Random Practice vs Metronome - when to do what?

[BlackInMind]

So a combination of Troy’s latest youtube video, plus a recent video Martin Miller posted on developing speed got me thinking about the merits of random practicing vs traditional metronome use. For most of my playing career I’ve employed the tried and true “John Petrucci Rock Discipline” start slow and build up gradually approach, but much more recently I’ve adapted this less structured, random, bursting kind of practice Troy and Martin talk about, and I have to say its something I wish I had been doing since my teenage years.

I’m not knocking using a metronome… I believe it really helped my timing and ability to play to a click track in a studio setting, but I’m wondering if there are times to apply one method and times to apply another? Or if they both more or less accomplish the same thing, its just a personal learning preference. I do find it takes me way less time to “warm up” nowadays since applying a more randomized structure to my practice.

I know we have forum members who may have very strong opinions on this, such as John from Mile High Shred, so I’m really curious as to what everyone seems to use. I will say, I do notice a pretty common thread with a lot of guitarists I view as at the very least “looking” like they have very natural technique, such as Yngwie, Gilbert, Michael Romeo, Emil Werstler, etc and those guys tend to describe their practice as more random and exploratory in nature. Actually, Emil was the one who got me hip to this approach through taking some skype lessons with him (He would say “hit it and quit it” in terms of how to cycle licks, as in, don’t spend too long on one thing).

If anyone isn’t familiar with the youtube videos in question, here they are:

> Blockquote

[Gtrjunior]

Good topic.
I’ve mostly done my practicing over the years opposite what you have.
I just recently began consistently using a metronome and I wish I had done it from the start.
But, I also feel that the burst method works well too. Sometimes with a metronome I feel like I just can’t overcome a bpm barrier. So usually I shut it off and just “go for it”. More often than not I’m able to pass my barrier, even if only for a short time. But it’s usually enough to give the ole confidence a boost, if nothing else!!

[Troy]

but much more recently I’ve adapted this less structured, random, bursting kind of practice Troy and Martin talk about

I wouldn’t say it’s less structured. It’s structured, it’s just a different structure.

Were you a gamer ever? What console? The first time you play a new game you have to learn the controls right? You probably mash a bunch of buttons haphazardly and see what happens, because you don’t even know what the controls do yet. Once you figure that out, you see, ok, that button is jump, the other one is punch, the two in a sequence means jump-punch attack, and so on. And then you learn what scenarios the different techniques work in. And to do that you probably have to die a bunch of times before you realize, ok the best way to do that is to let the other guy punch first and then you time it and do the other sequence in response. Over time, your approach becomes more consistent with less trial and error because you no longer need trial and error to figure out what to do in each scenario. You get better at spotting and timing the scanarios, and the hand movements become memorized.

This isn’t just video games, and it isn’t just guitar. It’s everything you ever learn that’s physical. Even if you have a really good teacher, they can cut down on the amount of this you will need, because they can show you what works. But you can never eliminate the TE part because you still need to do it all yourself, under your own power, and go “aha that’s what it feels like when I get it”.

When should you do this? Whenever you are doing physical learning. If the movement is not fully learned yet, or if it is learned incorrectly, then you need to learn it correctly. This involves capturing the click, i.e. “feeling of doing it correctly”, and then trying to replicate it again, after a period of time has passed, to test that you can reproduce it from memory by feel alone.

My exact structure for this involves capturing that click, doing it correctly for a few reps, and then putting the guitar down and stepping away. Then I come back a few minutes later and try to replicate. If I can’t, it’s back to trial and error. Play a different phrase. Play a different speed. Change the hand position. Change the grip. Change the pick. Keep searching for the click. Once you get the click, do it a few times and put it down. Yes, stop playing. This gives your brain a chance to “learn” what you have shown it. And what you have shown it was the very last thing you did before putting the guitar down. That’s the clearest signal you can send.

What I don’t do is capture and then keep playing until I lose it. That would be like pedaling the bike until you fall over. Instead, I like to pedal, get that feeling of balance, and then come to a stop and step off - without crashing. In guitar terms, I try to capture the movement feel, play a few successful reps, then step away. I want to send my brain the signal of what “correct” feels like, and give it a few minutes to incorporate that feeling. Then I come back, pick it up, and try to replicate.

You’ll flip flop back and forth around getting it, not getting it - sometimes for a long time. Especially if the movements are unclear to you and you have no or vague teaching. This is why we want to provide total clarity on the hand positions and movements. We want you starting halfway down the road to correct, not at the beginning of the road in total darkness.

But in all cases, there is this iterative process of capturing and replicating that I go through.

[Frylock]

This involves capturing the click, i.e. “feeling of doing it correctly”

This is where I think the dogmatic “play slow perfectly, and gradually speed up” goes astray. The “feeling of doing it correctly” of fast picking is something that you’ll never experience below a certain speed. Avoiding “fast” attempts in the name of perfecting “slower” speeds first is systematically depriving oneself of opportunities to capture the “click” of high speed picking. Once you’ve experienced that “click” you can try to hone in on the “click-friendly” movement even during slower practice, and you can systematically refine it to higher and higher speeds. But as you reiterated yet again in the latest crosspicking video, there’s no point trying to “perfect” a “fast phrase” at low speed if you don’t yet have any solid basis for self-correcting re: what the movement will need to look and feel like at high speed.

I think this is where things like the “bursts” Petrucci advocated in Rock Discipline really make sense. A wide range of trial and error that includes “fast” attempts in order to discover movements that have a chance of working at high speed, and methodical metronome work to refine a technique once you’ve begun to experience “the feeling of doing it correctly”.

I actually witnessed a sudden epiphany of spontaneous technique in a high-school garage band once. Our drummer (not a guitar guy), picked up my guitar on a lark during a break, and doing his naive air-guitar movements on my guitar, ripped out a very convincing sounding fast ascending run across a couple of strings. He was as amazed as the rest of us, and had no idea why it had come out sounding so cool. We were all laughing about it, and he said “I could never do that again in a million years.” He couldn’t articulate what he had done, apart from “moving my hands really fast”, and when he tried again a few times, it came out sounding like ass in the way one would expect. But I wonder now, if he had a little more prior guitar experience to provide self-awareness at that moment, whether he might have been able to pull some sort of useful insight from that “lightning strike” moment. As they say, even a blind squirrel sometimes finds a nut once in a while. To me, that moment really spoke to the lie of the idea that all change has to be gradual. Developing consistency with that kind of ability, maybe that’s more of a gradual thing.

I think guys like Michael Angelo Batio underestimate the capacity of inexperienced players to “bark up the wrong tree” with their slow metronome practice. I’m pretty sure one of Batio’s videos does mention at one point the idea that your slow practice technique needs to be the “same” as your fast technique for the metronome approach to give good results, but I don’t think it was emphasized enough for people to recognize its importance. And as we’ve been discussing, until you’ve experienced the “click” it’s virtually impossible to self-correct the slow practice, because until you’ve felt the click, you don’t really have an informed ideal of what movement/feeling to strive toward.

And regardless of how long the “click” thing has been established in academic literature, it’s been part of anecdotal sports coaching wisdom at least since I was doing recreational sports as a kid in the 80s: the emphasis on “remembering what that felt like” when you execute an action in a way that gives a good result and/or matches conventional wisdom (as interpreted by the coach) about “correct” technique. In complex movements, sometimes it’s still necessary to “slow things down” in order to instruct the principles of how to perform the movement correctly, but once you’re trying to bring things together at speed, getting the parts to fit together properly to execute successfully, and diagnosing your own mistakes is largely about having that “feeling of doing it correctly” as a reference, and you can’t get that until you’ve done some trial and error at speed (though coaching can help reduce the guesswork involved in that trial and error).

[BlackInMind]

Thanks for the response, Troy! So do you think the players that tend to recommend that kind of practice achieved the same thing that the dedicated metronome players did just through a different process? Where do you see the application of the standard metronome approach fall into this? Is it worth doing a little of both simultaneously or do you think it’s more important to replicate the “click” 100% every time before doing any kind of metronome work?

And I’m slowly building my retro NES collection these days!

[Troy]

So do you think the players that tend to recommend that kind of practice achieved the same thing that the dedicated metronome players did just through a different process?

I don’t think there is one group or another. Based on our interviews with researchers, I think this is how all mechanical learning works. Players either recognize that they went through this process or they don’t. Players who are really good at figuring things out by feel probably just didn’t perceive that’s what they were doing. Either because it was quick for them, or again because they just didn’t perceive that their various wrong attempts in the beginning were actually an important part of the learning process.

To be fair, researchers weren’t aware of this either until a couple years ago after one study in particular, done at Harvard, revealed it. They used to just think mistakes were mistakes and went away over time as ability improved. Nobody really considered that the mistakes were actually necessary to cause the ability to improve.

Where do you see the application of the standard metronome approach fall into this?

It’s a spectrum. If you are learning new physical movements, this is our best understanding of how that process works. If you already have those movements, and are simply trying to memorize new musical pieces, you could argue that’s not even the same kind of ‘practice’ at all. What techniques would you use to memorize the rules of a card playing strategy? Would it be the same or different from the way you would learn to perform fancy card shuffling tricks? Pretty different I would think.

The reason the ‘metronome method’ fails is because it assumes you already have most of the skill you are trying to learn. It is more applicable when you get to the end of the road and you are trying to test out your new movements at various speeds, or at the very very end when your movements are essentially done and you are tying to memorize new sequences of them.

Even then, you would need to show me hard evidence that using incremental computer-like increases in tempo actually produces results at any stage of learning. Because honestly no other aspect of human performance that I can think of really works in strict linear fashion like that.

[Troy]

What techniques would you use to memorize the rules of a card playing strategy? Would it be the same or different from the way you would learn to perform fancy card shuffling tricks? Pretty different I would think.

Not the greatest analogy! A better one, to keep things in the realm of physical performance, would be what it feels like to me at this point to learn new piano pieces. I don’t really need basic movements, per se. But I’m definitely learning to string together different sequences of movements I already have. And that can still feel foreign in a physical sense. However when your motor movements are learned like this, you can use slower tempos, and that is probably an aid to memorization and note-level correctness.

In other words, slower tempos and note-level correctness happen toward the end of the process. Fast tempos do too. But slow tempos don’t happen at the beginning because the movements aren’t really in place yet. How can they? So again, I think the reason “traditional” practice advice fails is because it more closely resembles what you’d do at the end of the road, rather than the beginning.

If you’re trying to learn a new motor skill, you need to practice like someone who doesn’t have it yet!

[BlackInMind]

Troy, I think I’m understanding much better now. The two biggest impacts on how I practice were the Rock discipline VHS and my lessons with Stump during my Berklee days and summer sessions. Stump’s metronome use was a little less regimented than Petrucci’s though, he seemed to make larger jumps in tempo and didn’t start at tempos that “put you to sleep” (in his words). So I’m trying to wrap my head around how to integrate this new information into not just my own practice, but also how I relate practice strategies to my own students.

[Troy]

Very simply, I would just say, what is it someone is trying to learn, and where along the process are they? If it’s motor movements, and they’re at the very beginning, then get them doing those things as smoothly and naturally as possible right away. You’re the teacher, you know what good form looks like. The coach knows what a perfect handspring looks like, the gymnast doesn’t. And more importantly, the gymnast doesn’t know what it feels like to nail one either. Get them out there on the mats, making sloppy but natural/realistic handsprings as soon as possible. Get them to recognize by feel when they’re doing it correctly. Use whatever random means at your disposal to achieve this. Don’t worry about getting every note correct at every speed. Get them moving naturally and make those movements realistic. Once they can recognize by feel what “correct” is like, they can start to slow down and clean it up.

That’s pretty much it.

[Frylock]

The coach knows what a perfect handspring looks like, the gymnast doesn’t. And more importantly, the gymnast doesn’t know what it feels like to nail one either.

This, and your previous mention of Teemu using foam wedges to help students feel out certain picking movements, reminded me of something I saw years ago about the availability of “pac-mac” shaped training mats for backhandsprings. Not sure whether their use is widespread or not, but the short video below illustrates how they can be used, along with some coaching of the “early” stage of working out a movement.

[Troy]

Not sure whether their use is widespread or not, but the short video below illustrates how they can be used

That looks fucking terrifying. Let’s try a dive - with no water!

[Frylock]

@Troy

That looks fucking terrifying. Let’s try a dive - with no water!

For the record, I did recreational gymnastics for a couple of years as a kid, and when they tried to teach us back handsprings, we didn’t use anything like this, it was just with an adult supporting our back with their hands. One of the hardest parts was being willing to “sit” down into that unbalanced “sitting in a chair that isn’t there” position (where if you tried to rest in that position, you’d fall down backwards), rather than doing a “balanced” squat. And no, I never actually learned to do one unassisted.