A Powerlifter's Guide to Velocity Based Training Pt 3

My n=1 Experience and Best Guess Practices

Life pro tip: Enable text to speech on your phone (iPhone or Android) and have that guy read the article to you. I do this all the time with Stronger By Science articles. Read smarter, not harder.

TL;DR

• VBT isn’t about speed-work, but you do have to move with fast intent and consistent form and kinematics (IE: either squat with or without a stretch reflex, choose one)
• Load-velocity profiles help translate %1RM to velocity. This relationship appears fairly stable over time and helps prescribe intensity according to what the athlete is capable of that session.
• VBT 1RM estimation allows an athlete to assess 1RM submaximally, sparing the potential fatigue and programming considerations for AMRAPs and 1RM testing.
• Velocity loss within a set or drops in opening/set average velocity can be used to target strength or hypertrophy adaptations and establish fatigue thresholds. This should not be confused with attempting to limit fatigue unproductively.
• How to improv a VBT workout if you’re trying to familiarize yourself with it
• Next week I'll actually drop some VBT programs

DISCLAIMER

I should start this by saying that the Spaniard VBT proponents are Gonzalez-Badillo, Medina, Marques, Pareja-Blanco, Rodriguez-Rosell, Gorostiaga, Izquierdo, Ibanez, and probably a few others. These individuals along with Cronin, Mann, Baker, and others to some overlapping extent (Haff, Kraemer, and Stone come to mind) are the forefront of VBT research. I say the Spaniard VBT proponents not as an insult to the Spanish academics, but ease of explanation. While I try to use some learned reasoning when I write these, I won’t pretend like my style is academic. If they were academic, I wouldn’t be dropping links to Murderbot Productions’ videos alongside journal articles.

INTRODUCTION

               It’s entirely appropriate to say that most of the benefits of using velocity based training (VBT) apply most appropriately to training for power and not powerlifting (PL). Luckily, there’s enough overlap between power and strength training since one tends to be programmed to potentiate into the other that we can still glean some utility out of it. I’m not going to make grandiose claims VBT for PL is going to prevent injuries, add kilos to all your lifts, or that all other methods of programming pale in comparison. My intent with the whole series is to spur interest, increase participation, and create discussion about VBT for PL. I do not claim to be the subject matter expert on VBT. I’m just someone with an interest in VBT and powerlifting trying to make lemonade from some lemons.

THE RATIONALE OF LIFTING FASTER

               If you intend to get the best results with VBT, one of the general things you might have to get used to is maximum intended velocity, or moving the weight as fast as you can with sufficient form regardless of fatigue. You're not attempting to game the movement. You shouldn’t be accelerating the bar so much that you’re bouncing it off your back or throwing yourself off the bench (squat and bench). Trying to move a heavy load as fast as possible sounds like a training obstacle, but when you consider that we’re talking about loads that are innately slow to begin with (EX: 80% 1RM loads), the velocity difference from as-fast-as-possible (AFAP) to self-selected pace is markedly similar to the naked eye. The Spanish VBT proponents found that moving loads with fast intent in the bench press, AFAP subjects were seeing an 18.2% (±11.9%) increase in 1RM versus 9.7% (±7.9%) for a purposefully slow concentric group. The point of this is that there is more than one reason to move the load as fast as possible while still maintaining technical proficiency and it extends beyond just enabling VBT utility. I won’t over-emphasize this point though, this should be considered more on an ancillary benefit. As usual, Greg Nuckols makes some very fair points on this, which I have no intention to cover, but every intention of linking to. This sort of runs counter to my assertions about the conjugate training's approach to speed work, but speed work in that case is often relegated to lower intensities. The point here is any intensity moved AFAP will have better gains, so why bother just limit it to the light stuff? There are time under tension concerns for practicing VBT, but that's a whole other can of worms. 

               I should also emphasize that some amount of consistency in form and technique is also required. If you’re going to use the stretch reflex from touch and go or bounce out of the hole, you need to be conducting this when you attempt to compare past performances to in-session performance. Generally speaking, including a stretch reflex does come at a detriment to the reliability of the measurement. If you change your technique or your form, you might find a disconnect between prescribed velocity and intended load. This will be covered in more detail later in the article.

OBLIGATORY SPEED WORK SECTION

              You can skip this if you don't care for conjugate method circle debates. 

               As I’ve said previously, I’m not a big believer in speed work. Many others have figured out how to get it to progress their training, but I haven’t seen any tangible improvements using it. Given that disclaimer, Louis Simmons popularized the idea and tends to suggest speed work in the confined terms that it increases your force output by spending equal emphasis in progressing the contribution of mass and acceleration. Louis Simmons seems to espouse the generalities of the force-velocity curve without regard to individual differences, differences by movement, etc. Maybe in private practice this isn’t the case, but the blanket coverage is to just aim for 0.70-1.0 m/s in the bench press, 0.7-0.8 m/s for the squat, and 0.9-1.2 m/s for deadlifts. Matt Wenning, on the other hand, tends to recommend anywhere from 1.0 to 1.2 m/s.

Force output is better attenuated by mass than it is by acceleration, but it’s not absolutely clear if there is some ancillary benefit to training within this range. In terms of force, the outputs don’t match the end result – but that might not mean anything. PUSH allows you to see force on exporting from the web portal, so the feedback is fairly delayed. Form Lifting also gives you force in pounds of force (Newtons be damned). OpenBarbell does not calculate force.

Force by percent of 1RM on a straight bar deadlift and a hexagonal deadlift.³

               Without injecting my opinion, there are some people that are still seeing improvements out of this method. One method of determining the most appropriate load via force output would be a Force-Load (F-L) profile, Force-%1RM profile, or you could just use the rigid speed zones. If you want an F-L profile, use a sample of 5-8 sets with 1-3 reps in a different zone of intensity - not much different than Load-Velocity or Load-%1RM profiles. Most devices don’t get instant readouts of force since it’s a calculated metric, not a measured metric. Once you identify your optimal force zone, you can likely use that velocity or load as a reference point in future workouts under the presumption that’s where the best training effect comes from. There is something to be said that training force output indirectly via below peak force output helps create that adaptation. I covered this in the previous article, so let's stop beating this dead horse. 

1RM ESTIMATION

Mladen Jovanovic and Eamonn Flannigan have an excellent review article on researched applications of VBT. What’s even greater about it is it gives very specific details on how to implement it. Even down to the detail of the formulas that are used in MS Excel. Both have an online presence or lecture on the topic of VBT in a way that has a high amount of utility to powerlifting. To summarize it easily:

We have recommend measuring mean concentric velocity at 4-6 increasing intensities of load ranging from 30-85% of actual or estimated 1RM to estimate load/velocity profile. To estimate 1RM for an individual, coaches need to know the MVT of the exercise which could be assessed through traditional 1RM test or reps to failure test.

               This is essentially a two-part test. My current practice is to leave a week free from training and conduct a load-velocity profile, followed by an AMRAP to determine minimum velocity threshold (MVT). Minimum velocity threshold is your typical grinder rep. Generally speaking, the velocity of your last rep in the tank at 75% 1RM and 100% 1RM are similar. An AMRAP is the more feasible way to go, but I’ve found it’s helpful to update your MVT whenever you see a measurement that was slower than one you performed. This is not to say that I’ve actively sought out to train to failure, but it happens often enough even when trying to limit fatigue through things like VBT (for example, hitting ~92-95% 1RM for singles for 8-10 sets). On the rare day that I do max out, it’s essential to get a measurement because that is the direct measure I want. This works differently than conducting weekly AMRAPs because it’s an incidental part of the training process. I won’t say it’s the most accurate way of progressing your training (at least no better or worse than adjusting training max off your rep maxes), but it’s accurate enough to be functional.

AMRAPs usually function as scheduled auto-regulation, add to or taking away from the training max and influencing training load for the following week. By changing out MVT whenever we establish a new one (AMRAPs at a point of our choosing or the surprise grinder), we implement this progression/regression as often as we deem necessary and affects our VBT estimated 1RM which functions the same as a training max. By using the %1RM-velocity relationship, we can get more granular and adjust our training load day to day.

               One of the more ironic things about 1RM I’ve noticed is that whenever I’ve gotten stronger, my 1RM has gotten slower (so much for velocity based training) and my velocity at standardized loads has become faster. This phenomenon (strong=slower 1RM) has also been observed by the Spaniard VBT proponents. Whereas some people view new rep maxes as a sign of progress, in the VBT context of things I have regarded new MVT’s or a sustained increase in velocity at a standardized load as progress. Velocity at a standardized load would be something like a cyclical check on velocity of 80% of your last known 1RM. If that velocity gets faster over time, you've probably gotten stronger over time. In my mind, a new MVT is more meaingful and is essentially a new PR. The general baseline for the population is that 1RM is near 0.15 m/s for the bench and deadlift, and 0.30 m/s for the squat. If your MVT isn’t near that on the device you measure it with, you could just chalk it up to measurement error. As long as your device is reliable, it shouldn’t matter.

Let's make sure we're comparing apples to apples here. Because someone hits 1RM on their squat at 0.3 m's and someone else hits 1RM at 0.26 m/s doesn't mean by default that the slower person is stronger. It's very specific to the individual. A better comparison is that someone that maxes at 0.3 m/s in a cycle is likely stronger by the time they hit a max at 0.26 m/s later. 

               Some people have advised conducting a VBT 1RM test before a workout to serve as a warm up, then conduct a prescribed load based on that estimate for a given number of sets and reps. I don’t think that’s appropriate for reasons I will explain in load-velocity profiling. Doing this test appropriately (not rushing through it and biasing the test) every session should extend your session in the neighborhood of 30 minutes. I would argue this is not the best use of training time.

AN ASIDE ABOUT BUILT IN 1RM TESTING

               This is a rant. You can skip this. It’s important to mention that some devices have apps with built in 1RM testing. There are multiple issues I’ve found with this. 

1           These tests seem to be inflexible if you’ve conducted a 1RM test before. It will base the loads that map your 1RM estimate on the last estimate. This is a big issue for powerlifters who have a gap between their sumo deadlift and their conventional deadlift. My sumo DL is 25 kg higher than my conventional. Because I made the foolish mistake of testing my sumo DL with the deadlift VBT-1RM feature, it prescribed sample loads based on my sumo DL 1RM estimate. This means rather than making the estimate from loads of 40% to 80-85% 1RM, it’s shifted it to around 50% to 90-95% 1RM. This isn't the worst, but I'd appreciate the flexibility to delegate load for testing myself. There is no option to test a different type of deadlift through this feature. And god forbid you ever mess up during the test and it somehow adds 50 kg to the estimate. There could be no going back. Ever since one app update removed the ability to change the prescribed loads, I've calculated the VBT-1RM estimate manually.

2             Rather than following the protocol described before, it assigns you to conduct 5 sets of 3 reps. This seems like an arbitrary call here, but I don’t see the purpose of hitting 85% loads for three reps to map the estimate. It’s extraneous work that may not be necessary. You can complete less than 3 reps, but the chances of the app returning an error instead of an estimate increases. Additionally, it might be useful to map 1RM using more than 5 sets (and therefore 5 different intensities), something I’m trying to transition towards. I’m also entertaining the idea of shifting the starting intensity to 55-60% 1RM since I’m not very concerned with what the corresponding velocity for 40% 1RM is.

3              Part of the reason I don’t like this feature anymore is because it’s a black box. You perform a bunch of inputs, it does some mathimagic, and bada-bing bada-boom, you have an answer. Not an explanation, just an answer. I don’t know if it’s averaging velocity amongst a set or if it’s using a maximum value. I don’t know what the confidence interval is (95% or 90%?) to derive the same answer. I don't know the standard error of the estimate. If I can’t reproduce the answer, I don’t know if I should trust it. And I’ve had some estimates that were off by quite a bit.

4                I believe it uses generalized MVT’s. This can both shortchange your estimate and determine a value below your actual 1RM or it can overestimate it and prescribe you death by bench. If your 1RM is faster than 0.15 m/s for BP/DL or 0.30 m/s for a squat, it’s certainly SkyNet becoming self-aware and trying to kill you with a barbell. If you’re slower than those velocities, the gains goblins are trying to pull one over on you. It shouldn’t be off by that much, but it might be off by enough to affect your training.

All and all, it’s a nifty feature and I’m glad developers are focusing on this. It would be more helpful if load-velocity profiles could be an integral feature. It would certainly make autoregulation easier and probably eliminate the needs for a spreadsheet in training. So far as I have seen, one bar speed tracker has attempted to do this. The app Powerlift stores your load-velocity profile from a previous session. Once stored, you can do a rep at any load and it will estimate what %1RM that was. It's all camera based, which I have railed against before, but they're a relative new-comer to the scene that have implemented such a simple feature straight off the bat. 

LOAD-VELOCITY PROFILING

               I tend to lump load-velocity profiling in with VBT-1RM testing. I consider my MVT measurement, which is required for VBT-1RM testing, a rolling measurement process. If I ever feel like it hasn’t been updated in a while, I’ll go for an AMRAP to make sure I have the most up to date measurement. If it's an exercise I haven't profiled before, I perform an AMRAP. Once I have a load-velocity profile, I tend to convert it to a %1RM-velocity profile. This acts almost like a translator for percent based training (PBT). This way I can still program by %1RM, but train in VBT.

               My current method of doing load-velocity profiling essentially the same as VBT-1RM estimates. Use a minimum of 5 sets at different loads, a maximum of 8. The rule of thumb is to try to decrease the velocity of your fastest rep by 0.05 m/s with each set. Low-intensity loads get 3 reps per set, moderate intensity loads get 2 reps, and higher intensity loads get 1-2. Here’s the example everyone in VBT screenshots:

From the SAME Flannigan & Jovanovic I keep linking

               Once you have those measurements, you’ll have to plug it into Jovanovic’s templates. He recently posted a step by step guide here, but you can also find it in his research review here. Here is a picture to demonstrate, but you should certainly check out Mladen’s article and his training blog.

Guess what? From the same article. Notice this profile is beyond 5 sets. Increasing
your sample size might yield better results. 

               It’s important when you do this that you do practice sets. This is different than warms ups. Practice a set at your starting weight and move it as fast as possible (again, maintaining sufficient technique). If you’re like most powerlifters, you don’t move fast for anything but cake. Practice sets will decrease the error you inject into the system. If you see your second set moving faster than your first set despite the fact it’s heavier, chances are you didn’t practice with enough sets. I prefer to keep the rest times insufferably high. Just like rest times for sprinters, weightlifters, or gym bros distracted by Instagram.

               The next thing is to conduct a MVT test, which is typically reps to failure. The general guideline is to use 75% 1RM or above. My general suggestion is that you use something you can crank out at least 9 reps. Let’s say at 85% you get an average of 0.04 m/s less with each successive rep. If your opening velocity is 0.45 m/s, that means your second would be 0.41 m/s, second at 0.37 m/s, and so on until you approach your MVT. My intuition tells me you can get closer to your actual MVT at 75% than you could at 85%. Alternatively, you could just do an AMRAP at both intensities and take an average of both mean velocities of the final reps. It might also be helpful to use a rep max calculator as a second estimate.

               If you don’t want to do an MVT test, you can use a generalized MVT of 0.15 m/s for bench and deadlift, or 0.30 m/s for the squat. I’ve said before that VBT essentially is snowflake training, so this probably misses the point of using VBT to begin with. Additionally, there is some danger with trying to use generalized velocities for assistance exercises. In my comparison article, I mentioned that I had issues with front squat load prescription. An additional issue I had with my front squat load prescription is that my MVT for the movement is 0.10 m/s higher than it is for my back squat. The result was always that my front squat load was over-prescribed. Mistakes were made.

               One thing I suggest is updating your MVT on the fly throughout your training. This will likely happen often enough that doing AMRAPs become less necessary. I would track when you conducted your L-V and AMRAPs/MVT tests so you know how out of date they are. Maybe you are setting new MVT’s regularly on your main lifts, but a newly adopted assistance lift might not get the same attention – so you might have to implement AMRAPs as necessary, which can be part of your normal training. 

               I do suggest converting an L-V table into a %1RM-V table once you have an acceptable estimate. I was skeptic about the stability of this, but decided to guinea pig the assumption after establishing a 12 kg PR on my bench. I kept the same %1RM-V table. Unlike front squats, it did not kill me ded. After finishing the cycle, I redid the L-V profile, but found no noticeable difference with %1RM-V profile. My estimates for load were off, but my %RM was spot on. Yes, this is a n=1 response, so take it with a grain of salt. All things considered, if your MVT lowers and your L-V profiles shifts with it, your %1RM-V profile should be sufficient. 

EARNING YOUR KEEP: YOU HAVE TO ACTUALLY TRY

               I’m obliged to tell you about the positive effects of feedback. There’s plenty of articles about feedback augmenting performance in team sports, but none of it applies as readily to powerlifting performance. I like VBT because it helps keep you honest. If the estimate says you should be able to move 80% 1RM at 0.43 m/s, in build-up sets you’re hitting 0.5 m/s at 70% 1RM, you need to kick it up. Maybe you're just really stressed, fatigued, or whathaveyou. But also, have you tried trying?

               There’s also the caveat that you could cheat yourself. If you want to cheat yourself for an individual competitor sport, I don’t think VBT is going to do any worse than applying the same level of effort using another programming method. You can cheat yourself with RPE, AMRAPs, or your percent based training. Seems a moot point. 

               One thing that isn’t as easily accomplished is loading more than you’re prescribed on the bar. If you load higher than your prescribed velocity allows, you’ll miss your target velocity.  VBT allows you an objective way to earn your keep - whether it's a higher load or more volume. If you want to lift a heavier weight that day, you have to objectively lift faster. If you want more reps, you have to fight to keep velocity across the set. Bryan Mann hits the point of feedback augmenting performance better than I could. His degree is in sports psychology if I remember right.

VELOCITY LOSS AND SET AVERAGE VELOCITY

               Velocity loss is the percent of velocity that is lost from the opening rep to the last rep. I’m not trying to insult your intelligence, I'm just trying to be painfully clear.

               Example: opening velocity of 0.43 m/s and an ending velocity of .27 m/s. This comes out to 37% velocity loss.

Relevance? A study by the Spanish VBT proponents looked at training effects from velocity loss. Training was auto-regulated, with subjects hitting reps within a set until they reached their respective training threshold (20% or 40%). You would think working close to failure would translate directly to strength sports, but that wasn’t the case. The group that terminated sets at 20% velocity loss has higher strength gains despite performing fewer repetitions. The group that performed reps until 40% velocity loss had more hypertrophy ended adaptations.

Limiting reps to 20% velocity loss isn't about limiting your volume. You can still get the same amount of volume, but you'd have to do more sets. I'm not going to beat around the bush: it's annoying. You can perform these as straight sets, stopping well below failure, or perform them using rest-pause/cluster sets. I'll cover this more in depth in a program design article.

If we take that at face value, there are two ways you can use this that I can think of. Either you can use it in real time or you can use the feedback after the fact. I’m not aware of a device that gives you velocity loss in real time (for no reason, it's an easy feature to implement), so the easiest way I’ve come up with is establishing a working weight off target velocity, then taking 80% (assuming we're going for 20% loss) of that as a stopping point. In this case, if your opening velocity was 0.43 m/s, an optimal STOP velocity would be 0.43 m/s * 0.80=0.34 m/s, giving you near 20% velocity loss. This takes a lot of focus because you really have to push through the lift as fast as possible, while making sure you’re not applying an undue stretch reflex (for bench press) or rebounding (squat), as well as making sure your form doesn’t fall apart. The fatigue is more on the neurological side than it is on the metabolic side, so it might not be something that’s as easy to feel out.

Using 20% is supported in the literature, but you'll find it's fairly restrictive and you're terminating sets well above failure. Another method I've considered with no support by the research is by determining velocity loss across your AMRAP set. If during your AMRAP you lose 55% of your opening velocity, than a more appropriate velocity loss threshold might be 50% for hypertrophy/work capacity work and 25-30% for strength work. Using this method or the original method might provide advantages for obtaining sufficient training volume, or it might not make a difference. I don't think it's necessary to confine ourselves strictly to what is only supported in the literature. The research is a reference point, not dogma. 

For systems that don’t give real time readouts or log too many ghost reps to really track velocity from rep to rep, you can just go off of delayed feedback. Complete a set halfway to failure, then see what the velocity loss is after completion. Then you can scale reps up or down to see what’s appropriate, but you'll have to adjust as you get fatigued and figure out if you have a fatigue wall you can overcome after which performance increases again. You could also use Mladen’s exertion tables, but in the words of Mike T, maybe that’s over-fitting.

The positive part of this method is it functions really well when you pair it with RPE if you’re working at high enough intensity. The downside of that is now you’re working with two systems of feedback and that could be confusing. This is a quick way to turn your training into rocket surgery. In my experience, Beast could give you too many ghost reps to make real-time feedback unfeasible. PUSH will give you real time measurement for some exercises, but don’t expect all your assistance movements to be supported. This could also be a benchmark for when you terminate your total number of sets. If you fail to achieve velocity loss in successive sets (it's alright to accept one set as a fluke), it might be time to move on or move to back off sets (decrease the load and finish off with a few more sets).

VBT is about quality of work, not necessarily limiting work. That said, it would be convenient if the amount of quality work was pretty high. Using this method on its own could limit your volume and limit you from achieving enough work to develop a solid training stimulus. You'll likely have to augment this method with something else, like cluster sets, back off sets, or some metric of baseline volume or number of lifts. The theoretical argument is something along the lines of minimum effective dose of training or maximum recoverable volume, but one practical application is intensity and number of lifts (INOL).

In the case of hypertrophy/work capacity, I think the more appropriate value to use might be set average velocity or a decline in opening velocity. There doesn’t appear to be any established protocol on this, so you’re sort of making educated guesses here. My general inclination has been to use the 10% rule. If opening velocity or set average drops from one set to the next by more than 10%, pack it in. The alternative is just using RPE in some way – which is my preferred way. If velocity loss remains the same, load remains the same, and RPE rises, then you should terminate the exercise or initiate volume protocols mentioned in the sidenote (the tangents annotated with smaller text). Since the point of back off sets is additional volume, my inclination is to say dropping the load and hitting the same number of reps for a fixed number of sets (1-3 sets), but that could be variable. I’m hoping someone releases a study or develops a better way to navigate this. Until then, it might be helpful to utilize different methods or find your own way.

THE UGLY BABY: THE DEADLIFT EXCEPTION

Velocity loss doesn’t work well for deadlifts, due to how erratic they respond. This output is from PUSH, but I've both seen and heard this trend  I’ve mentioned this before as the ugly baby, a technical term in the scientific world. No one wants to tell their best friend their baby is ugly, but they should hear it from someone that loves them first. As everything I’ve mentioned here, this might be my n=1 experience, but the impression I’ve gotten from everyone I’ve talked to that’s used a VBT device tends to be that deadlifts just act this way. I believe this is something that is indirectly reflected in the literature. Take for instance the last article I cited. They tested the squat and bench. These obey the rules of VBT very readily. They could have had a full power intervention, but instead chose to limit it to the exercises that comply. That says something to me.

Additionally, the literature and the manufacturers appear to employ two different protocols for 1RM testing: using the highest rep velocity within a given load or using the average of all mean velocities within a given load. Averaging essentially drops the magnitude of the fastest rep. Using the highest mean velocity rep in deadlifts seems to ignore the first rep, which is an issue because the first rep is the last rep for a 1RM. I can see the argument for usng either method, but so far my inclination is to use averaged sets and using a strict 5 sets x 3 reps sample across the different intensities. I haven’t had an issue with establishing MVT with reps to failure (if you ignore the fact that RTF with deadlifts is twice as horrible as it is for squats or deadlifts), but for VBT-1RM testing I’ve found the best results come from doing full resets to the floor. If you do full resets to the floor, you’ll still get some variation within set (and first rep to following reps), but it won’t attenuate velocity of the second rep due to stretch-reflex contribution or sacrificed momentum. I’m not advocating against touch and go (absolute zero fucks given on this argument), just that this is a test that struggles on its own. I’ve been able to get within 3 kg on my VBT-1RM estimate with a low standard error of the estimates (SEE) several times so far, but the VBT-1RM test is a learned skill. If you’re new to adopting VBT and plan to map out your %RM-V profile for deadlifts on a schedule, try to plan for an additional buffer day because it’s highly likely you’ll have to spend some extra time figuring out how to make VBT deadlifts work for you. If you plan to use an opening velocity as a gauge of intensity, it's probably best to use the second rep.

Again, this is all my n=1 experience. I’ve done my best to ask coaches that use VBT to figure out how they’re implementing it, and after two years of working at it, this is what I have. If anyone disagrees on this, I’d love to see the corroborating data and make the correction for the record.

AN INTRODUCTION TO VBT: OFF THE CUFF

               I’ll be honest, this is easier with PUSH since it has much of this built into the UI already, but you can make a spreadsheet with a little bit of forethought put into it. This isn’t the optimal way to do it, just the easiest way to do it.

             Week 1 is all testing. Do a set of reps to failure of each of the competition lifts at 70-85%. Write down the velocity of your slowest rep or MVT. Since this is all improv, you don't necessarily need to do a L-V/%RM-V table. This is designed to familiarize you with VBT. 

Let’s start off with a 4 day a week template: 2 days on, 1 off, 2 on, 2 off. It’s your typical MTThF routine. Monday is squats for intensity and bench for volume. Tuesday is bench for intensity and deadlift for volume. Thursday is bench for intensity and squats for volume. Friday is deadlifts for intensity and bench for volume. Add in whatever accessories you want, but make all the lift designations the competition lifts for simplicity’s sake. Run this set up for two or three weeks after completing your RTF sets.

The easy rule of thumb to follow here: High=MVT + 0.10 to 0.15 m/s. Low=MVT + 0.20 to 0.30 m/s. This will get you in the ball park of intensity that you need for a high intensity or low intensity load. So if your MVT for bench press is 0.12 m/s, on a high day you're started at 0.22 to 0.27 m/s. On a low day, that's 0.32 to 0.42 m/s That velocity is your opening velocity, or the velocity of the fastest rep when you're moving each rep with intent as fast as possible (while maintaining sufficient technique). 

Monday	Tuesday	Thursday	Friday
Squats (high)	Bench (high)	Bench (high)	Deadlift (high)
Bench (low)	Deadlift (low)	Squats (low)	Bench (low)
Accessories of choice (untracked)	Accessories	Accessories	Accessories

When you’re starting your actual work weeks, use build up sets to get you in the neighborhood of the target velocity, give or take 5% of that velocity. I tend to think of build ups as probes, only hitting them for 3 reps to be sure I’m not inducing any extraneous fatigue, but you might need a more thorough warm up. If you overshoot, adjust back down by a smaller load. Once you’re near target velocity, that’s your working weight – stay there. If your reps start getting close to your MVT, terminate the set. 

The back to back upper and lower days, if we’re thinking that way, is to create a training stimulus that purposely induces fatigue for the following day. It might seem like too much benching, but I tend to find that benching responds better to a higher frequency. As fatigue is induced, you should see some variation in the number of lifts you’re able to accomplish. I’m going to think in GZCL terms here and say your targets for high intensity are 10-20 total reps and 20-40 total reps for volume days. I generally worry about going under that range though, not so much about going over by a little. I would also use the second rep of your deadlift as the gauge to target velocity. Worst case scenario, since this is just an introduction, if you’re confused whatsoever, just use %1RM and keep an eye on velocity so you understand how it operates within a set and under fatigue. Worst case scenario, don’t do another rep if you think it could be close to failure ("/1 fail" in Juggernaut terms). How many sets you do is entirely up to you. I generally keep going until I cross the upper threshold of the range mentioned before, and have often exceeded it for curiosity’s sake.

Once you have that, you should watch how volume behaves from week to week, but you might need a longer sample to see any patterns. I’m not claiming this is the best program, it’s actually quite bad. I think it's important to have some sort of orientation to VBT before you actually undertake a specifically VBT program. The amount of work you have to put in on the front end with this style of "programming" is pretty minimal. You only have to test 3 lifts, there’s enough volume and consecutive upper/lower efforts to induce a good amount of fatigue to affect the next day’s lifts – requiring auto-regulation. If you don’t see this effect, back off on load after you reach a fatigue point, and add 1-3 back off sets to add some interference with follow on training. If using back off sets, just make sure you don’t get too close to MVT – which you can do by feel alone.

This could get really complicated and you can load 3 primary lifts and 6 assistance lifts that all require testing. There’s significant investment required out of VBT, both monetary and of time, but once you have that established it’s fairly easy to run from one program into the next with available data. Once you have a general idea of how VBT works and how you respond to feedback from VBT, you can use information in this short orientation period to feed forward into a more VBT oriented program. Just make sure you carry those MVT's you obtained from your RTF sets forward into the next cycle. Building an effective VBT program is mostly about maintaining momentum to mitigate the time eaten up by submaximal testing. Time spent on testing is time not spent on training.

At the start, you want 3 profiles for the main movements. The next cycle, you can profile 3-6 more movements. You now have profiles for 6-9 movements total. The next cycle you can change your exercise selection and add 1-4 more movements. At some point, you're going to have profiles for so many different movements and you'll only have to worry about updating old ones that are chronically out of date. 

So why not just train with VBT in exactly this way all the time? It's a stupid way to train, that's why. If you go strictly by the gosepl, and use the velocity ranges (0 - 0.5 m/s for absolute strength), you're probably not going to get enough variation in your training to really progress. There's also no range for hypertrophy in the classical Russian velocity ranges. But most importantly, it's too hard to really get organic variation of intensity and volume. You can make progress up to a point, but it really is helpful to designate velocity as a throttle to progression. If you designate the velocity, volume and intensity fall into line. Again: the above plan is pretty bad. You should have a better plan. This is not a year round training plan. Don't follow this plan any longer than you need to.

WRAPPING IT ALL UP

               I started with an obligatory speed-work explanation. I won’t pretend I have extensive experience with this, but if any conjugate method proponents want to add their two cents on VBT in speed-work they can pick up the slack. In order for me to have a qualified response to this, I’d have to run a mesocycle using conjugate and I think my time is better spent really getting a handle on things that would fill that gaps in VBT – like RPE.

               The key advantages of VBT are hard to be gleaned unless you run your program with a spreadsheet in MS Excel or Google Sheets. This should be a non-issue at this point, since most smartphones can support spreadsheets. Trying to get it simple and readable is another issue. The nuts and bolts of it, in the purest VBT way, is going to take preparation with load-velocity tables, VBT-1RM estimation, and repetitions to failure. This requires roughly the same input to a comparable system like RPE, but the same amount of testing. If you plan to use VBT to augment your training and not drive parts of it, then there’s significantly less work involved. Once you complete this initial test, you can easily understand exertion and intensity in VBT terms of opening velocity and velocity loss (or successive rep velocity decrement).

               Once you understand that feedback, you can tell when fatigue from previous days are affecting your neuromuscular readiness (intensity) or work capacity (exertion/number of reps). The point is not to avoid fatigue, but help gauge it. Once we can gauge it, we can increase or decrease load to induce the training adaptation we want. If the training load is not sufficient, we can increase it by using back off sets.

There is also another table called the load-exertion table, but I chose not to include it and simply adhered to the velocity loss rules. I will mention that it might be very useful if you just want to stay a designated number of reps away from failure instead of following velocity loss designations, but I never got that granular.

For some reason, I chose to inject a rant about built in 1RM testing, but I should be clear that I think this is a feature I think all manufacturers should include in their UI. Probably this, load-velocity/%RM-velocity mapping, MVT designation by exercise (or allows users to conduct reps to failure to change MVT), and automatic calculation of the percentage of velocity loss. If it had the ability to use AMRAP sets to determine load-exertion tables, I'd probably be more apt to use them. If all these features were built in, VBT would be so much more accessible. Instead, many devices are just following transliteration of the force-velocity curve.

I also included a part about feedback and its role in sports psychology. Then I promptly passed the buck onto an expert.

The last section gave a very poor sample VBT program, meant to introduce people that wanted to test drive VBT to follow. The next article will (might? I’m a wordy bastard) be brief. It’s just sample programs I made. It should be more content than it is words. Special shout out to Bryce Lewis of The Strength Athlete for making some great videos on how to become a powerlifting excel wizard and Cody Lefever since I essentially started by stealing his template design and vocabulary.