Strength Training Manual: Planning – Part 1

I am very happy to announce that I am finishing the Strength Training Manual. I decided to publish chapters here on Complementary Training as blog posts for two reasons. First, I want to give members early access to the material. And second, this way I can gain feedback and correct it if needed before publishing it.

I look forward to hearing your thoughts.

Enjoy reading!

Before jumping into deep waters of strength planning, it is important to cover a few “Small World” models of the training dose and training response that are lurking behind all of our planning decisions.

Training dose

To make it distinctive from the term training load, which I have used to refer to the weight on the barbell (see Figure 4.1), I will use the term training dose to indicate a construct that represent some type of stress and/or stimuli that athlete experiences when training for strength (or training in general). That being said, it is really hard to have a precise definition of a training dose and quantify it. It is particularly short-sighted to represent training dose with a single metric. Thus, I will represent the pluralist viewpoint by using multiple “Small World” models as potential tools.

Figure 5.1 contains hypothetical components of the training dose construct (keep in mind that this is also a “Small World” representation).

Figure 5.1. Training dose components

Intensity, Volume and Density

Simply, Intensity as a component of training dose represents quality, while Volume represents quantity. As explained in Chapter 4, intensity represents a complex interaction of (1) Load, (2) Exertion, and (3) Intent. For example, lifting 100kg (here 100kg could be considered “intensity”) for 5 reps with maximal intent versus submaximal intent would represent qualitatively different training intensity and thus qualitatively different training dose. The same can be said for lifting 100kg for 3 with 3 reps in reserve, versus 1 rep in reserve. Intensity is usually represented with the average relative intensity metric (aRI) introduced in Chapter 4. Using relative intensity allows for comparison between exercises and individuals with different 1RM. For intra-individual monitoring of intensity, one can use average intensity (aI), which represents absolute load (in kilograms or pounds).

Volume represents quantity or amount of work done. As explained in Chapter 4, volume is usually expressed through number of lifts (NL), tonnage, impulse and INOL metric (as well as with novel metrics such as exertion load). In an ideal world, volume would be represented with work done in Joules, but mentioned metrics are satisficing proxies.

Intensity and volume can be combined by using “zoning”, or providing volume metrics per intensity zone. For example, NL in 70-80%, 80-90% and 90%+ 1RM zones. But, a workout consisting of 3 x 10 @70% will have the same volume indices (and volume distribution) as workout consisting of 10 x 3 @70%, although we know experientially that these are qualitatively different. Chapter 4 introduced the novel “exertion load” (XL) metric which gives non-linear weighting of the reps depending on their proximity to failure (using RIR).

Volume metrics tend to use intensity cut-off point (e.g., not counting reps below certain %1RM) , which is usually around 50% of 1RM for grinding lifts. This depends if one uses dynamic effort method and wants to keep counting reps under 50% 1RM. It is thus important to clarify what this threshold is. For example, if someone says that the weekly NL for bench press was 50 lifts, it is natural to ask “What counts as a lift?” or in other words asking about the intensity cut-off point. The same is true for any other volume metric.

What about finishing 5×5 @80% workout in 10 minutes versus 15 or 20 minutes? These would have the same volume and intensity, but they would have different density. Mathematically or physically expressed, density can be considered a proxy to average power, since it is work done (or proxy metric to work done) divided by the time it takes to complete the work. As mentioned in Chapter 4, density metrics are not really common, but they could be particularly used in the Mongoose Persistence methods (e.g., muscular endurance, power endurance). The concept of density is also crucial element in Charles Staley’s Escalatory Density Training (ESD) method (Staley, 2005).

Within quality saturation-distribution

Besides the above use, density is an interesting component of a training dose, particularly because it depends on the time frame, and thus expands into the concept of distribution of the training dose as well as frequency of training sessions (e.g., among how many training sessions a certain training dose is distributed). For example, in a given session one can distribute all particular lifts of one exercise into one time block (e.g., 5×5 @75% of Squats), or combine multiple exercises in a superset or a circuit fashion. In motor learning and skill acquisition, this is termed blocked practice versus random practice (Davids, Button & Bennett, 2008; Renshaw, Davids & Savelsbergh, 2012; Chow et al., 2016; Farrow & Robertson, 2017). Blocked practice involves solving one particular task or performing single particular skill for blocked period of time. Random practice refers to randomly solving or performing multiple skills and tasks. It has been shown that from skill retention perspective¹, random practice is better. This could be a useful tip when coaching someone the basic skills of lifting (i.e., some type of superset, quad-set or even circuit could be better from motor learning perspective).

Across multiple days, training dose aimed at the given quality (or method) can be distributed or saturated (see Figure 5.2). I like to refer to this as distribution-saturation continuum or complementary pair.

Figure 5.2. Saturated-Distributed continuum of distributing training dose (using NL metric) of a single quality

Mike Tuchscherer refers to this as frequency-based (e.g., 2nd row in Figure 5.2 which utilizes distributed or complex loading) versus fatigue-based (e.g., 4th row in Figure 5.2 which utilizes saturated or unidirectional loading) training cycles (Tuchscherer, 2008). In Figure 5.2, I’ve used NL as a training dose metric, and assumed zero NL on days without training². Then I calculated common summary statistics: (1) Total NL (i.e., sum of NL for a week), (2) NL per day (i.e., average NL per day), (3) SD NL, or NL standard deviation, (4) CV, or coefficient of variation (which is equal to SD divided by the average NL), (5) Monotony, which is equal to the average NL divided by SD, (6) Strain, which represents Total NL multiplied by Monotony, assuming that monotonous training increases strain for the same training dose, and finally (7) Gini Coefficient (or Index), which represents inequality metric³. Please note that these metrics refer to a single quality or method solely, rather than overall training (e.g., the NL used in this example could be NL in the back squat, or total NL in the upper body pressing movements⁴).

All these metrics are useful, but in this particular example, Gini index represents a metric of saturation of a training dose (at particular quality, assuming NL represents dose). Note that monotony and strain are higher for the distributed examples while Gini Index is lower. The opposite is true for the saturated example (Gini higher, strain and monotony lower).

Some evidence-based lab coats would want us to believe that there is the optimal frequency and distribution to maximize strength or muscle mass gain. I buy that story only as a useful “prior” or heuristic, but keep in mind that there are non-linear effects and interactions. This is particularly evident when multiple qualities or methods are involved in the game. All examples in Figure 5.2 have 40 NL, but we can’t assume that this dose will create equal response in all examples, even if the intensity is kept constant across days. If we add changes in intensity we will create a havoc in training dose metrics. Moreover, a hard lower body session might negatively affect next day upper body session. Also, someone might prefer and flourish on “microdosing” (top row in Figure 5.2) on every day, while someone else might need a single heavy or two medium-heavy days back to back to induce response (adaptation). This can also be exercise-dependent as well (e.g. squat might thrive on 2x per week, while bench press needs at least 3-4x per week sessions). Also keep in mind the “real life” thing, where someone might not afford to lift frequently, or can’t afford big training doses on a particular day, either due to job requirements or due other skill training (e.g. professional soccer player), or can find only 30-40min to exercise (see Top Down vs. Bottom Up in Chapter 2). Remember the saying “In theory, there is no difference between theory and practice. But, in practice, there is” which lab coats don’t seem to get, but rather prefer mental masturbation of finding the “optimal” distributions and securing tenure and fundings.

The very useful approach, particularly for those who are able to implement this strategy, is to alternate between frequency-based and fatigue-based cycles over a longer period of time (e.g., 6 months). And as already stated, this can be implemented on different levels (e.g., qualities, methods, exercises and so forth). See Table 5.1 for an example:

Table 5.1. Alternating between frequency-based (4/week) and fatigue-based cycles (2/week) over a longer period of time using upper body pressing and squat movements. Weekly dosage (using NL metric) is equal across cycles (60 NL in a week). Fatigue-based and frequency-based distribution can be applied to both exercises in this example (Cycles 1 and 2), or individually (Cycle 3 and 4).

“Saturate and Separate” heuristic and “complex/parallel and unidirectional” continuum

So far we have mentioned only single quality and it’s a dose (e.g. bench press NL). The problems start when there are multiple qualities that need to be developed (which is usually the case). Then, the non-linear effects and interactions become really complex and finding the “objective” optimum is only lab coats’ wet dream. We need simple heuristics, and one such commonly used heuristic is the following: “The higher the level of the athlete, the higher the dose, the higher the saturation of a given quality (relative to other qualities) and separation of the developmental loads that is needed to squeeze out response and adaptation” (see Figure 5.3). I call this the Saturate and Separate heuristic. I will explain this heuristic, but I am not a huge believer in it, although I find it an interesting tool rather than the factual state of reality.

Figure 5.3. According to “Saturate and Separate” heuristic, as one makes progress, training dose at a particular quality should increase, together with saturation of that quality relative to other qualities

Here is how the “Saturate and Separate” story goes. As a novice, one can train multiple qualities at the same time⁵, since the dose that is needed to induce adaptation is low. The combined dose (of multiple developmental doses aimed at different qualities) is thus tolerable and there is less negative interaction (see “Novice” on Figure 5.3). As one improves, developmental loads (training dose that is needed to induce adaptation) becomes higher, and thus one needs to start saturating and separating developmental doses. In the beginning this can mean having designated day to emphasize development of a given quality, and this can extend to week emphasis for more intermediate athletes (see “Intermediate” on Figure 5.3). At the highest levels, one needs to saturate and separate for a longer period of time (e.g. focus on improving your squat for a month; see “Advanced” on Figure 5.3). This reasoning is perfectly outlined in books by Mark Rippetoe: “Starting Strength” and “Practical Programming”(Rippetoe & Kilgore, 2011; Rippetoe, Baker & Bradford, 2013), which I consider must reads.

The “saturate and separate” heuristic is the basis of the Block Periodization concept (Issurin, 2008a,b, 2013, 2019), where with the higher-level athletes, developmental loads of the compatible qualities needs to be saturated in the separated blocks. Please keep in mind that “saturating” in this case means “emphasizing” development of a few compatible qualities. This emphasis can be done with both frequency-based and fatigue-based approaches. For example, let’s take Table 5.1, and implement alternating emphasis (in this example by using higher weekly NL) of either upper body press or squat movements: