Advanced Strength & Power Concepts For Combat Sports w/ Dr. Greg Haff (SSOF Ep. 39)

Dr. Greg Haff0:34

My wife probably is one of those people that I'd put in that category. So really, we spend all our time just trying to figure out how the body works and how to make it work better.

James de Lacey3:07

Now that's awesome. That sounds like an interesting project too. When you're talking AI and machine learning, are we going past, like we're talking past the typical velocity based training apps kind of thing and prescribing maybe loads based on what's happening there, or are we talking something completely different?

Dr. Greg Haff3:07

That's part of the question, and the problem that we have right now is that we can't find any velocity devices that are reliable enough to actually do what we want to do or accurate enough. What we find is that from a velocity measurement perspective, the construct that force and velocity are intimately related is a fascinating one because it makes so much sense. We know that in the weight room, if you lift a heavy thing, it's going to move slower. The question that we have to ask then is how accurate are those metrics for prescribing training load?

And right now you have two camps, really. You have one camp that's like, let's just ditch everything we know about strength training and just go full on in with velocity. And then you have the other camp that's like, whoa, slow down, this is too much too fast. And the reality is, this is not a new question. This question's been around since the 70s when the Russians first started using linear position transducers in weightlifting, and they found that when you did that, you didn't get as strong as you did when you went off of percentages.

And then the Americans tried it in the late 1970s with Carl Miller and found basically the same thing. The Germans tried it in the 90s with the VS scope and found the same thing. We tried it in Professor Stone's lab in the early 90s, found the same thing. And it's just kind of this rehashing of the same problem. The reality is that velocity will change before strength changes. And the people that promote velocity based training are really saying that strength is this wildly changing construct that from day to day it's going to vary really dramatically.

We have a series of papers right now that we're in the process of completing. We have one that's already published that looks at the one rep max on a Monday, Wednesday, Friday, and it's pretty bang on around five kilos error rate, which is natural variance. So a lot of people questioned us on that and said, why just a one rep max? That's not that big a deal. There's not a lot of volume. So we decided to go—and that was in the back squat—and we said, all right, fine.

So we've got a series of studies with my PhD student Stuart Guppy, who is looking at the 3RM deadlift, which we know people don't like to deadlift to max multiple times a week. And we did that three days, Monday, Wednesday, Friday, and bang on, doesn't change, much less than 5%, maybe 6%. Same thing, we said, all right, 3RM, okay fine. So we went to 6RM, bang, same thing, about 5%, 6%.

And it falls right in line with what Professor Stone started teaching in the 1970s, early 80s, that light-heavy days earlier in the week you can go heavier, drop 5 to 10% for the end of the week, you'll account for fatigue. So the reality is strength's not that variable, but velocity is all over the shop. So when you start to use velocity metrics to predict load, it becomes problematic because are you actually getting a load that's actually related to the actual athlete's capacity?

Now, you go out to the Google Earth view, so to speak, and you think about it in the context of overtraining. And you look at the seminal work of Professor Andy Fry from the 90s. What you see is that overtraining in the progression of overtraining, that strength is the last factor to actually fall off. Rate of force development is first—well, actually psychological variables are first. You start to feel like crap.

And then you start to see changes in rate of force development, which then show you changes in movement velocity. So we would expect movement velocity to move earlier than strength, but everybody seems to disregard that. In fact, we did a study in the early 2000s where we looked at sleep deprivation in weightlifters, and we found that even 48 hours of sleep, people could still hit their maximums. They felt horrible doing it, but they could actually do it.

So for us, what we've been trying to do is figure out what's the best way to use velocity measurement devices, because we need inputs to create our AI models for training. So we've been doing a lot of validation work on different devices. We've been working with different companies. We've been looking at how does this change in relation to strength, because we need inputs. If we don't have inputs, we can't create the actual programmatic structures that we want and have machine learning and AI kind of work from there.

Obviously we can easily measure psychological factors. We can measure sleep, we can measure heart rate variability, we can measure resting heart rate. There's tons of devices that do that quite accurately. But to be honest, in the weight room there isn't much that actually works on a reliable and really accurate way. Most of the validation studies for a lot of the devices that are used by proponents of velocity based training are done in the Smith machine. They're done with a pause between the eccentric and concentric phase.

In their papers, they actually state they do that to improve the reliability. Well, not many people A, train in a Smith machine, and B, pause between the eccentric and concentric phase of the lift on a regular basis. Now, in competition a powerlifter will do that on a bench press, but fundamentally, if you are a proponent of ground-based weightlifting training, which most of us are, that's not how you train with the weightlifting motions. You're always moving as fast as possible.

So from there, I think that's where we're spending a lot of time at the moment. And the reason we do that is because we think it has defense applications. Because when we look at, for example, soldiers, special forces operators, they have very variable schedules with training that has a multitude of factors from high-intensity interval training to long-distance cardiovascular to high power output to high force. And how do you get all that to work together to optimize that individual and keep them optimized on a regular basis?

So it's kind of a package we're trying to unpack. Started looking at some of these factors with Duncan French at UFC, and we spent a fair bit of time there prior to COVID, which was really kind of fun, whiteboarding and trying to figure out the idea of emphasis periodization and how that would relate to MMA but also to defense applications. And so for us, that's kind of where we're at.

Right now is to try to figure out how to use technology to vary training within an emphasis model for sports that require a multitude of physiological factors to be at a high level. Because it's easy to do weightlifting. I mean, like, weightlifting coaching is pretty simple. I mean, I know my wife and purist weightlifting coaches might question that, but it's pretty easy. Master techniques, get strong as humanly possible, and it's pretty easy. Just takes time and developing work capacity.

When you start to put in multiple factors, it gets a little bit more tricky.

James de Lacey10:19

Yeah, for sure. That's funny how you mentioned the I guess the ebb and flow of velocity-based training coming and going. You see it now again within weightlifting, within the training halls, a lot of people using various devices—V-Max Pros, Gym Wares, whatever. I guess maybe it's a track they training through there, but what is needed for these devices to become more accurate?

Dr. Greg Haff10:19

So linear position transducers, they measure displacement, so they're probably the most accurate, but they're also the most expensive. And the question that you have to have is what are you going to use them for? Are you using them for motivation, or are you using them for programming?

From a motivation perspective, yeah, you can set target velocities. But in weightlifting, what we find is that people change their technique to get a higher velocity and they become less efficient. I could put a really efficient weightlifter on a velocity meter and they're not going to look as good as a really inefficient person as far as movement velocity goes. We saw several teams in Australia that tried to use velocity meters and their technique went sideways.

So for me, when I start to look at how we use them, we think that they may have application in season when we're trying to balance fatigue. Because remember, if we look at our overtraining spectrum, velocity is going to be one of the early things to go.

And so what we suggest is that we use, what we call, velocity-based training with guard rails. So thinking of it in this capacity: set our training loads in traditional ways based on percentages, percentage ranges. I'm a big fan of Mike Stone's system where we set 5% ranges and let the athlete auto-regulate within that range. Set your ranges there and then use the velocity meter to kind of either raise or lower the intensity for a band of velocity that fits in that range.

I'm not a proponent of this idea that we're just going to do as many reps as we can until we get a certain velocity loss. There's a recent paper that came out out of Spain that suggests that we should not ever use percentages and that velocity—just program velocity percent losses. So let's say I'm going to have a 20% loss in velocity. They said from rep one, but the literature would say from rep two. And do as many reps as you can until you get to 20%.

And as a lifter, I'm like, that makes no sense because I know the more reps I do, the more fatiguing it gets. So for me, set the rep range, make it static, adjustable, go to your intensity based on what you've programmed with your percentages, and then use velocity maybe in season. That's where I think we might have application or as a motivation tool where we have, for example, our rugby players come in and we want them to do a bench throw and make a contest on who can get the highest velocity or the highest power output. From there, I think that would work.

But I'm not ready to make the leap into full-on, let's program everything off velocity. I think the data is too variable and it's innate problems with the devices. Like, for example, IMUs and accelerometers are really rubbish at determining the beginning and the end of a lift.

James de Lacey13:14

This is why I found that, yeah.

Dr. Greg Haff13:14

If you look at the research to make it reliable, they artificially put in constraints of pauses at the beginning of the rep and at the middle of the rep and at the end of the rep. And people just don't train that way. That's not how we maximize strength. So that's been the issue, and I think I'm not sure you're going to ever get that out.

Now they're trying to use artificial intelligence, machine learning, to try to smooth out some of that error, but I just don't think it's going to work. Now, the linear position transducers has been around for a long, long, long time. I just think that we know people who have been strong or are strong know that you just kind of got to grind sometimes. And you can't overcome will to do something.

And if you look at, for example, Special Forces operators, they do things that no human would want to do, but they do it and do it well. And they do things in the gym, in the weight room, that you just kind of scratch your head going, how did they just do that? Me, we have studies that we've done where people are doing 90-kilo squats for sets of 80. I mean, that's insane.

And so when we start to look at this, I'm not sure you can get it to work. And Professor Stone and I have talked about this a lot. In the '90s, the big rage was optimal load training. So you're going to find your optimal power output, you're going to train at it, and people tried it. Yeah, it works for a little while, but then you tend to lose strength and then power goes down.

So I think if you're always optimizing on velocity or power—because really, what is power training? It's velocity training, it's just packaged in a different way. If you know that force times velocity equals power, you turn out power, it's velocity training. And so I think with some exercises, yeah, it's necessary to train power, so to speak. But I'm not sure we can ever get the devices to work.

I mean, because think about it: if you had to have a linear position transducer for every single device, every exercise that you want to perform, that's somewhat problematic. The other issue that you run into is that you've got to create a load velocity profile for every individual and every exercise you want to use it with.

If I'm an American strength coach who works in football and I've got a hundred football players, that's a hundred people I got to test, and I'm probably going to have a hundred exercises. All I'll be doing for the rest of my life is testing. And there's been some work by Jonathan Weakley and his colleagues and my former doctoral student Harry Banyard who have suggested that one-RM testing is time-consuming and da-da-da-da-da. Yet they propose to do velocity-based profiling by doing a test for one-RM and then coming back three days later and doing another one-RM test. Well, why don't I just do one-RM?

Again, the practicality of all this is difficult. I think there's a rush to put technology into training, and I think it's good. I mean, I think we need to have data, but I'm not sure how we use it at the moment is the best way to go.

James de Lacey16:47

Interesting, interesting take. I kind of wanted to dive into as well the idea of programming off one-RM. Obviously, as you mentioned, there's a shift. People program one-RM, people program velocity, people not using anything, just using RPE. I guess, as you mentioned, obviously within the week that you might take 5% off near the end of the week to I guess account for fatigue. So how do you go about programming then typically? Is it just kind of that one-RM later in the week you take around 5% off, or give a range? And then do you ever use RPE within that as well?

Dr. Greg Haff17:16

Okay, so there's a good one to unpack. So programming, the classic weightlifting way is to, you've got your snatch and your clean and jerk and your squat one, and then you program it. And you look at the repetition to intensity, basically, relationship. And obviously the heavier the load, the less reps you can do. So that's the way a lot of people do it.

The way I'm probably more inclined to do it is what we would say off the x-RM. So this is the classic way that Professor Stone really came up with in the early '80s at the NSCA research center at Auburn and at LSU. And, wildly successful programming style.

So what you do is you create bands. So it's like, let's say we're going to have a moderate day, and that's 80 to 85% of RM. So if I'm working on singles, then it's 80 to 85% of a one-RM. If I'm working on doubles, it's 80 to 85% of a two-RM. And use a basically a sliding prediction.

And then we do that regardless of the rep range. So it creates a band of intensity, and then we can vary that band of intensity. So let's say I have a moderately light day. So that would be, what, 70 to 75% of the RM.

Let's say the first day of the week is moderate. Let's say Monday is our squatting day and we're going to have a moderate day, so we'll train that. Thursday would be a moderately light day, so we've got that built-in change. So we create that light-heavy conundrum. And if you think about it, it really fits nicely with some of the work of Verkhoshansky with the idea of polar training where you have

High intensity days and low intensity days, and Louis Simmons made that quite famous with some of the things that he did in powerlifting. But it's basically the same construct. We're just using off of the RMs.

Now that doesn't mean that you have to always test the one RM. You could fold in an RM into the training program. So let's say I'm doing sets of five in the back squat. Well, I could make it a max day and just go to RM fail, and then I can recalibrate all my other training loads with predictions.

Now people are like, well, but the predictions, and people can do different repetition ranges and things like that. I think as a coach you start to figure that out. You can actually say, all right, my athlete doesn't always hit the same target, so we can adjust the load. Because again, I'm a firm believer that while there is some mathematics to training and training theory, it's also about coaching and understanding your athletes.

I use the analogy, and this will tie in with the RPE, we've got two types of athletes. We've got what I would call a marshmallow and a brick. So the marshmallow is all soft and squishy and is going to quit on everything and everything is really difficult and I'm never going to push themselves. I've got to hit them with the stick to get them to go.

And then I've got the brick, is that individual that no matter what I give them, they're going as hard as they can and they're going to push themselves too far. And I'm going to be constantly pulling them back. And so when I start to look at this construct, when I have those type of people, I've got to understand who they are and know how they move and how they lift the weight and how they operate.

And then I can start to figure out, okay, they can lift at a higher percentage of their maximum for more reps. And it also has to do a little bit with their physiology.

I had a track cyclist that I worked with from Canada, and he could do almost 90% of his one RM in the back squat for ten. He had very low end one RM, but the problem for him wasn't his legs, it was his core. His core couldn't handle the loading for heavy singles. And so that was something we had to adapt. Obviously he would be somewhat underloaded with high volume if he has such a high 10 RM, but we could actually adjust it.

And Christian Thibaudeau out of Canada has a wonderful table where he's got people who are high fast twitch percentages for 1 RM, low fast twitch, high slow twitch, and then kind of a mixed. And so we have the same thing, a similar table within our new book that's coming out next year.

But how does this relate to RPE? Well, when we start training RPE, one person's pain is another person's pleasure. And some people push really hard and will rate it low. I remember when my wife worked at the West Coast Fever with Norma Plummer. I mean, there was one netball player there, if I swear if she had a baby fall out of her, she'd rate it a two. And nothing was ever above a two when we would do the monitoring and help Aaron with that. And so if she jumped to a three, my wife was in panic mode like, what's going on, what's wrong with her?

But then there was other people that come out of the change room and they're rating it a nine just walking out. So for me, RPE is a tool, but I think I've got to contextualize it to the individual and understand their peculiarities in the sense that, I know when I work with military people, they're on that brick side. And I mean, pulling them back is difficult because there's an ethos there that if they don't do it then they're letting their teammates down. And also the type of people that attract to that.

So I've got to understand that. So do I use RPE? Yeah, I use it as a monitoring tool and I may use it with loading in some situations, but I've got to ground it. My wife kind of did, she was trying to figure a way to make it work. And what she did was like on an 80% day, she'd say, all right, this is an eight, okay? Feel what that feels like and kind of give it a feeling.

Because what we know with RPE, it's very subjective. And the original Borg scale, the 6 to 20 scale, was based on heart rate. So you would basically take someone on a treadmill and say, all right, you're going to a 12. So you'd take them up till their heart rate was 120 and you'd ground it there. And so you could use that as a guide.

In the weight room it's a little bit different. No one, that I know of, has come up with a way to ground it. Now I know the group out of Pittsburgh, Robinson, had done a lot of work on this, but most of it's in untrained people. And it's not people at the pointy end of the spear of strength development, and so that's a problem.

Now the other issue that I think is really interesting at the moment is repetitions in reserve is kind of a hot thing. And for me, I'm 53 years old now. I've been lifting weights since I was 11 on a regular basis. I can't tell you how many reps I have left in the tank until I'm done. And I've been in the weight room probably more hours and days than some people have been alive.

And so I've talked with Professor Stone about this. We don't even conceptualize how you can actually design a study that isn't confounded with that, in the sense that, and we've been trying for the last five years to come up with an idea and we're just stuck. Maybe with COVID gone we might have an opportunity to get in a room and figure it out, but we haven't been able to do it virtually.

So if you look at repetitions in reserve, the idea is you're going to lift until you have X amount of reps left in the tank and that's going to be your intensity. And it was made famous out of powerlifting and rather simple movements. Well, in the weightlifting movements you can just toss it because who knows? Because technique will break. And oftentimes with weightlifting, it's not the strength that's the problem, it's the technique breaks. And as far as the full lifts go, unless you're dealing with really highly developed athletes, then it becomes a strength issue. Strength is the underpinning factor.

When I think about it from a repetitions in reserve perspective to create a study to actually test it, you've got one or two things. You got to ask them before they do the set how many reps they think they can complete and then they do it. The issue there becomes, now they've psychologically predisposed themselves or biased the study. If you ask them in the middle of the set, then you distract them from actually performing and they're thinking about something other than performing.

If you pause in the middle of the set to measure it, then you're creating a cluster set, which then changes the whole conundrum. So I'm actually struggling with how you could actually do a repetitions in reserve. We tried to do something with Amador García-Ramos with velocity and could we use velocity declines to predict how many reps you had left. And no, didn't work, too much variability.

So for me, when you start to look at psychology, resilience is an interesting phenomenon. And in the weight room, I'm a firm believer that strength training builds resilience when it's done appropriately because you have to learn how to fight. And when you have people that are really resilient, I use special forces operators because I think they're really quite on that really extreme spectrum. They have a way to do things and push through that a lot of people don't have. And your super elite athletes often have this as well.

But when you look at novices, they don't know what hard is or they don't know where their limits are. And in talking to people that go through extreme sporting events, they don't know what their limit, how far they can push themselves until they actually push themselves. I think David Goggins is a great example of someone who tries to find where his limits are.

I mean, he's a little extreme, but to be honest, I've worked with a lot of special forces operators that are in that same spectrum. That is not uncommon. He's just a little bit different. But his resiliency is a lot different than other people.

So for me, the psychological factors, do I measure them as a strength coach? Absolutely. Because it gives me evidence that I can use to create a model of my athlete and couple it with what I know about my athlete. So for me, that's how I would use it. And obviously we use it to calculate training load, arbitrary load.

The biggest thing that I know of, no studies, and we've wanted to do it for a very long time, and we tried but one of the grad students messed it up, was does volume load monotony and strain correspond to arbitrary load from session RPE monotony and strain? We've never been able to, we've never measured it. We're assuming that the mechanical work corresponds, but we've never actually tested it.

Obviously external load influences internal load, external load being the volume load of lifting, internal load being the RPE-based training load. We assume that those are the same, but are they the same? I don't know. And I think that's a study that needs to be done down the line.

We did do a study in rugby that we haven't published because we don't have the statistical power, where we compared training on RPE versus percentage-based training. And what we found was that your personality type related to it, so your personality constructs dictated how effective either training strategy was. And what we did find generally is people with session RPE, they're either overtrained or undertrained. And the percentage-based people, we could control the inputs a little bit better.

So from that perspective, I mean that was in rugby players. And we found that was an interesting thing. We just don't have the power to say confidently that the findings are what they are. We intend to redo the study, it's just since COVID it's more difficult to do training studies. People aren't really interested in participating in things anymore. You'd be surprised, we pay people and they still don't want to do stuff.

So from my perspective, yeah, that's how I use RPE.

James de Lacey28:51

Nice. I like that. And while we're on the topic of weightlifting as well, weightlifting derivatives, I know you're deep in that world too. So I wanted to maybe just have you explain to the audience why use derivatives over the traditional Olympic lifts, that's from a sporting context, and are there any advantages or disadvantages of doing so?

Dr. Greg Haff29:20

This is a great question. The first thing I have to give Dr. Tim Suchomel a hard time, because we were using derivatives when he was still in nappies. So he's not the founder of this. Professor Stone was doing this.

You can look back in Strength and Health, and York Barbell was doing derivatives back in the '50s, so it's not a new conundrum. Derivatives are a great tool. Now the way I've always looked at it is that derivatives gives me the ability to overload the lift.

And for example, pulling in the clean, the limiter is usually the catch and standing. The pull is usually not always the problem. So to overload the pull, I can pull more than I can clean. So I can use that to create strength, and we get our triple extension, and that triple extension has translation to sporting performance.

The issue that you have is that I don't think it's a one or the other equation. I think it's a blending of the two. I think you need to catch as well. And we have this wonderful study that's under review at the moment that I worked with the German University and the German national weightlifting team to do a study. And we figured out a way to actually use EMG to look at how the core muscles contract when you catch for bracing.

And we compare derivatives to catching, and what we're finding is that there is a unique activation pattern of the core muscles to stabilize the spine on reception that is very different than if you are doing derivatives. So when I look at derivatives, I think of them as tools for developing aspects of the lifts. And for non-weightlifters, they're very useful.

In fact, when I was a lifter, we spent majority of our time in Professor Stone's group doing derivatives and a minimal amount of time on the full lifts. Now I think we were a little bit too extreme in that regard because timing of your lifts will go off if you don't do enough of them. But actually I think many people in weightlifting probably do too many of the full lifts and not enough of the pulling and the pressing and the squatting movements, because really that's the engine builder.

I think my wife actually explained it one day to me in a way that made me think a lot. She said the snatch and the clean and jerk are the technique work, it's your sport. The derivatives, the pulling, the squatting, and the pressing, is your strength training for your sport. And I think that's a good way of looking at it.

So for non-weightlifters, yeah, if you have somebody who technically can't perform a clean, let's say you've got a rugby player who's got a wrist problem where they just can't get into the catch position, well the derivatives is a fine alternative. You're going to get a lot of the benefit of the triple extension. In fact, as a weightlifter, I would argue that all weightlifters should be doing those exercises, and I know a few that don't.

So from that perspective, I think it gives you more tools in your toolbox so that you can actually plug and play things so that you can target training adaptations. It's not so much always the exercise, it's the adaptation that you're trying to get. So we use those a lot.

And think about it, there's some wonderful work from Dan Cleather and others that shows that the push press corresponds or translates or has dynamic correspondence with vertical jumping. So if I'm working with AFL players, Australian Rules Football players, push press is a great exercise for them. So things like that, I think it's trying to find what exercise is going to fit with what you're trying to get.

So to sum it up, derivatives, great tool, use them. They should be programmed heavier probably than the full lifts if you do them well. But you don't want to sacrifice technique just to lift load. I think that's where some people make big mistakes.

For example, on a clean pull, and this is where Professor Stone and I get into some really hot debate to be honest, I don't think you really would pull more than 120% of your clean. If you're capable of pulling more than 120% of your clean, then your clean technique is inefficient and you've got bigger issues. And so from a weightlifting perspective, that's how I look at it.

Now the big issue that we have right now is, and I've actually asked Paul Comfort this, who's my very good friend and one of my most cherished colleagues, and Tim Suchomel this as well, if you can't catch, how do you program it? Because all the programming is based off of your full lift. If you can't catch, how do you program it? And there isn't a good answer to this.

This may be where velocity is where we could get something. Set a velocity band and make it work. There's a group that's proposed using body weight, which I think is probably the dumbest idea I've ever heard, because that totally forgets the idea that people have different strength-to-weight ratios. So just programming derivatives off of a percentage of your body weight makes zero sense to me.

So that's an issue that with the derivatives, if you don't have mastery of the actual catching, how do you program it? And right now the only way I could think of is with velocity, and that maybe we're velocity meters have a use. But we have to be cognizant of the fact that they are variable and or inaccurate at times.

James de Lacey35:05

I know Mona falls into the same camp as you regarding the percentages for pulls and things above the clean. I know for her preparations and things, she doesn't pull that much heavier above her clean and snatch.

Dr. Greg Haff35:05

She doesn't pull. If you look at the Russian literature, they suggest 90 to 110% is what you would train off of, because that has the best transference. And we track that way back, and there's a position stance on weightlifting coming from the NSCA that we were all co-authors on.

Professor Stone may go as high as 140%, but where I would go that high would be when I'm pulling from blocks at the knee or the thigh, which obviously you've taken out some of the sticking points. But fundamentally, I think if you go that heavy that often, you become too much on that slow strength area, which that high force zone of the force-velocity relationship, which is important, don't get me wrong, I'm not discounting that.

But to get the transition to the actual full lifts, I think you got to be somewhere in that 90 to 110% is where it's going to be. You might push it as high as 115 or 120 if your athlete is quite proficient. But that's pretty well documented now.

There is a neat paper by Urs Granacher that talks about heavier pulls at the first pull just to get the lift off more effective, and he's gone as high as 130. But that's a very finite movement. It's almost the question that I always ask, and this is part of the debate that Professor Stone and I had, is when is a pull end and a deadlift begin?

Because really if you take it heavy enough, it becomes a clean deadlift, it's no longer a clean pull. So when is that going to happen? What speed does that happen at? And I don't know if we have a good feel for that. And so, what are you using the pull for? Is it simply to develop strength, or is it to develop speed strength or strength speed? All those things, you've got to make those decisions.

James de Lacey36:55

So how then would you... I know you mentioned obviously you need to know, to grow percentages you need to know your clean or snatch, but

Is there a way for, for example, an MMA fighter listening to this where they wanted to use some kind of Olympic weightlifting derivative that they would be able to figure out the loading? Or would you be like, okay, you need to kind of look at velocity or at least something to go off?

Dr. Greg Haff37:18

If they've got no catch, then I'd probably use velocity as some sort of barometer just to get an idea. I think you might be able to predict it off of a squat or a clean. I mean, a squat or a front squat, but it's going to be rough because we know we have relationships. We could back mathematically calculate—we know there's relationships between the squat and the front squat and the snatch and the clean.

So you could say, all right, roughly your squat is like 120% of your clean. So if you squat 120 kilos, then your best clean is 100, figure from there, and that might get you in the ballpark. And again, I don't think you have to be spot-on to get benefit. I think if you're in the ballpark, you're going to figure it out.

It's not a one-for-one kind of relationship where it has to be perfect. I think the body's going to adapt to stimulus and we'll figure it out eventually. And I think velocity actually could be used there. I don't know if anybody's done it with the research study, so it's kind of out there. It's just ideas that we're playing with in our lab group at the moment.

James de Lacey38:37

Gotcha. And you also mentioned cluster sets earlier. I'm a fan of cluster sets. I know you've done—I think you've probably done the best review or analysis on, or published the best one on cluster sets. I think you got another one coming out? You had one come out recently?

Dr. Greg Haff38:37

My doctoral student Yoshi Nagatani just came out with—he's got two. He's got one in SCJ, Strength and Conditioning Journal, from the NSCA, but I actually think the better one's actually in the UKSCA journal, the Professional Strength and Conditioning Journal. I actually like the second one better.

Yeah, we've been playing with cluster sets for a long time. I mean, I remember when I was lifting in the 90s, Professor Stone used to do them with me quite frequently. And a lot of the stuff that we research—stuff we've done in training or tried in training that we've gotten from different sources around the world. Cluster sets are really fascinating. It's not a new construct at all. We've tracked it back to 1924, and it's been around for some time.

I've been working with Jan Todd at the Stark Center. We're trying to find the origins of certain kind of things. I was supposed to go to University of Texas at Austin around the time COVID hit to spend a couple weeks at the Stark Library tracking a couple things back. But they had a student who worked with me a little bit on this, and what we found was that with the cluster set, Perry Rader talked about it in Iron Man magazine in the 1950s based on some work from the 1920s.

And it was really in the bodybuilding spectrum, and it was called the rest-pause method basically. You do a few reps, rest or pause, and then you do a few more reps. So we became interested in it from a weightlifting perspective in the 90s, and we did our first study in the early 2000s.

It was a really interesting project. And we used velocity tools for our measurements. We actually used the vector scope out of Israel, which was probably the best velocity measurement device ever made. It was really accurate. It was really good for weightlifting. I mean, you actually got a bar trajectory instantaneously that was accurate. But it was $155,000. It was unbelievable, but we used it quite a lot.

So yeah, cluster sets are really interesting. There's a lot of work that still needs to be done in them. Yoshi, my doctoral student, is going to be doing a series of projects with cluster sets for his PhD. He did a series for his master's degree under me, and he's really quite fascinated with it. So yeah, I mean we could talk a lot about cluster sets.

James de Lacey41:01

Let's try to keep this practical. Do you have any favorite set-rep structures, maybe to develop the strength side versus maybe the velocity or power side?

Dr. Greg Haff41:31

That's a good question. So I'll start on the conditioning side. So we use them a lot for conditioning in the strength-endurance block. So one of the ones I've been using a lot lately is a little bit different cluster structure.

So typically a cluster set, you'd either keep the reps the same throughout the series. So let's say I'm doing a set of 10. I could do a rep, rest 20 seconds, and then accumulate 10 reps. And basically that short rest in between repetitions allows you to have partial recovery of phosphocreatine so that you can maintain the quality of the repetitions across the series.

Now, where most of the research I think is flawed significantly is they equalize the training load between the traditional and the cluster set. One of the main benefits of the cluster set is you can use a higher load, and then you can accumulate more work, which means you can accumulate more adaptation.

So for example, a traditional cluster set would be single reps. You could modify that. You could do five pairs for a total of 10 reps in a strength-endurance phase, or you could adjust the reps and not do 10. You could do 12 reps to do four sets of three or three sets of four, depending on the physiological outcome that you want.

Obviously, the more reps you do in the cluster, the more fatigue you're going to create. So that could—if that's what your aim is, that's what you're going to do. What we've been playing with more recently is ascending clusters where the load changes from rep to rep.

And so let's say I'm going to do a set of five. I may have 70% on rep one, 73 on rep two, 76 on rep three, 78 on rep four, 80 on rep five, and so the load goes up across the set. And then the next set, what we see potentially is a higher velocity at the lower load. So we get PAP built into the set structure.

And so that was Yoshi's thesis. We didn't load it correctly and we didn't really see the actual percentage, because the loading is the issue. When you start to do a study, you got to make it comparable to the traditional set, so the average loading gets a little wonky. But we found that it works pretty good.

Now, where we've been shifting now is we've been doing ascending clusters but changing the rep scheme. So ascending load, descending repetition. So we do 4-3-2-1 for a set of 10. And that's more on that conditioning end of the spectrum.

Okay, so we do that. Now for strength development, I tend to go more with the individual repetitions, maybe up to two reps in a cluster personally. And I try to load them a lot higher than most people.

I would train it off the number of reps in the cluster, not the number of reps in the set, if that makes sense. So if I'm doing a set of five, instead of going off moderately heavy off of a set of five—let's say that's 85 to 90%—I would go off of the number of reps that I have in the cluster. So if it's singles, then I'm going to go 85% off of a single on each repetition. So the loading is relatively the same, but I'm getting more work in.

Now, depending on how you want to do it, the more strength or power focus, the longer the rest between the cluster needs to be. Obviously, I don't usually go above 30 to 40 seconds because at that point it becomes short rest interval training and it's not really a cluster set.

And so that's kind of how we would do it. So we use ascending, we use undulating where we go almost a pyramidal fashion where we go up and down, and the back end is potentiated. And another one that Yoshi's talked about, but we haven't done, is descending. So warm up to a high set, and then the cluster set is reducing loads. And we've talked about that, so get that big stimulus on the first rep and then more speed as we go across, but we just haven't done the work on that yet.

James de Lacey45:25

So would the theory on that last one be almost like to potentiate the speed side of that cluster?

Dr. Greg Haff45:54

Yep. Yeah. The whole theory is based on the paper that we did with Professor Stone at the USOC with the weightlifters up there, where we did down sets and we found that we got a PAP effect in the down sets. It's where we started to work on strength-power potentiating complexes.

So the idea there is, could we create PAP or speed-based training within the actual set structure and get loading as well for more efficiency? So that may work. We're not sure, but we've got to try it. In the lab, we often try it in training first.

James de Lacey45:54

Then, are there any protocols that you'd like for speed and power development? I'm assuming maybe we're looking at doing various jumps with the clusters or something like that?

Dr. Greg Haff46:20

Yeah. For me, this is an interesting one, because we've—

We've done a fair bit of work with the cluster set in the squat, and I actually really don't like doing cluster sets with the squat because the problem that you have is you have to rack it, unrack it, and that creates fatigue unless you have a monolith, like a Rogue Monolith or something like that where you just stand up and then squat. Because then you've got to account for the amount of fatigue that you're creating going back and forth in the squat rack, so you've got to lengthen the rest interval a little bit.

For me, I think it's more about the combination of the exercise you choose to do with it and then the rep and set scheme that you put into the cluster set. So if I want to build power, I will use it with my power clean or my pulling derivatives or I guess I could use it with a jump squat or something like that if I chose to. And then how does that load—what is the load that I'm going to use and how am I going to make that stimulus relate to what I'm trying to accomplish?

So for me, with that cluster set, first thing I'm going to do is choose the exercise. I tend to default to ground-based type activities, so snatching, cleaning, power cleaning, power snatching from the floor or blocks and my pulling derivatives. Now what I would then do is then figure out: okay, what block of training am I in? Am I in a strength block, a power block, or speed-strength, strength-speed? And what's my repetition scheme that I'm actually going to implement?

And so let's say I'm in a strength block, I'm going to do fives. Then okay, I'm going to either do singles or three cluster set to put a split there. And then it becomes the loading—what is my loading? Now I can use ascending clusters to develop high-end strength under fatigue. I can use descending clusters for PAP. I could use an undulating for PAP, so that would be more in that speed-strength side or strength-speed. If I'm trying to build just brute force, then I might keep it the same at some target load.

Now the other thing that I could do then is if I'm going to go into a purely speed-strength, I might only do triples. I may keep the load light and work really high on speed, and that could be in a jump squat or something like that. So it's about blending the two programming structures.

James de Lacey49:05

Gotcha. And you've mentioned a lot here on PAP or post-activation potentiation, obviously depending on the different camps. So you have post-activation potentiation enhancement now. Obviously there's always—I think someone published a paper just on the terminology recently as well. But regardless of the terminology, post-activation potentiation, do you want to maybe dive into: okay, what it is and different ways to go about it within a training program that anyone can literally listen to this and start to implement within their training?

Dr. Greg Haff49:39

So post-activation potentiation enhancement—I think that's probably the best term. I do agree with some of the people have changed that term. It's basically we're trying to precondition the muscle so that it is better able to express higher speeds or higher power outlets. So typically that's done with a high force activity followed by some sort of ballistic or plyometric activity. And the time frame between those two activities is really largely dictated by the individual's ability to recover or their strength seems to be a major indicator.

Stronger people tend to be able to dissipate fatigue faster it appears, so typically if you're stronger you can display your highest PAP enhancement or post-activation performance enhancement somewhere between 3 and 6 minutes. Now weaker people, it could be as high as 8 to 12 minutes between the two activities.

The problem that I have with that is that as a strength coach I don't want my athletes sitting around for 8 minutes. Because a) I've worked a little bit in professional rugby as a consultant and I've seen what they do when they have nothing to do in the weight room—it becomes quite chaotic. I've worked in American gridiron and it's not much different. When the inmates have a little bit too much time, chaos occurs.

The reality then becomes from a programming perspective: how do we optimize that? Well first, for me, one of the classic ways I always do PAPE is I do a heavy loaded quarter squat or some sort of squatting motion, couple it with some plyometric motion like box jumps, hurdle hops, things like that. That's an example.

There are benefits even if it's a shorter window between the two exercises, so your standard 2-minute rest interval should be fine. You'll still get benefit from it, it's just not the most benefit from it. And I think that's something you have to consider.

Now my wife used to do stuff with netball where she would do a heavy quarter squat and then a box jump, and that was one that she used a lot. Or she used a box drop jump into a vertical jump and a change of direction, so she could get that to match what they do in netball, so almost like a transition. And I think that works quite well as well.

And the idea then comes: could we optimize that? Now my former doctoral student, Dr. Laurel Sights, has a theory from rugby that I think is really quite interesting. He suggested that you would do a heavy quarter squat, go to a bench press, go to a jump squat, go to a bench throw, so almost like the French contrast. And to create your PAP that way. And I think there's some benefit to that.

And I think if you structured that correctly you could optimize the time between activities, so you would lengthen the duration using a non-fatiguing exercise in another part of the body. And I think that would work quite well. I mean, there's very limited research on KD's French contrast theory, which is basically sequencing exercises. There's a Russian method, there's a Bulgarian method if you read the various texts. But fundamentally it's high load to more explosive exercises and creating a sequence. And I think there's some really good benefits there.

I think you can overdo it. Like you don't just do everything as a circuit like that or as a complex. I think there's a time in place. I like to use PAP complexes or Professor Stone will call them strength-power potentiating complexes. I like to use them in-season because I can get a strength dose and a power dose and I can modulate: is it a strength-speed training work day or week, or a speed-strength depending on how I create those?

Now, Jimmy Radcliffe, the famous strength coach from University of Oregon, who in my opinion is probably the world's best strength coach—I find him to be one of the most inspiring strength coaches I've ever met. I met him in Houston at the GAIN conference, must be 5 years ago now.

James de Lacey53:55

Good. He's unbelievable.

Dr. Greg Haff53:55

He is one of the most knowledgeable but one of the most humble strength coaches I've ever met. When I was President of NSCA, I gave him an award—I owed him an award every year that he turned down. Every year he won't take a Lifetime Achievement. It was my dream to give him an award because I think he definitely deserves it more than anybody I've ever met.

But regardless, Jimmy did a wonderful study back in the day where he coupled power snatches with plyometrics and found great benefit there as well. So you could do Olympic lifts and couple them with high power activity, so high power, high power. And you could also look at it from an exercise ordering perspective. Loren Chiu actually talked about this as well. He did a study where he did pulling in vertical jump or pulling, snatching, pulling, jumping, things like that.

So again, post-activation potentiation enhancement is really about sequencing exercises to stimulate something or basically high load, high power, or some sequence. And so I think you can get quite creative with it. I think there's a lot of work that still needs to be done there. I mean, we just got a paper out—Dr. Nick Poulos, who's one of my doctoral students, just published a paper in JSCR. We did some of these sequencing ideas with AFL players, but the problem was Australian Rules Football players are just too weak to get the maximal benefit from it.

They're more on the endurance spectrum and we just didn't see what we had hoped we would see. I would think in rugby players or gridiron players we would probably see a greater benefit because they're just stronger individuals.

James de Lacey55:53

And then what about in the fighter population? I guess you would probably class more on the AFL side there regarding strength, depending on the sport. Like grapplers may be a little stronger. But what can weaker individuals do to take advantage of this? Can they take advantage of this? I know you mentioned longer rest periods, but that doesn't really help them take advantage of it. Is that something that they kind of have to work up to?

Dr. Greg Haff56:21

Yeah, I think—well, first off, I mean I am 100% in agreement with Professor Dip Marsham Baker on the fact that strength is the mother of all characteristics. Strength is the foundation for everything. And obviously we are on a sliding scale of strength depending on our athletic backgrounds, and what's strong for an MMA player is very different than what's strong for a gridiron player.

So I think you have to kind of think of it in that construct. I think there's place for—you have to go back to the base paradigm. When we talk about post-activation potentiation enhancement, there is time frames for optimal expression of power or velocity in the test exercise or the second exercise. So is there benefit? Yeah, there's still going to be benefit. It's just how much fatigue did they—are they under? So you may have to play with the loadings a little bit to make sure they can recover, and I think there's benefit.

I also think there's benefit for time efficiency. Now, on the flip side of that, if we create things like, for example, 5-4-3-2-1 cluster sets to try to get a PAP, it's got a double effect. Because what is a fighter going to have as part of their conundrum is they have to express forces under fatigue. So in that model you create—you're pre-fatiguing, and then you're trying to express high forces when you're a little bit tired. So that actually may have some really good transference.

I don't know of any studies that have looked at that directly, but from a conceptual perspective, I think that's where we could look at it as well—is creating cluster sets or these complexes where they have some fatigue to create force. So I think there's definitely benefit, but how you modulate it is depending on what your outcome measure is targeted.

James de Lacey58:20

Gotcha. Man, this has been an awesome chat. I mean, we could have gone another one, two hours just on these topics as well. But we're going to end here—I actually got to pick up our puppy from puppy school in a little bit as well. But if anyone wants to find you and follow you and see what you're up to, where can they do that?

Dr. Greg Haff58:20

Well, we're on Instagram at Doc Haff. We are on Twitter, not as much on Twitter anymore because it's a little bit odd now with all the weird advertisements that pop up. But definitely on Instagram—we put up stuff occasionally.

We're probably going to be starting a new thing in the next year, mTOR Strength Sciences, which is kind of a side project for us. We're starting to build out that. My wife and I, our weightlifting team is mTOR Barbell Club or Weightlifting Team, which is kind of a play on sports science, which is quite popular. So we're hoping to have that running in the next year, once I get a little bit of time away from ECU and do some of my own things. But definitely on Instagram.

James de Lacey59:13

Awesome. I'll link that up in the description anyway, but thanks for coming on, Greg. Really appreciate it.

Dr. Greg Haff59:13

No worries. Thanks, James.