starting strength gym
Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: Training Optimality

  1. #1
    Join Date
    Nov 2009
    Location
    Texas
    Posts
    3,111

    Default Training Optimality

    • starting strength seminar april 2024
    • starting strength seminar jume 2024
    • starting strength seminar august 2024
    A Study Proposal from an Economics Perspective

    by Nicholas M. Racculia

    “The motivation for discovering optimal programming is rational. Though our life span is the longest thing we experience for sure, it is relatively short in terms of the limited number of training sessions available. We want to get the most out of each training session. Getting strong or increasing power or endurance should happen as efficiently and safely as possible.”
    Article

  2. #2
    Join Date
    Sep 2014
    Location
    Champaign, IL
    Posts
    469

    Default

    The pursuit of optimal training efficiency over a lifetime, or lifting career, seems rational. But if you’re going to limit your definitions to just the novice stage of a lifting progression it seems a bit extraneous.

    Hypothesis: The Starting Strength Novice Program is Optimal with respect to Strength acquisition.
    ... The goal of each experiment is to reject the hypothesis.
    I’m probably just missing something obvious but why aim to prove something about predictable maximal efficiency for novices? It just seems like a very high bar (hurdle not squat) when a much lower one would suffice.

    Suppose you can demonstrate a Starting Strength progression gets 70% of trainees in a given demographic group to 225/315/405/495 within 18 months. I would think a lot of serious trainees would trade the potential of a better program for a high degree of certainty in results. People make analogous choices with their financial investments all the time.

    It might be more productive to quantify what your protocol does, let ACSM demonstrate that their program is better. I would think there are enough novice trainees with certified coaches already to yield some reliable data. Maybe this has already been done, or maybe no one cares to do it.

    Thanks for an interesting article.

  3. #3
    Join Date
    Feb 2012
    Location
    Greensburg, PA
    Posts
    400

    Default

    Thanks for the comments AJVD, some thoughts:

    Quote Originally Posted by ajvd View Post
    The pursuit of optimal training efficiency over a lifetime, or lifting career, seems rational. But if you’re going to limit your definitions to just the novice stage of a lifting progression it seems a bit extraneous.
    There are some important and useful reasons to limit (at least at first) the analysis to the novice stage.

    First, if the sample size is large enough, as long as the population is composed of untrained individuals, then the characteristics between each comparison population (Protocol A and Protocol B) should be approximately equivalent. I think once you introduce intermediate and advanced lifters it becomes too complicated to disentangle confounding effects.

    Second, any linearity or quasi-linearity that is quantifiable during a novice phase becomes much more difficult to track when introducing more complicated programs like Texas Method. I don’t think it’s impossible, but let’s establish a first best path to get to intermediate levels of strength first.

    Third, I selected novice because I believe that optimal should also be non-detrimental with respect to other aspects of fitness. For example, suppose a program which promotes the quickest path to maximal endurance seriously depletes a person’s strength or power levels. Is that really optimal fitness for an average person??

    Finally, the novice stage is the most important stage of development for the average person. It's the novice stage that allows us the most progress per workout, so why not maximize results here first?

    Suppose you can demonstrate a Starting Strength progression gets 70% of trainees in a given demographic group to 225/315/405/495 within 18 months. I would think a lot of serious trainees would trade the potential of a better program for a high degree of certainty in results. People make analogous choices with their financial investments all the time.
    I think the financial analogy works well, but only so far. Investing in our “physical 401k” for smooth future physical activity is parallel to investing in our 401k for post-retirement consumption smoothing. But the analogy breaks down when you look at the nature of risk and reward in financial markets versus health and fitness. In finance, risk and expected return are positively related…in other words, to entice an investor to purchase a junk bond (rather than a safer AAA-rated bond) there must be an expectation of earning a higher yield. This higher expected yield comes with higher uncertainty (from the risk of default); it’s the uncertainty that drives the expected returns.

    For physical returns, higher uncertainty is inversely related to higher returns. The high expected returns of Starting Strength flow directly from the predictability of results. Contrast this with the unpredictability of, say, Crossfit programming (or really any programming that doesn’t allow for sustained progress). Are many 500 lb deadlifts produced with random programming? Contrast this with the number of 500 lb deadlifts at Matt Reynolds’ Strong Gym.

    In fact, that’s exactly my point when I say we should be measuring programmatic volatility. If a program claims elite fitness results but the actual results deviate so far from the predicted path (even when members follow the program as described) then I would consider this a bad program with a low degree of certainty of results.

    So if we can measure programmatic volatility, then I think we can arrive at both a “better program” and a “high degree of certainty of results.”

    It might be more productive to quantify what your protocol does, let ACSM demonstrate that their program is better. I would think there are enough novice trainees with certified coaches already to yield some reliable data. Maybe this has already been done, or maybe no one cares to do it.
    It would be fantastic if there was a thoughtfully designed database with granular data that tracked results over time. In fact, I would argue that such a database would be a game changer. We would be limited only by the creativity of the questions we could ask it. If only…

  4. #4
    Join Date
    Sep 2014
    Location
    Champaign, IL
    Posts
    469

    Default

    Finally, the novice stage is the most important stage of development for the average person. It's the novice stage that allows us the most progress per workout, so why not maximize results here first?
    I didn't intend to suggest a harder study of intermediate/advanced progression, just that proving something about efficiency wasn't as important for a necessarily brief progression.

    I tend to think of the novice stage as brief by definition, and the Starting Strength novice program as sub-optimal by design. (Specifically I’m referring to linear increases in strength where the theoretical potential for exponential increases exists.) In my mind, the most important factor of a novice progression is that trainees reliably complete it. In this short phase I would put more value on reliability than efficiency, in the longer training progressions efficiency would have a much larger effect.

    After another read I’d also say your definition of training optimally includes this point.

    It would be fantastic if there was a thoughtfully designed database with granular data that tracked results over time. In fact, I would argue that such a database would be a game changer.
    It is an interesting situation. You have a fairly strenuously certificate group of coaches all administering a relatively standard program to novices in a wide range of demographic groups. All of these novices are presumably required to keep a log book. It is almost as if the only thing standing between you and that data is the fact that it’s logged on paper and not a smartphone.

  5. #5
    Join Date
    Feb 2012
    Location
    Greensburg, PA
    Posts
    400

    Default

    Quote Originally Posted by ajvd View Post

    I tend to think of the novice stage as brief by definition, and the Starting Strength novice program as sub-optimal by design. (Specifically I’m referring to linear increases in strength where the theoretical potential for exponential increases exists.)
    AJVD, have you ever completed a linear progression? I ask this sincerely, as anyone who has would never think that any sort of exponential increases could possibly exist.

    What theory are you referencing that suggests this? I can't imagine a rate of return that would generate better returns than linear progression (over the same interval) and could be maintained (with compounding) for any length of time. Can you?

    I even say that SS is quasi-linear, since as we get closer to the end of our novice progression, the jumps become smaller, the law of diminishing returns takes hold. But is this suboptimal?

    Think of an average guy who follows the SS program (no idiosyncratic volatility). Let's say on Day 1 he squats 135 lbs for a work set. If he is able to follow linear progression according to the SS model he will have squatted 3 times a week for 4 weeks for 3 months at 5 lb increases per workout (for simplicity let's say there are no 10 lb initial jumps). Three months from now he will be squatting 315 lbs for sets across. Can you describe me a program that gets you to 315 faster? or that gets you to increase your squat by more than 180 lbs in three months?


    Quote Originally Posted by ajvd View Post

    In my mind, the most important factor of a novice progression is that trainees reliably complete it. In this short phase I would put more value on reliability than efficiency, in the longer training progressions efficiency would have a much larger effect.
    I think you are referencing compliance rather than reliability. This is what I'm trying to separate with the two expressions of volatility (programmatic and idiosyncratic). Measuring "idiosyncratic volatility" really measures compliance, doesn't it? And I'm arguing that a prioi the levels of idiosyncratic volatility should be roughly the same from one program to another (though I think we can relax that assumption once the data are collected).

    "Programmatic volatility" is just the program produces the results it promises. Reliability here is (or implies) efficiency. Please do not take the finance analogy too far (reliability and efficiency are not tradeoffs here), it breaks down in this aspect (and other aspects)

    See the difference?

  6. #6
    Join Date
    Sep 2014
    Location
    Champaign, IL
    Posts
    469

    Default

    Nicholas, Thanks for engaging with me, I do appreciate it. I think I understand what you're saying. Briefly what I'm getting at is just this:

    Three months from now he will be squatting 315 lbs for sets across. Can you describe me a program that gets you to 315 faster?
    My point is that the value of the program is not that it gets you to 315 in 3 months, but that it can get lots of people to 315. Reliability is worth demonstrating regardless of efficiency, and it's probably simpler to prove anyway.


    I may be very mistaken in my understanding of the SS model. Maybe this picture will illustrate what I'm trying to get at.




    The picture is my poor attempt to replicate the Strength Performance curve found in many of Rip's books, let's say figure 8-1 in PPST3. Again I may be very misinformed but I'm trying to make two points:

    1) There is a space, indicated by a green arrow, for a faster increase in strength than a linear progression. No doubt any program in that space would be very difficult and perhaps practically unattainable. But by definition the space is there and if we looked we shouldn't be surprised to find it.

    2) Because the novice phase is so brief, it's not important that you take either the red line or the green line, it is important that you eventually hit the purple line. The most substantial problem with a a lot of training programs is not that they take too long to make you strong, it's that they don't make you strong at all even if you are perfectly compliant.

    AJVD, have you ever completed a linear progression? I ask this sincerely
    In terms of my own training experience I've found that the pursuit of the optimal program is an easy way to waste a lot of time. It's much better to find something that is working and has worked for others, and then just be patient. To my limited understanding this patience is baked-in to a SS novice progression - a feature not a bug.

    I have completed a linear progression, probably not very well. I didn't drink a gallon of milk a day though so I believe YNDTP applies to me.

  7. #7
    Join Date
    Apr 2009
    Location
    England
    Posts
    733

    Default

    Hope you don't mind me laying this actual drawing on the table, taken from PPST3 and overlaid with 3 coloured lines.



    The novice phase is made up of linear progressions which track very closely to the strength increase curve in the diagram. There is no 'space' available in which to speed up the progression.

    Within the novice phase itself the gains slow down, so the red line could represent the fastest strength increases using 3x5 3 times per week, you may introduce a light day by the orange line and back off sets by the yellow line. The rate of gain is slower but still linear, and these adjustments are needed to reflect the slowing down of the possible rate of strength gain as one gets stronger.

  8. #8
    Join Date
    Feb 2012
    Location
    Greensburg, PA
    Posts
    400

    Default

    Happy to talk with you about it.

    Quote Originally Posted by ajvd View Post
    My point is that the value of the program is not that it gets you to 315 in 3 months, but that it can get lots of people to 315. Reliability is worth demonstrating regardless of efficiency, and it's probably simpler to prove anyway.
    These are not mutually exclusive characteristics. I'm defining reliability as the program actually produces what it says it will produce. You are defining reliability here as getting people to show up. How about we call that compliance? The experiment I'm proposing can (and should), as a corollary, test for compliance. This is the very idiosyncratic volatility I'm defining.


    Quote Originally Posted by ajvd View Post
    2) Because the novice phase is so brief, it's not important that you take either the red line or the green line, it is important that you eventually hit the purple line. The most substantial problem with a a lot of training programs is not that they take too long to make you strong, it's that they don't make you strong at all even if you are perfectly compliant.
    The length of the novice phase is a function, in an ideal world, of genetics (if all other variables like diet, sleep, form, willpower, etc. are perfectly measured and attained) and it varies from person to person. We all want our novice phase to be as long as possible, and it is probably realistically longer than most people believe it to be. That said, why on earth wouldn't we want a program that maximizes returns on time and effort invested?

    Resources are scarce. Money is scarce. Time is scarce.


    Quote Originally Posted by ajvd View Post
    In terms of my own training experience I've found that the pursuit of the optimal program is an easy way to waste a lot of time. It's much better to find something that is working and has worked for others, and then just be patient. To my limited understanding this patience is baked-in to a SS novice progression - a feature not a bug.
    An experiment in place that weeded out non-optimal programs allows for you to not waste your time. That's the entire point of the paper. I'm conjecturing here, but I can't imagine a programs that makes me stronger faster. And with more reliable results. See what I mean?

    I see the graph you are drawing from (thank you Dan Lightbulb). Make sure you understand that your blue line demonstrates diminishing marginal returns. The "linearity" is really not linearity in the truest sense, but quasi-linear. Inter-workout weight jumps begin big, then drop to medium, then to small and then finally it takes longer than that brief respite will allow...that's the definition of the end of the novice phase. That's when more complex programming becomes necessary.

  9. #9
    Join Date
    Aug 2012
    Location
    Moscow, Russia
    Posts
    699

    Default

    Quote Originally Posted by danlightbulb View Post
    Hope you don't mind me laying this actual drawing on the table, taken from PPST3 and overlaid with 3 coloured lines.



    The novice phase is made up of linear progressions which track very closely to the strength increase curve in the diagram. There is no 'space' available in which to speed up the progression.

    Within the novice phase itself the gains slow down, so the red line could represent the fastest strength increases using 3x5 3 times per week, you may introduce a light day by the orange line and back off sets by the yellow line. The rate of gain is slower but still linear, and these adjustments are needed to reflect the slowing down of the possible rate of strength gain as one gets stronger.
    Do you realize that your 3 colored lines can't represent actual progress?

    Once you get to the right of the "ideal" adaptation curve, your best progress from that point is the same adaptation curve but now shifted to the right. Therefore you have no way to get back to the original ideal curve once you get out of it, as you can't exceed maximum rate of adaptation at any given point in your training. Speaking more math, the first derivative of an adaptation curve shouldn't get to the right of the "Rate of Adaptation" curve.

    On your drawing you need to cut every straight line when is starts to get closer to the optimal curve, draw a copy of the original curve shifted right to run through this new point, and start next linear line from there, once again below the newly drawn curve. You will finally get similar curve to the original, but made up from straight line, and that reflects slower progress all the way.

    That said, it's not evident that the lines should be straight in the first place, as It's the weight on the bar that is increased linearly in LP, not actual trainee best performance.

  10. #10
    Join Date
    Apr 2009
    Location
    England
    Posts
    733

    Angry

    starting strength coach development program
    Quote Originally Posted by osv2 View Post
    Do you realize that your 3 colored lines can't represent actual progress?

    Once you get to the right of the "ideal" adaptation curve, your best progress from that point is the same adaptation curve but now shifted to the right. Therefore you have no way to get back to the original ideal curve once you get out of it, as you can't exceed maximum rate of adaptation at any given point in your training.

    Yep totally agree. The curve represents ideal performance. The straight lines I drew represent the program if done perfectly and if you even miss one workout then that would shift the line ever so slightly to the right, permanently, given, as you say, that the rate of catch up could never exceed the ideal rate.

    Quote Originally Posted by osv2 View Post
    That said, it's not evident that the lines should be straight in the first place, as It's the weight on the bar that is increased linearly in LP, not actual trainee best performance.
    True but since weight on the bar is what we can measure then I think we can make the two equivalent? I guess the graph is really showing 1rm which should be closely correlated to the 3x5 rm? Anyway the point is that one would struggle to get any more efficient than the program in strength gains.

    It would be interesting to take dates and squat weights from the novice logs here and plot actual curves. Maybe someone with time on their hands could design a google docs spreadsheet capture tool. I may have a go if no one else does.

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •