View Full Version : "Intervals within Intervals"

Steve Shafley
10-20-2006, 11:50 AM
Could you elaborate on this a bit, Robb? Maybe a bit of the reasoning behind it and then an example or two?

I'll re-read "Power Bias 2" while I'm at it to see if I missed it the first time.

Robb Wolf
10-22-2006, 01:34 PM
I guess the way I am looking at this is...take Helen: 400m run, 21KB Swings, 12 Pull ups-3 rounds. Now the fact one is breaking up the runs, swings and PU is a form of mixed-mode interval training, however this is performed as quickly as possible, which necessitates a SUSTAINABLE pace.

By contrast this "intervals within intervals" idea is to break the sub-sets of something like Helen into discrete pieces:
400m run-2min rest (max speed)
21 KB swings-2min rest (max speed)
12 pull-ups-2min rest (max speed).

So we have the interval aspect of the mixed-mode approach AND the emphasis of max power production during those sub-elements because we have some built in rest periods, which are necessary to maintain that PEAK power output.

Like I mentioned in the second PB piece the original Helen is a serious fanny kicker (high AKP) but the emphasis on max power output in this alternative format is even nastier...and it offers some adaptations that are simply impossible to achieve when the objective is a "good" average time.

Let me know if the helped or muddied things!

Steve Shafley
10-22-2006, 09:40 PM
It's interesting to play around with that stuff:

If you assign numbers to the movements in the Helen workout, where:

1 = 400m run
2 = 21KB Swings
3 = 12 Pull ups

Regular Helen is (1+2+3)*3, where success is measured by total time for completion

Power Bias Helen is (1+1+1+2+2+2+3+3+3) where success is measured by minimizing the time for each individual exercise's set.

Or is the power bias Helen (1+2+3) with full rest in between, and then with a repeat? [(1+2+3)+2 min rest]*3

Two very different things, for sure. Or even three.

Yael Grauer
10-22-2006, 09:55 PM
I thought it was (1+rest+2+rest+3+rest)*3, but math has never been my strong suit.

Eva Twardokens
10-22-2006, 09:55 PM
It was great to see you this weekend and I will give this a whirl...it seems right down my alley for training right now. Will report back at the end of the week.

Steve Shafley
10-23-2006, 05:05 AM
Actually, Yaul, that's what I meant in the first string of +s and #s. I didn't clarify it. Sorry.

Yael Grauer
10-23-2006, 10:17 AM
I tried a variation of this out on some kids I was volunteering with, and the way I did it without having to give everyone a stopwatch was by putting them in small groups. In this case, it was two teams of two and they were racing each other, but I could see it working with four-person teams (still in groups of two.) The way I worked in intervals without having to use a stopwatch is by having them take turns. They did two rounds of 25 situps (they held each other's feet), a shuttle run and 15 rolls (their choice) but had to wait for their partner to finish before they could go. I think it was ~ 2 minutes for the situps and rolls, the run was a bit quicker. If I was really good, I could have divided them by ability level, had everyone go at the same time and described the second exercise to each group in exactly two minutes after they finished the first one.

Anyway, my goal was to come up with a cooperative group learning activity that was appropriate for the age group, was okay for different skill levels and incorporated intervals within intervals as well as all seven aspects of cooperative learning, according to this really awful educational psychology textbook I'm reading. (Group heterogeneity, group goals/positive interdependence, promotive interaction, individual accountability, interpersonal skills, equal opportunities for success, team competition.) It worked pretty well and got them working together instead of hitting each other, but I think they like it when we kick the soccer ball around better. Also it was too short and I should've added rounds.

Just thought I'd share this because it seemed like a good way around the "reasons this isn't used more" part of the article...for kids, anyway.

Ryan Atkins
10-23-2006, 12:42 PM
Another option for getting this to work for a group of kids may be to assign a number of reps for the group as a whole to accomplish. For example, you could use 200 squats for 4 kids (or 400 for 8 kids). Once the total of the group reaches that level they all get to rest for 2 minutes before doing the next round or exercise. Hopefully this would promote some in group competition (somebody gets to brag about contributing the most to their team), some competition between groups, promote some team cohesion and might even refine adding skills among the participants. Assuming they're motivated, all the participants would be working at or close to peak power capacity during the work timeframe, regardless of their varying athletic capability. It also has the potential for cutting down on the number of stopwatches required.

Just a thought,


Robb Wolf
10-23-2006, 05:54 PM
Hey Eva!
We were stoked you could stop by. Keep us posted on the OL's and minimalist met-con!
Good to see you amigo!

Joe Hart
07-13-2007, 12:59 PM
So I gave Robb's interval with in an interval a try. Let me just say it was interesting. I used Helen, because she is my fav, with 2:00 rest between exercises. I liked it quite a bit. I certainly looked forward to each event. 2:00 seemed lilke a long time. I had to make myself wait. Strangely my 400m was faster in each round than the first 400m of the previous Helen bout. It was like blasted off and crashed on the way back. I would get about 50m from the finish line and crash. The KB swings were all unbroken, where in the last bouts it was 21, 14/7, 7/7/7. Pullups are not my strong suit, they went quick for a non-kipper (more like a carp on a stringer).

All in all I liked it. Now I have to figure out how to do the other girls that way.

Russell Greene
07-13-2007, 04:32 PM
I like this idea. Something I've been thinking about and am about to try is

1 minute 135 lb. thrusters
1 minute pullups
1 minute rest
for rounds

Or ripping off Diane:

1minute 245 lb. dl's
1 minute HSPU's
1 minute rest

or Elizabeth

1 minute 155-175 lb. cleans
1 minute ring dips
1 minute rest

I leave the upper body movement as prescribed since I am more in need of practice on the rhythm and endurance of them than strength.

I just broke 4 minutes on Fran and am going to use this to help me get to sub 3, in addition to a bunch of other stuff of course.

Matt Cricchio
07-13-2007, 05:38 PM
I was very interested in this kind of training when I read the Power Bias article so I decided to incorporate it into my new cycle. I went through the majority of the "girls" and some other CF classics and attempted to make them fit the power mode. I used Robb's examples with Cindy and Helen and for the rest I cut volume as well as adding intensity and plyometrics where I thought it might work. I'm not sure if I was on the right track but I will leave that for you guys to decide-

Power Girls:

Angie: 5 rounds of 10 pull-ups, 10 push ups, 10 sit-ups, 10 squats in 20lb weight vest. 2 minute rest between rounds.

Barbara: 5 pull-ups with 20lb vest, 10 clapping push-ups, 20 GHD med ball sit-ups, 30 squat jumps. 3 rounds, rest 3 minutes between each round.

Chelsea: On the minute every minute for 10 minutes perform 5 clapping pull-ups, 10 clapping push-ups, 15 jump squats.

Cindy: 5 rounds of 5 clapping pull-ups, 10 clapping push-ups, 15 jump squats.

Jackie: 5 rounds of 200m row (damper at 7), 10 95# thruster, 6 pull ups with 20# vest. Rest 2 minutes between rounds.

Nancy: 3 rounds. Run 400m. Rest 2 minutes. 15 overhead squat 95#’s. Rest 2

Helen: 3 rounds. Run 400m. Rest 2 minutes. 21 Kettlebell Swings. Rest 2 minutes. 12 pull ups. Rest 2 minutes.

Kelly: 5 rounds. Run 400m, rest 2 minutes. 15 box jumps to 32’ platform, rest 2 minutes. 15 wall ball shots to 15 foot target, rest 2 minutes.

Filthy 20: 20 Box jumps, 32’ platform. Rest 2 minutes. 20 weighted pull-ups 20# vest. Rest 2 minutes. 20 kettlebell swings. Rest 2 minutes. 20 step Walking Lunge, 45# bar. Rest 2 minutes. 20 GHD med ball sit-ups. Rest 2 minutes. 20 push press, 75#. Rest two minutes. 20 back extensions, 45# bar. Rest 2 minutes. 20 wall ball shots to 15 foot target. Rest 2 minutes. 20 burpee box jumps. Rest 2 minutes. 20 tuck jumps with 20# vest.

Micheal: 3 rounds. 400m run in 20# vest. Rest two minutes. 25 GHD med ball sit-ups. Rest two minutes. 25 back extensions, 45# bar. Rest two minutes

Tabata This: 4 tabata rounds of squat jumps. 2 minute rest. 4 tabata rounds of rowing at #7 damper. 2 minute rest. 4 tabata rounds of clapping pull-ups. 2 minute rest. 4 tabata rounds of GHD med ball sit-ups. 2 minute rest. 4 tabata rounds of clapping push-ups. 2 minute rest.

Karen: 3 rounds. 30 wall ball shots to 15’ target. Rest 2 minutes between rounds.

Robb Wolf
07-13-2007, 06:58 PM
I will definitely give some of those a shot! Good stuff.

Marc Moffett
07-14-2007, 11:33 AM
Curious what you think of this one based on the filthy fifty:

30 box jumps
2 min. sprints (tabata intervals)

rest 2 min.

30 jumping pull-ups
2 min. sprints

rest 2 min.

30 wall ball shots
2 min. sprints

2 min. rest

30 burpees
2 min. sprints

2 min. rest

30 double unders
2 min. sprints


I was shooting for something that was all about running and jumping as conditioning for basketball. It was pretty tough to complete the sprints with adequate speed; might be better to increase the reps on excercises and decrease the number of spring intervals per time (maybe, two to three). Thoughts?

Chris Forbis
07-14-2007, 03:52 PM
I've done tabata sprint intervals before. I don't like them, as by the 3rd or 4th round they no longer really resemble sprinting.

Maybe replace the 2 minute tabata intervals with:
Half-court back, full-court back (about 15 seconds)
30 seconds rest
Half-court back, full-court back (another 15 seconds)

Then have your 2 minutes of rest.

You could do a 100m dash for each one instead... but I really like the need to decelerate, plant, and accelerate for the half-court back, full-court back.

That will give you 10 high quality bouts of sprinting. May need to tinker with the 30 second rest in between... make sure it is long enough so you can really hammer the sprints.

Dave Van Skike
07-14-2007, 04:46 PM
was thinking the same thing...sprints are speed work..tabata is speed endurance work....

Marc Moffett
07-15-2007, 09:32 AM
Thanks Chris, That seems right. Not enough time to adequately recharge on the tabata intervals, which undermines the attempt to make the whole thing a power bias workout.

Suppose we change it along the lines you suggest. The result is a triple interval structure: (1) a basic interval structure, (2) a multi-modal interval structure, (3) a basic interval structure within one or both modes.

If you keep the couplets (jump+sprint) short enough that, when you include the sub-intervals, you can really go with max effort, then the whole workout has a power bias structure: couplet-rest-couplet-rest-couplet...

David Gutierrez
07-15-2007, 11:16 AM
Long time lurker, big fan of the PM ...

Background ... I try to keep with a 3 on 1 off cycle as much as possible, but it doesn’t always work. It usually jumps between 2-4 days on, 2-4 days off, then 2-4 days back on. The first day back is usually a crapshoot and the workout is scaled to compensate for a shitty sleep cycle, fatigue and poor nutrition. I'll usually cut rounds, reps or weight accordingly to allow for a bit of “reintegration”.

Because of this, I have started using the “intervals within intervals” as a way to either regulate fatigue, or maximize power output depending on the day and how I feel. I've kept it pretty simple, generally applying it to WODs where there is a run followed by a weighted exercise/gymnastic movement. Example:

5 rounds:
400m run
95 lb thrusters x 15
pull-ups x 15

If I'm feeling good, I'll usually do a 1:1, maybe 1:2 work/rest ratio (or I'll just pick something like 1 or 2 min as the standard). What I’ve noticed here is that the times will remain fairly consistent across the board, with the exception of the last 1 or 2 rounds—there I’ve noticed a significant burn out in the runs the last 100 meters. The times in the other exercises have remained steady.

If I'm not feeling as sharp, I’ll increase to a 1:2, 1:3 or full recovery between intervals. I'll usually drop the weight or cut a round off the workout as well.

So far, this has been helpful in regulating fatigue. I could use a bit more strength-specific work ... perhaps ME/OLAD integration, but that is a different topic. Not sure if the poor food/sleep cycle would result in this causing more harm than good. Either way, I am a big fan of the "intervals within intervals" concept and will start using this, as well as the power-bias influenced workouts more often. Input always welcomed.

Scotty Hagnas
07-15-2007, 02:03 PM
We use the Power Bias stuff quite a bit. I detailed a whole different way to break up "Helen" a couple of weeks ago on my site:

"Speed, agility, and peak power are all important attributes to develop if we are chasing well rounded athletic development. How can we plug a bit more of this type of work into our WODs? There are many different possibilities, but one is to break a workout such as the classic “Helen” into mini rounds done in interval fashion.

“Helen”, as the traditional benchmark WOD, is three rounds for time of 400 meter run, 21 kettlebell swings, and 12 pullups. You do this as fast as possible, eventually straight thru without stopping. Under these circumstances, though, your runs will be anything but true 400m sprints. Full speed sprints will need recovery time between efforts. One way to increase the peak power of a workout such as this was outlined by Robb Wolf in an issue of the Performance Menu from last year. Basically, in this version you rest 2 minutes between each exercise. This will allow you to go pretty much all out on each exercise, knowing that you will be getting some rest. The metabolic cost by the end is even higher than if you had done the workout straight thru!

Our version here splits the workout into 9 mini rounds, with 40 seconds of rest between rounds. Each round is a 40yd zig zag sprint out and back, 7 clubbell side swings(or power bomb KB swings), and 4 pullups. I averaged about 38 seconds per round when I tried this. You rest while your partner goes thru his or her round.

The rest period will allow you to put full speed into your short runs, and crank the other drills with max power. This, like the other interval variant, delivers a potent metabolic wallop by the time you are done. Give it a try!"

We've done the dirty 30 this way, also, as well as others.

Scotty Hagnas
CrossFit Portland

Robb Wolf
07-16-2007, 03:02 PM
Great first post and yes, sleep deprivation can crush your performance. We just took a week off over July 4th and we literally slept 12+ hrs per height the whole week. I did No training, walked into the gym and smoked stuff I was struggling with previously...one week back to our 5am wake-up and looong work days and I feel like a 3pack/day smoker.

What is nice about the way you are breaking things up is the intensity is high and the total workload is high but it is not as systemically crushing.

The joys of micro-gym/fitness publication lifestyle!

Good stuff on the Helen modification.

Ron Nelson
07-17-2007, 08:51 AM
I like what you and Scotty have come up with. It reminds me of EDT with a slight power twist. So, it's what I tried yesterday in the gym. Limit time, reps, and rest to maximize power. Go hard with moderate weight, keep reps fast; keep effort at a maximum, or near maximum.

I did this with SGDL's at a moderate weight to avoid injury, and paired it with db push presses (single arm). Did 3 reps of each per round. Limited rest to about 20-40 seconds between rounds. Got 11 rounds in 15 minutes.

Tried the same thing with pull ups and split squats, but got 12 rounds.

Think I'll try the Helen break up one day this week.

Robb Wolf
07-17-2007, 09:36 AM
I think it's a pretty cool way to plow through a boat load of work but keep things short. Is it how you will develop a top tier Olifter or sprinter? Na, but it does shift things more towards the power side of things will allowing for some met-con type elements.

Dave Van Skike
07-17-2007, 10:34 AM
I think it's a pretty cool way to plow through a boat load of work but keep things short. Is it how you will develop a top tier Olifter or sprinter? Na, but it does shift things more towards the power side of things will allowing for some met-con type elements.

truth. density work is inherently power biased and somewhat self regulating

I've never done the alternating form of density training like Staley suggests but am currently trying out a bryce lane style progression for power clean/push press and squats. (ye olde 20/50) Love it. Haven't tested my max yet but I suspect I'll see some increase in my max on squat. Someone more advanced might not get the same reaction bump but I'll bet I get at least a 10 pound bump in my squat.

William Hunter
07-17-2007, 06:16 PM
Tried the PB version of Helen today (400m, 2min, 21 swings, 2min, 12 pu, 2min for 3 rounds) and got nicely smoked without getting flattened. I loved this version, probably b/c I'm notoriously weak, and ended up slogging through OH (original helen), practically shuffling my way through the 400's. My biggest concern/hang up about the CF WOD''s was that in order for me to get a "powerful" workout I had to scale waaaaaay back on the weights in order to keep momentum going. The PB version allows me to keep the weights on the distant side of respectable. It's probably why I've enjoyed dabbling in Staley's EDT from time to time, where I can better manipulate weights and total reps.

Two thumbs Way Up!

Marc Moffett
07-18-2007, 01:41 PM
I wanted to follow up on my previous post which was unavoidably cut short.

It seems to me that real world work (whether it be in the form of recreational sports like basketball or outdoor adventure like bowhunting or simple manual labor) typically has a very complex interval structure which is generated by a basic goal-subgoal psychological structure in conjunction with the complex behaviors required to achieve a given goal/subgoal. So, if my goal is to do X, I must typically do Y and Z. Y+Z together constitute a complex behavior which allows the individual to achieve the goal. Here is a simple example: When I collect firewood, my goal is to load the trailer with logs. Doing this generally requires that I do two things: first, carry the logs to the trailer and, second, lift/load the logs onto the trailer. I generally don't to this all at once, but rather carry a bunch of logs, then load a bunch. Iterating Y & Z constitutes one type of multi-modal interval, an interval which is inherent in the activity itself.

Now, consistent with the power bias picture, these intervals will typically come in macro-intervals. That is, it is pretty common to do a number of reps of Y, rest (optional), do a number of reps of Z, rest, repeat.

But finally, it seems to me that there is frequently (almost always) a kind of micro-interval structure to the subactions Y & Z. In the case of carrying logs it is built into the action itself: carry, walk back, carry, ... But even in the case of lifting logs, I almost never find myself lifting and loading all the logs I have piled up in one long stretch of work. Instead, I work in short bursts with subgoal of loading 4 or 5 logs, resting, loading 3 or 4 more, resting, etc. until the pile is gone and I can go back to carrying some more logs down.

If you think about a wide variety of actions, I think you will find that this pattern is pretty common. In fact, consider somebody who is just starting Crossfit and who is doing a power biased WOD. Since it is unlikely, that this person will be able to do unbroken sets for all or most of the actions they perform, they will naturally add micro-intervals into their work. As the person gets better, the micro-intervals will disappear. But I am not sure that this is a particularly desirable thing. Instead, at least from a functional point of view, a case can be made that the person should at this point (or perhaps prior to this point) up the weight they are using in order to force themselves to break up the sets again. [Actually, if you think of someone just starting Crossfit, they will probably come very close to naturally imposing the power-bias macro-intervals as well...]

OK, now for a bit of speculative theorizing. In the literature on evolutionary biology you will find lots of work on "optimal foraging theory" which basically looks at the way animals forage from the perspective of energy efficiency. I speculate that the three-tiered interval picture is pretty optimal in terms of amount of work done per unit time. I further suspect that this is reflected in our psychology: we naturally break complex actions up into their component parts and perform them serially, we get bored if perform the same action too redundantly, we naturally set micro-goals to achieve, etc.

Robb Wolf
07-19-2007, 07:35 AM
I've had something like this rattling in my head...almost an aversion to things like the standard 3 round couplet or triplet. "constantly varied if not randomized functional movements..." What you mentioned in your post takes this to another level. It is much like the JKD axiom of dispensing with the classical mess. We obviously want to (need to?) quantify and codify...but once we do, once fitness has become a sport...we are playing to a mechanized, non-organic and random construct...Amazingly insightful Marc.

Ron Nelson
07-21-2007, 11:27 AM
Had an epiphany this morning (got up, went to the bathroom, almost went away):

What if you took the 45 reps from Fran and divided the work into either sets of 3x15 or 5x9? Want kind of rest interval would work for a good power bias workout?

Scott Kustes
07-21-2007, 01:35 PM
Ron, CrossFit did an experiment with Fran probably 2 years ago. I recall doing a version of Fran at 5 sets of 9. At the time, my Fran was in the 7-8:00 range. I found that the 5x9 cranked up the metabolic demand because it decreased the rest time. Since I only had to get out 9 thrusters, I could straight set them as opposed to doing 12, resting, doing 5, resting, doing 4 to get to 21 (or however I broke it down) and then doing the same with pullups.

Ron Nelson
07-22-2007, 07:42 PM
I remember doing that. I hated the 45 straight through (took me like a day or so) and didn't fare well with the 3x15.
Would be willing to do the 5x9, but I think the 9x5 would be a little more powerful.

Scotty Hagnas
07-22-2007, 08:03 PM
The 9X5 isn't much fun, but the switch time between exercises does give you something of a break. It's easier to load this version up, though. We recently did this with 135 lb thrusters and tucked front lever pullups.

Scotty Hagnas
CrossFit Portland

Sam Cannons
07-30-2007, 12:33 AM
I have done a 5x10 Fran before wasnt fun.

Joe Hart
07-30-2007, 02:13 PM
Did the 9x5 Fran today. She is the short fat fran. Much likened to mopeds. It was sort of fun right up to the 9th round. Then I had to wake up and chew my arm off.

Robb Wolf
07-30-2007, 02:55 PM
Did the 9x5 Fran today. She is the short fat fran. Much likened to mopeds. It was sort of fun right up to the 9th round. Then I had to wake up and chew my arm off.

That is an amazing sell point!:)

Patrick Donnelly
08-18-2007, 05:45 AM
This seems interesting.

And like it was mentioned somewhere earlier in the thread, I'd be one of the weaklings who ends up breaking up a lot of sets, where it may be better to break between them...