Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

All training and programming related queries and banter here

Moderators: mgil, chromoly, Manveer

Post Reply
User avatar
GlasgowJock
Registered User
Posts: 1621
Joined: Tue Sep 19, 2017 9:15 am
Location: Glasgow, U.K.
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#761

Post by GlasgowJock » Mon Jun 29, 2020 11:44 am

This thread serves as a continual reminder of external bench wisdom, and internal bench frustration. Programming bench for progression is exasperating. I feel there's no 'middle ground' for it ie strength training due to injuries (typical 2-4 rep set work with some @7-9 singles like my squat and dead lift) though rather it's alternating practicing 'skill' singles at highish RPEs or accumulating a bunch of volume/ hypertrophy work in a session.

JeanLannes
Registered User
Posts: 67
Joined: Thu Apr 16, 2020 8:18 am

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#762

Post by JeanLannes » Mon Jun 29, 2020 6:59 pm

Hanley wrote: Sun Jun 28, 2020 3:05 am No, no.

Volume will vary tremendously by lifter, but roughly

30-50 reps per session with 70%/12RM
25-35 reps with 75%/10RM
15-25 reps with 80%
12-16 reps with 85%

Above 85%/5RM depends totally on lifter.
How different is this philosophy for lower body? Based on your own logs, there's less volume for lower body than upper body

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#763

Post by Hanley » Mon Jun 29, 2020 7:10 pm

JeanLannes wrote: Mon Jun 29, 2020 6:59 pm
Hanley wrote: Sun Jun 28, 2020 3:05 am No, no.

Volume will vary tremendously by lifter, but roughly

30-50 reps per session with 70%/12RM
25-35 reps with 75%/10RM
15-25 reps with 80%
12-16 reps with 85%

Above 85%/5RM depends totally on lifter.
How different is this philosophy for lower body? Based on your own logs, there's less volume for lower body than upper body
Not too different. Maybe less volume.

I'm not looking to hit lifetime 1rms on squats or deads...so I just sorta fuck around on those lifts. At some point I'll get serious and use "real" programming on deads.

dw
Registered User
Posts: 1502
Joined: Wed Jun 17, 2020 1:35 pm

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#764

Post by dw » Tue Jun 30, 2020 6:58 am

@Hanley have you looked at Average 2 Savage 2.0?

I'm not that well versed in programs but to me it looks remarkably Hanley-esque (e.g. sets across until you reach @7 RPE). All performed as explosively as possible.

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#765

Post by Hanley » Tue Jun 30, 2020 8:19 am

dw wrote: Tue Jun 30, 2020 6:58 am @Hanley have you looked at Average 2 Savage 2.0?

I'm not that well versed in programs but to me it looks remarkably Hanley-esque (e.g. sets across until you reach @7 RPE). All performed as explosively as possible.
I've only looked at it briefly.

A bunch of evidence has come out in the last ~2 years supporting the efficacy of low(ish) fatigue sets for strength (and power)...so I'm not surprised Greg is incorporating the approach.

I stumbled onto the approach as a way to avoid arthritis flare-ups. It just happened to work better than anything I'd tried previously.

AlexG332
Registered User
Posts: 40
Joined: Sat Apr 25, 2020 12:40 am

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#766

Post by AlexG332 » Tue Jun 30, 2020 2:01 pm

Hanley wrote: Mon Jun 29, 2020 11:07 am
KarlM wrote: Mon Jun 29, 2020 10:07 amare you finding that in general people respond better to this low rpe, high volume approach?
Generally, yes, I think so. I have some of "the data", but I don't have the null. So...eh, I'm pretty sure, yeah.

And lots of confounders. And so many people cutting (shakes fist at sky).

KarlM wrote: Mon Jun 29, 2020 10:07 amAre there outliers?
Totally. I have clients that could probably do 25 reps with 70% 1RM and others that could do maybe 7-8 reps. It's crazy.

I don't think high-volume low fatigue works all that well for either. With the 25 reps @ 70% guy...I can't fucking find the "stimulus zone" for a set that's also "low fatigue". Like RPE 4-6 zone is kinda meaningless with a 25RM. With this person, I think you need to push pretty close to failure on sets. It's just so easy to miss that stimulus zone and spend lots of time underworking.

The 8RM @ 70% guy is probably just better off doing low-volume double progressions or Hepburn type shit. There is no "high volume" for this guy.

Maybe there's a way to reliably adjust a HVLF format for these outliers...but I can't figure it out.

Then there are outlier "success" cases. Folks who seem to rocket up from years-long stalls to hit novel lifetime PRs (I'd put myself in the category) with a first run of HVLF. I don't know what's going on here. Maybe a high proportion of type II muscle fiber and this sort of format offers the perfect stimulus? IDK.
Is there a lower volume version? My work capacity seems pretty crappy. Getting through 10 sets of 3 or 2 is relatively fine, but trying to around 10 sets of 5+ is torture.

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#767

Post by Hanley » Tue Jun 30, 2020 2:19 pm

AlexG332 wrote: Tue Jun 30, 2020 2:01 pmIs there a lower volume version?
Just do less volume.

Or run a base-building/work-capacity focused block for 4-6 weeks (generally...3-4 sets per exercise in RPE 7-9 zone with loads anywhere fin 60-72ish %).

User avatar
tsor
Registered User
Posts: 177
Joined: Fri Mar 16, 2018 11:20 am
Location: NC
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#768

Post by tsor » Sat Jul 25, 2020 12:28 pm

Is there a point where a trainee is simply too weak for this kind of approach to be productive?

I took an extremely long layoff for several months due to illness and surgery (thanks, Crohn's disease) and lost around 80lbs. By the time I was able to start lifting again a couple of months ago, I was completely detrained and was essentially starting from scratch.

I've been doing an LP of sorts (something like GZCLP https://old.reddit.com/r/Fitness/commen ... thod_from/) for the last month or so and that's been going well enough, I suppose. My bench has always been abysmal, though, so I'm wondering if something like HVLF would set me up better for eventually having a respectable bench.

My E1RM is only around 185lbs right now at a similar body weight, though, so I'm not sure if doing sets of 120 would be a good idea. I was thinking of just using the base program, but with a more aggressive loading approach, like adding 5lbs to my E1RM every workout. Feel free to let me know if this is a dumb idea and I should just LP until my numbers are better, though.

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#769

Post by Hanley » Sat Jul 25, 2020 12:38 pm

tsor wrote: Sat Jul 25, 2020 12:28 pm I should just LP until my numbers are better, though.
I'd honestly just do this. If you can LP without crushing yourself, I think you should. Then move onto the more complicated stuff when you have to. For your "abysmal" bench, I suppose you could just "LP" these 3 sessions to try-out the HVLF approach:

1: Sets of 5 with 12RM until RPE 7-8ish
2: Sets of 2-3 with 8RM until RPE 7-8ish
3: 10-15 reps with 5RM in singles and doubles

^ just run those (in whatever order you like best) with 48-72 hours between sessions, and add about 2% load every time you repeat the sequence.

User avatar
tsor
Registered User
Posts: 177
Joined: Fri Mar 16, 2018 11:20 am
Location: NC
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#770

Post by tsor » Sat Jul 25, 2020 12:48 pm

Hanley wrote: Sat Jul 25, 2020 12:38 pm
tsor wrote: Sat Jul 25, 2020 12:28 pm I should just LP until my numbers are better, though.
I'd honestly just do this. If you can LP without crushing yourself, I think you should. Then move onto the more complicated stuff when you have to. For your "abysmal" bench, I suppose you could just "LP" these 3 sessions to try-out the HVLF approach:

1: Sets of 5 with 12RM until RPE 7-8ish
2: Sets of 2-3 with 8RM until RPE 7-8ish
3: 10-15 reps with 5RM in singles and doubles

^ just run those (in whatever order you like best) with 48-72 hours between sessions, and add about 2% load every time you repeat the sequence.
Gotcha. Thanks, Hanley.

dw
Registered User
Posts: 1502
Joined: Wed Jun 17, 2020 1:35 pm

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#771

Post by dw » Sat Jul 25, 2020 12:57 pm

@Hanley

Would you say in general doing a high number of sets, say 10 per movement, as explosively as possible, and taking each set to RPE 7 is a reasonable distillation of HVLF? Or would RPE 6 say be better?

The reason I keep asking all these questions is I believe this approach works but I don't want to just abandon the program I'm on because I'm still linearly progressing on it. I'm hoping to sub in some HVLF replacements as needed.

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#772

Post by Hanley » Sat Jul 25, 2020 1:20 pm

dw wrote: Sat Jul 25, 2020 12:57 pm Would you say in general doing a high number of sets, say 10 per movement, as explosively as possible, and taking each set to RPE 7 is a reasonable distillation of HVLF? Or would RPE 6 say be better?
Eh, not really. I'd qualify "as explosively as possible" and I don't like using RPEs here.

On paper, you want to use "maximal volitional intent". But in reality, I think about 80% "max volitional intent" is better. Otherwise you'll start bouncing like crazy off ligaments at the transition from eccentric to concentric...and, generally, shit can go to hell pretty quickly. So, yeah "80%" is fine.

I've always used about 20% velocity loss from the fastest rep of the set (not always the first rep) as the standard for a "set". That usually is way less than RPE 7. A crude heurisitic might be: a set = "reps until appreciable bar speed slowdown". Then I'd do "sets" until RPE 7-8ish.

dw
Registered User
Posts: 1502
Joined: Wed Jun 17, 2020 1:35 pm

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#773

Post by dw » Sat Jul 25, 2020 1:39 pm

Hanley wrote: Sat Jul 25, 2020 1:20 pm
dw wrote: Sat Jul 25, 2020 12:57 pm Would you say in general doing a high number of sets, say 10 per movement, as explosively as possible, and taking each set to RPE 7 is a reasonable distillation of HVLF? Or would RPE 6 say be better?
Eh, not really. I'd qualify "as explosively as possible" and I don't like using RPEs here.

On paper, you want to use "maximal volitional intent". But in reality, I think about 80% "max volitional intent" is better. Otherwise you'll start bouncing like crazy off ligaments at the transition from eccentric to concentric...and, generally, shit can go to hell pretty quickly. So, yeah "80%" is fine.

I've always used about 20% velocity loss from the fastest rep of the set (not always the first rep) as the standard for a "set". That usually is way less than RPE 7. A crude heurisitic might be: a set = "reps until appreciable bar speed slowdown". Then I'd do "sets" until RPE 7-8ish.

Ok thanks, I think I understand.

413x
Registered User
Posts: 125
Joined: Wed Mar 25, 2020 4:50 pm

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#774

Post by 413x » Sat Jul 25, 2020 3:16 pm

Hanley wrote: Sat Jul 25, 2020 1:20 pm
On paper, you want to use "maximal volitional intent". But in reality, I think about 80% "max volitional intent" is better. Otherwise you'll start bouncing like crazy off ligaments at the transition from eccentric to concentric...and, generally, shit can go to hell pretty quickly. So, yeah "80%" is fine.
Do you give 80% only during the transition, or the whole time?

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#775

Post by Hanley » Sat Jul 25, 2020 7:27 pm

413x wrote: Sat Jul 25, 2020 3:16 pm
Hanley wrote: Sat Jul 25, 2020 1:20 pm
On paper, you want to use "maximal volitional intent". But in reality, I think about 80% "max volitional intent" is better. Otherwise you'll start bouncing like crazy off ligaments at the transition from eccentric to concentric...and, generally, shit can go to hell pretty quickly. So, yeah "80%" is fine.
Do you give 80% only during the transition, or the whole time?
Huh. I honestly don't know. I feel like I sorta set the "effort parameters" before I start the rep and check-out while actually moving the bar. I'll try to see what that actually means when I bench tomorrow.

gtl
Registered User
Posts: 1699
Joined: Fri Sep 15, 2017 10:04 am
Location: Naptown
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#776

Post by gtl » Mon Jul 27, 2020 3:23 am

Hanley wrote: Sat Jul 25, 2020 1:20 pm I've always used about 20% velocity loss from the fastest rep of the set (not always the first rep) as the standard for a "set". That usually is way less than RPE 7. A crude heurisitic might be: a set = "reps until appreciable bar speed slowdown". Then I'd do "sets" until RPE 7-8ish.
When HVLF calls for 7-11 sets, is there a target for # of sets with 20% velocity loss? Like 3-5 sets, but it might take 4-6 sets to get there? Or do you see the 20% velocity loss more as a result of load selection?

IE:
225x7 13%
225x7 15%
225x7 14%
225x7 17%
225x7 20%
225x7 21%
225x7 24%

vs.

235x7 20%
235x7 23%
235x7 22%
235x7 25%
235x7 24%
235x7 25%
235x7 30%

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#777

Post by Hanley » Mon Jul 27, 2020 5:46 am

gtl wrote: Mon Jul 27, 2020 3:23 am
Hanley wrote: Sat Jul 25, 2020 1:20 pm I've always used about 20% velocity loss from the fastest rep of the set (not always the first rep) as the standard for a "set". That usually is way less than RPE 7. A crude heurisitic might be: a set = "reps until appreciable bar speed slowdown". Then I'd do "sets" until RPE 7-8ish.
When HVLF calls for 7-11 sets, is there a target for # of sets with 20% velocity loss? Like 3-5 sets, but it might take 4-6 sets to get there? Or do you see the 20% velocity loss more as a result of load selection?

IE:
225x7 13%
225x7 15%
225x7 14%
225x7 17%
225x7 20%
225x7 21%
225x7 24%

vs.

235x7 20%
235x7 23%
235x7 22%
235x7 25%
235x7 24%
235x7 25%
235x7 30%
There's a lot of arbitrary is my parameters. Even the 20% is arbitrary.

What I want is:

1) high peak force reps in a "set" until a marked fall-off. I use 20% because 20% velocity loss tends to be beyond the threshold of rep-to-rep random fluctuation in force/velocity. I've had plenty of sets of - say 8 @70% - where the 4th or 5th rep was the fastest. I don't want to prematurely stop a set because of "noise". 20% is just a practical, utility-value threshold.

2) maximize session sets with the condition that I want near-complete intersession recovery.

^ that's it. That's my big-picture goal with session design. All the other stuff is fiddling with measuring tools/conventions

gtl
Registered User
Posts: 1699
Joined: Fri Sep 15, 2017 10:04 am
Location: Naptown
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#778

Post by gtl » Tue Jul 28, 2020 6:47 am

Hanley wrote: Mon Jul 27, 2020 5:46 am
gtl wrote: Mon Jul 27, 2020 3:23 am Specific Question
Response
Sorry, to clarify maybe in a more general sense, would you say that sets below an arbitrary barspeed loss should not count as work, or are they valuable in itself due to max force and accumulative intraworkout fatigue. IE, a range of sets is suggested (7-11), but is that actual # more dependent on how many sets are around that targeted barspeed loss?

User avatar
Hanley
Strength Nerd
Posts: 8752
Joined: Fri Sep 15, 2017 6:35 pm
Age: 46

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#779

Post by Hanley » Tue Jul 28, 2020 11:52 am

gtl wrote: Tue Jul 28, 2020 6:47 am or are they valuable in itself due to max force and accumulative intraworkout fatigue. IE, a range of sets is suggested (7-11), but is that actual # more dependent on how many sets are around that targeted barspeed loss?
Yeah, they're valuable. Do you have bar speed data? It'd be interesting if you weren't hitting ballpark velocity loss. Some people are super fatigue resistant and those peeps need some modifications.

Here's some of my old data:


gtl
Registered User
Posts: 1699
Joined: Fri Sep 15, 2017 10:04 am
Location: Naptown
Age: 38

Re: Experiments in High-Volume, Low-Fatigue Bench Programming (more "Montana Method" nonsense)

#780

Post by gtl » Tue Jul 28, 2020 12:32 pm

Hanley wrote: Tue Jul 28, 2020 11:52 am
gtl wrote: Tue Jul 28, 2020 6:47 am or are they valuable in itself due to max force and accumulative intraworkout fatigue. IE, a range of sets is suggested (7-11), but is that actual # more dependent on how many sets are around that targeted barspeed loss?
Yeah, they're valuable. Do you have bar speed data? It'd be interesting if you weren't hitting ballpark velocity loss. Some people are super fatigue resistant and those peeps need some modifications.

Here's some of my old data:

Yeah, I got my OBB. The raw data is not always clean (I don't use the inputs or delete reps) and I'm getting an error exporting.

Idk how this will turn out copy/pasted from my last 2 squat/bench sessions:

TNG Bench
Reps maxV minV Vel Loss % Loss
5 0.56 0.43 0.13 23%
5 0.53 0.4 0.13 25%
5 0.53 0.39 0.14 26%
5 0.5 0.41 0.09 18%
5 0.5 0.41 0.09 18%
5 0.54 0.4 0.14 26%
5 0.51 0.38 0.13 25%

Reps maxV minV Vel Loss % Loss
7 0.57 0.45 0.12 21%
7 0.56 0.4 0.16 29%
7 0.54 0.4 0.14 26%
7 0.54 0.39 0.15 28%
7 0.54 0.38 0.16 30%
7 0.53 0.39 0.14 26%
7 0.54 0.3 0.24 44%

SSB Squat
Reps maxV minV Vel Loss % Loss
5 0.61 0.54 0.07 11%
5 0.58 0.5 0.08 14%
5 0.57 0.49 0.08 14%
5 0.54 0.47 0.07 13%
5 0.58 0.49 0.09 16%
5 0.61 0.52 0.09 15%
5 0.57 0.48 0.09 16%

Reps maxV minV Vel Loss % Loss
7 0.7 0.62 0.08 11%
7 0.68 0.6 0.08 12%
7 0.69 0.6 0.09 13%
7 0.69 0.59 0.1 14%
7 0.65 0.56 0.09 14%
7 0.65 0.51 0.14 22%
7 0.63 0.52 0.11 17%

Post Reply