source

If you are focusing on the power output (or force even) and other metrics that calculate based on load, it’s probably not appropriate to take some of that information seriously when it comes to accommodating/variable resistance. That would include bands and chains. It will record velocity fine, but power accounts for mass as well. Usually, it’s measuring power as the product of force and velocity. Since force is the product of mass and acceleration and it has no way to track the change in resistance from the top of the lift to the bottom, that measurement is close to meaningless. VBT isn't the best way to track this change in resistance. You might be able to use velocity to autoregulate the training, but it might be more appropriate to use another method like RPE. I guess you could also do a 3D video analysis, manually calculate to include elasticity of the band, and make your workout as math heavy as launching satellites into space. Maybe as long as variable resistance is a tightly fixed percentage of bar weight this might make sense. Maybe keep it simple, stupid?

If we're going to use biomechanics to drive the training process, it's best if we have something that does the thinking for us. I don't advise mixing physics and exertion.

In any case, I think it's disingenuous to transliterate velocity to training quality in all respects. It sounds good on paper if you accept the table of training quality and movement velocity as dogma. But I would venture to say using variable resistance and inertial considerations is better at training a weakness in a range of motion (EX: speed out of the hole) rather than some classical Russian classification of strengths. I like my Russian ideologues, but I disagree with citing them dogmatically. The right answer probably starts with, "Well, it depends on..." Many of the folks espousing these ideas insist you need a good grasp of physics to think about powerlifting. Probably. But physics brings up many counter-arguments to speed work. Short of calling speed-work bullshit, it's an unanswered question.

Furthermore, VBT for power works well as a guage on force in movements where you don't leave the ground. If you leave the ground, like a squat jump, you're no longer exerting force on the ground. Think of jumping up and down on a bathroom scale. When you do the counter-movement and load the legs, the scale will read higher for a split second. When you've actually jumped, the scale will quickly return to zero. If you did this with a bar on your back, velocity would still be reading over the whole length of your jump. So if you're 1.67 meters tall, jump 0.3 meters for a quater of a second, it has averaged in 0.5 seconds of flight time where there is no force being exerted on the ground. There are other ways to use VBT here, but it's best not to take the readout at face value. This is also why Bryan Mann likely uses peak velocity for cleans.

BALANCE COMPONENT EXERCISES

Exercises that have a balance component might not be appropriate either. Adding the extraneous detail of maintaining balance skews things in terms of velocity. While it might be hard on a motor pattern level, the muscle might already be strong enough to gauge its objective feedback without VBT. This applies equally to those pistol squats as it does to those SSB Squats you added into your routine that you’ve only started doing. Usually, you want movements you’re proficient in, and that’s more of a judgment call. It’s for this same reason that VBT's not appropriate for beginners. There needs to be a certain amount of confidence with the movement before the metrics become meaningful.

If I had to guess why movements that have a high balance component don't fit it, I would guess it's because the constant movement in multiple planes skews the vertical velocity. That movement in other planes usually isn't part of consideration for vertical velocity. In theory, this might be more pronounced with systems that measure movement in less planes. Accelerometer systems, for example, usually measure in 3 planes. Tethered systems usually measure in one plane (Tendo, T Force, Open Barbell), but GymAware measures in two. Bryan Mann has said in the past he also has issues with front squat velocity, and I think it could be related to this. Bryan Mann also uses the Gym Aware.

One way to remediate this situation is to decrease the balance required. Squats with a safety squat bar (especially ones with a radically forward camber) are a great example of this. The movement generally is heralded because it pulls you out of position on purpose, forcing you to have greater stability. IF you wanted to smooth out velocity across a set (make it so you have a solid read on successive velocity decrements each rep), you can separate the eccentric and concentric portion. So instead of doing regular SSB Squats, do SSB Squats from pins. Lower the bar to some safety pins/straps, pause, then lift the load. This does change the training stimulus though. The training stimulus is even less so if you modify it further. For example, if you wanted to make sure the camber was challenging you in the bottom position of the squat, you could lower to just above the safeties (not touching them), hold for whatever amount of time, lower the last inch down to pins, then lift the load. I've done this successfully with my SSB Squats from pins, but then you have to live with the fact that you're maring such a beautiful bar.

This is a point of contention emerging in the VBT circle. Banyard et al recently put forth that velocity is less stable in free weight exercises because of the stretch-shortening cycle (SSC). Their main contention is that most experiments have been conducted on a smith machine (which provides some study constraints that help elucidate factors but also loses some context). Their case sort of sits upon the idea that the SSC is mitigated by the smith machine. I will say that's a pretty fair point. You can demonstrate it easily by performing a free weight bench press followed by a pin press. A pin press will generally create a higher R^2 for its load-velocity relationship. By contrast, though, the popularity of dive-bombing your squats on the platform to overcome your sticking point in powerlifting isn't nearly as common as it is in something like weightlifting or general strength and conditioning. Long story short: don't bounce out of the hole or off the chest. 

You could apply this same concept to other variations that have a balance component. Reverse Grip bench press is a possible example (I've never done it, but worth checking out). Front squat might be a bad candidate for this though. If you were to front squat to pins, lowering the bar onto the pins could only amplify the balance requirements by pulling the bar out of the front rack position. There is a possibility you can negate that with a cross-armed position or straps around the bar, but I also haven't tried this. I've heard from some other lifters that lessening the shoulder/torso balance component of the front squat (front squatting with straps) mitigates the noise of successive velocities.

CONCLUSION

VBT can be a useful way to auto-regulate training. If you find that more frequent feedback helps enhance your training or that external, objective feedback really helps, then VBT could be a good fit. If you prefer dedicating yourself to a set plan and can’t be sure to trust the information external, objective feedback, it probably won’t help you. To some extent, implementing VBT can mean diverting from fixed planned training and individualizing your training practices, but the extent to which it does that depends on how you implement it. There’s no guarantee that you’ll make gains by injecting VBT into your program alone. Ask your doctor if VBT is right for you.

VBT has the potential to give you quantitative ways to understand the appropriate intensity, fatigue, exertion, and technical proficiency. They aren’t the only way, but the feedback can be succinct and actionable. It is meant to be a training tool, not a training necessity and any attempt to make them essential to that process is likely to be met with limitations. These limitations can include limiting how tempo reps are used, VBT regulation of a general physical preparation component, VBT regulation of isolation movements, and balance predominant movements. Additionally, it’s important you establish an understanding of what metrics you need it to collect as features are not universal across all VBT devices.

At the end of the day, VBT is a training tool for the tool box. If VBT is the most insightful part of your program, your program is probably lacking. I would argue that how you distribute your training load and your exercise selection will have a greater impact than just adding a VBT element. Despite how much I'm also laying out the case for process-oriented training, I'm also arguing the importance of planned training. If you don't have experience with percent based training, you'll probably struggle to find the appropriate way to implement VBT. I would advocate the appropriate transition to VBT is to take an existing percent based training program (or design one), and then add elements of VBT as you go. Where VBT doesn't fit, use traditional progressions and loading schemes, or attempt to blend other autoregulation methods like AMRAPs and RPE.

Plenty of people have been breaking records and getting strong without VBT. As interesting as VBT might be, there's not an extensive roster of people that are dominating national or international powerlifting meets and espousing VBT as the reason. My qualifications as a powerlifter are questionable as well, even when you narrow the consideration down to my weight class. My position as an "expert" in VBT for powerlifting is not backed by my achievements. If you asked me, I don't think you're training wrong if you aren't using VBT. My purpose here is to give an accessible understanding of how VBT can be used in powerlifting, and also just refer people to links because it's not something that can be explained easily in conversation.

While this article mostly concerned reasons to implement VBT, the next one will build on that giving the nuts and bolts of what makes the whole system work. Eventually, this will get to a point where I'm giving example programs I've run to achieve different training goals. The current plan is to release everything in 1-week intervals, and this series will come to an end at some point.