Starting Strength 3rd Ed: The Best Middle School Science Textbook Ever (No, Really)

[TommyGun]

Coach,
Wanted to share this with you. (FYI Big fan of the program and what you have created with the books and SSC's. I always learn a lot.)

So my two kids in 7th Grade were studying for a science test on levers and force, and class 1, 2, and 3 levers. I was an engineer in college and although I hated the courses (being immature), I believe a fundamental understanding of Newtonian physics was crucial. When I first read SS 3rd Ed. years ago, seeing the moment diagrams blew me away, it simplified everything about bar path. I suddenly got it, and I will always be grateful that you included moment diagrams in the book.

So as I helped my kids study, I looked at their science textbook. Mediocre at best in how it described a force acting on a body, and its description of fulcrums were disappointing.

Me: "Kids, wait here, I have a good book we can study"
Kids: "Dad, is it one of your old textbooks?"
Me: "Sort of".

Anyway, we used your Leverage and Moment section on the Squat chapter as a review. God damn wonderfully written and illustrated stuff on fulcrums, force acting on an object, and moment diagrams. Written so clearly that two 7th graders understood what you wrote.

I am ready to bust out Practical Programming when they learn about homeostasis and Selye's Law. Thanks for the great writing and teaching Coach.

[Mark Rippetoe]

I'm happy to hear this. I've always hated shitty teachers, and I've had lots of them. I've tried to learn how to teach better than I was taught.

[Scaldrew]

I've always hated shitty teachers, and I've had lots of them. I've tried to learn how to teach better than I was taught.

Me, too. My science teachers always hated me for some reason, so I hardly cared to learn more than I should have. In contrast, I had great English teachers. They were always nice to me and I respected them and what they had to teach me. My late grandfather even introduced the English language to me and he and I were always close, so that helped. When it was time to think about higher education, I quickly decided on an English degree. It was all I was good at then and the only thing I really cared about. Were it not for all of that, I'd probably be in biology classes or law or something lame like that.

And not to get all mulchy, but if it weren't for this community, I'd probably not give two shits about trying to pass on all the valuable information to others. It's true that strength is worth it for its own sake, but I also feel somehow indebted and want to help in any way I can. If I didn't feel that way, I'd probably just keep all of it to myself. I don't know; I could be wrong about that. I'm not old enough to know myself at all.

[Mulgere Hircum]

Maybe I just don't read the board enough, but I don't know that I've ever seen any comment on the way Rip uses the physics as a pedagogical tool (though I think Steph may have alluded to it once). I think that's actually the most interesting part. SS2 was the first weight training book I had read that discussed the basic physics with anything like accuracy. None of the rest would even pass the laugh test in terms of basic mechanics. I made a couple of comments and was surprised at just how much time and work Rip was putting into the physics for the 3rd edition. I don't know much about the mass-market fitness publishing industry, but I strongly suspect that economics would never justify that kind of care (judging by my other books, publishers want to spend that money on photography).

The usual approach for a popular book, if the physics is not simply ignored, is to present a bit of sloppy almost-mechanics in the first part of the book without it really being necessary for the practical sections that follow (this is actually good, since the author's grasp of mechanics is inevitably wrong in the details anyway). I had one hard-to-find book (on rotator cuff rehabilitation) that did the science section *very* well, but what I didn't fully realize until I took the SS training was that Rip ha gone much further and actually built his pedagogical method around the basic mechanics. That's *much* harder than simply interpreting the physics for a non-technical audience, because it requires actual creative use of the physics, not simply description. SS Coaches in particular really ought to see and appreciate how much went into that part of the class.

I don't know that I ever convinced Rip of my position on why the bar really stays over the mid-foot, but I look forward to an additional section on dynamic stability and control theory in the fourth edition of Starting Mechanics. :-D

[Mark Rippetoe]

I don't know that I ever convinced Rip of my position on why the bar really stays over the mid-foot, but I look forward to an additional section on dynamic stability and control theory in the fourth edition of Starting Mechanics. :-D

Refresh my memory. Let's give it a shot.

[Mulgere Hircum]

Refresh my memory. Let's give it a shot.

OK, I'm home on Christmas vacation *and* trying to lift as much as possible while work isn't in the way, so I'm happy to think about this stuff right now.

Here is what I remember. At the time, your explanation of why the bar *must* move vertically over the mid-foot (barring mechanical limitations as in the bench) was that that path minimizes the effort in lifting a given weight (and so maximizes the 1RM). This was justified by a statics argument involving taking the moment about the mid foot--since that arm is zero on that path, naturally the vertical mid-foot path is the one that minimizes the "total moment" (intuitive notion so we don't have to do calculus) along the path. (Something like a minimum energy path, though I suspect it's more like minimizing the maximum percentage of RM1 at any point along the bar path.) My copy is in storage so I can't easily check, but I think that's the argument of SS2, since that's what I had to read and that's what you were rethinking and rewriting at the time. I argued a quite different line and I think you found some of it convincing, but I'm not sure offhand how much.

My memory is a rather special flower (am I allowed to say that here?) and I remember the salient points we argued then more than I remember how the positions evolved. I also remember that it wasn't until I took the class that I realized how thoroughly you used the physical arguments in the pedagogy. I was sorry there was limited time to revisit the topic afterwards in light of everything I saw that weekend, because it only got more interesting after I realized just why this was important to the program as a whole. I don't think I fully realized that there *was* a larger pedagogical program than just writing books until I took the class so I had a lot to digest at the time. Right at the end you made a fine point about symmetry (that I was kind of professionally miffed not to have seen first) and that was well worth thinking through further. So we never had a chance to discuss your evolved position. I will pull out SS3 and read the discussion of bar path so I can interact with your current, or at least a more recent, line of thinking. And review how much of my argument you bought--I wasn't kidding about not knowing. My wife isn't impressed when I say that all I have to remember is that I'm married and have children and can just look up the names as I need them. I wish that were more of an exaggeration than it is.

[Mark Rippetoe]

At the time, your explanation of why the bar *must* move vertically over the mid-foot (barring mechanical limitations as in the bench) was that that path minimizes the effort in lifting a given weight (and so maximizes the 1RM). This was justified by a statics argument involving taking the moment about the mid foot--since that arm is zero on that path, naturally the vertical mid-foot path is the one that minimizes the "total moment" (intuitive notion so we don't have to do calculus) along the path. (Something like a minimum energy path, though I suspect it's more like minimizing the maximum percentage of RM1 at any point along the bar path.)

This is basically still the argument, along with some newer acceleration mechanics stuff we've developed. But your copy of the blue book is dated -- we've added several sets of corrections and updates to that text over the past 6 years, to the extent that you really need a new copy. E-mail me and I'll get one to you.

[Mulgere Hircum]

This is basically still the argument, along with some newer acceleration mechanics stuff we've developed. But your copy of the blue book is dated -- we've added several sets of corrections and updates to that text over the past 6 years, to the extent that you really need a new copy.

So noted. I was wrong about my SS2 being in storage. I found it, and it looks like the discussion of the mid-foot position is basically prescriptive, there is no real discussion of why it must be so. Stipulating that my copy of SS3 is obsolete, the discussion there is interesting and clarifies why I was a bit uncertain whether we entirely agree or not.

Pages 11-12 are excellent. While the point perhaps only matters to me, the explanation that technically it is the center of mass of the bar-lifter system that must rest over the mid-foot is nicely done, particularly figure 2-4 (as usual, diagrams are the key to clarity). The paragraph beginning at the bottom left of page 12, "The body prefers stability to pretty much everything else," concisely the states the essence of my position, and my claim was and is that this is the *entire* story about the mid-foot position (i.e. efficiency is not a consideration). So to that point it sounds like we're in complete agreement.

The first full paragraph on p13 is where it sounds like we diverge. The discussion of leverage decreasing as the bar moves forward beyond the mid-foot is the part I did and still do disagree with, not that it isn't true (it certainly is) but rather that it is relevant to the necessity of the mid-foot bar position (I claim it isn't). I think optimal leverage is a consideration applied only after (in priority, of course the brain is doing all this simultaneously in time) the mid-foot constraint has been applied strictly on stability grounds. I really think of this in terms of degrees of freedom, an extremely powerful notion that unfortunately isn't a usable concept in a discussion for a general audience. We can try to restate it more simply later after I've made things clear in my own mind.

Whatever the shape of the bar path, it is a simple one-dimensional curve, so one degree of freedom is required to do any lift (movement along that path requires a change in angle of at least one joint). A squat has three, so let's sneak up on it by first considering a simpler lift with a single degree of freedom, say a strict preacher curl to make the picture clear. There is a single degree of freedom, the angle around the elbow, and we need one to follow any path at all, so there is no further freedom for the brain to optimize the movement. Thus there is neither choice of path nor optimization for efficiency; all the brain can do is move the single joint and accept the resulting path.

Interestingly, this is *not* true of a strict standing barbell curl, perhaps with the elbows jammed against the pelvis to guarantee that the shoulder is not involved. Suppose we do the movement slowly, so the body has to stay in balance throughout. What happens? The bar path is no longer completely determined by the elbow angle, because the body moves a bit in the front-back plane. The path is still curved, but less so. What is happening? The ankle angle in the front-back plane provides a second degree of freedom, and the brain is using it to keep the bar-lifter system C.M. over the mid-foot. Exactly the same thing is easier to see with a good morning--the hips go back to maintain the C.M. over the point of maximum stability, the mid-foot. My claim is that it must do precisely this *regardless of efficiency considerations*. If a second degree of freedom is available, it will be used to satisfy the stability constraint. The point I'm making is that given only one choice to make the brain chooses stability, not efficiency. (The other possibility is that the mid-foot leads to the most efficient movement, and so optimization, not stability controls the use of the DoF. I think it's pretty self-evident that this isn't what is going on, but I think it's also not hard to argue against it.)

Now consider the squat. Assuming the back is kept properly locked (using up all of the spinal degrees of freedom in maintaining a neutral position for safety reasons), there are three degrees of freedom available since we have three moving joints, ankle, knee, and hip. As we have seen one is required to make any movement at all, but now we have two degrees of freedom for the brain to choose freely rather than zero (preacher curl) or one (standing curl). So the entire question comes down to what precisely the brain does with the extra degree of freedom. My claim is that stability takes priority and so one degree of freedom is consumed by satisfying the mid-foot stability constraint, regardless of efficiency. The third degree of freedom is indeed chosen for efficiency, but in a a well-executed lift this happens only (logically) "after" applying the stability constraint. That is to say, stability is never traded off for extra efficiency. (People who consistently miss lifts forward could well be doing so because the brain is attempting to trade some stability for efficiency, but I assume we're talking about correct lifts here.)

With that clear (well, at least precise), I can state my objection to the line of reasoning taken on p13. The line taken is that the mid-foot position is the most efficient, but that is in tension with the assertion on p12 that "the body prefers stability to pretty much everything else." If that is the case, as I claim, then it isn't on-point to justify the mid-foot position on efficiency grounds. *The bar would be there regardless of efficiency.* Optimization certainly takes place, but only with the mid-foot constraint already applied. For the efficiency argument to be relevant to the mid-foot constraint, it would have to be true that the correct bar path *could* deviate from the mid-foot position if that were more efficient, so it truly is efficiency keeping it there.

The strongest objection to the above argument is that we all know from experience that missing forward does in fact kill your drive, seemingly providing observational support for the efficiency position (and perhaps motivating the discussion of p13?). Since this is already unreasonably long (my hallmark, unfortunately) I'll defer my answer to that objection, but I hope I've laid out precisely, if perhaps not as clearly or concisely as I'd like, what I think the precise disagreement is. I may be wrong, however, since as I read the text p12 makes a strong statement for the stability constraint dominating, while p13 takes a line that implies that efficiency dominates or at least contributes. To me they are at least in tension, if not at war, and so I cannot be quite confident that I grasp the details of the intended position.

Dustin

[Pawn]

Whatever the shape of the bar path, it is a simple one-dimensional curve, so one degree of freedom is required to do any lift (movement along that path requires a change in angle of at least one joint).

I would say this is the weakest part of your position on this. There are multiple kinematic solutions for the lifts discussed in SS3. This is also inherent variance between a single lifter's movement patterns (day-to-day, set-to-set, and rep-to-rep). If this was not the case then there wouldn't need to be a discussion about technique... simply setup.

About efficiency and the mid-foot... the combined CG of the lifter and bar need to be between the toe and the heel in the hole and at lockout (for standing lifts) -- Newton and all that. The mid-foot argument is describing a potential location for the CG to target during the movement. This description is an excellent tool for visualization and teaching, but (and correct me if I am miss-characterizing) the 'actual' location of optimal efficiency / stability will vary based on anthropometry. (aka not 0.5*LengthOfFoot) Not that that distinction gets you anything in real-world application.

I believe a solid argument implying the link between stability and strength output / the human brain is the lifting belt and its impact on the squat. (Re-you squat more with a belt... but your quads / glutes didn't get stronger... there is a protective / limiting mechanism in there somewhere)

If we want to look at what the human brain is optimizing for in a movement I would argue that you need to look at what happens with things go sideways. (failed reps, 1RMs, form breakdowns) The body moves from position of stability to position of stability until it either becomes completely unstable or it sorts out the problem. Muscles that too weak for a given position get compensated for by other muscle groups which may end up in an out of control system / a viral video on instagram.

[Mulgere Hircum]

I would say this is the weakest part of your position on this. There are multiple kinematic solutions for the lifts discussed in SS3. This is also inherent variance between a single lifter's movement patterns (day-to-day, set-to-set, and rep-to-rep). If this was not the case then there wouldn't need to be a discussion about technique... simply setup.

With respect, it's actually the most indisputable part, but I can guess why it doesn't sound very convincing. At a guess, you are familiar with elementary Newtonian dynamics and understand that F = m*a gets quite tricky when the situation gets complex or when one wants general answers to general questions rather than specific answers to specific questions, but don't have much experience with the bag of tricks to bypass F = m*a and avoid that complexity. Probably you know that conservation laws are very powerful: when they can be applied, they bypass all the complexity of the dynamics and and they give general answers to general questions very well. Degrees of freedom analysis plays a slightly analogous part in applied mathematics in general--the method is so general and so independent of details that when it can be applied it cuts the Gordian Knot of mathematical details (it actually doesn't even depend on the dynamics being Newtonian, so it is *more* general than conservation of mechanical energy), but IIRC tends to only be taught considerably later in the standard curriculum (it gets hammered very hard in Lagrangian and Hamiltonian dynamics junior year in the physics department, and the weeping, wailing, and gnashing of teeth tends to be audible all the way across campus). My conclusions from DoF analysis do not depend on anthropometry, bar path, or really much of anything except that only three joint angles are changing throughout the lift. During a well-performed lift that is more than accurate enough for accurate answers. This is analogous to the fact that the work a lifter does against gravity is completely independent of bar path and can be computed trivially from the mass and the change in elevation--the details are simply bypassed by using the deeper property of conservation of energy. The bottom line is that unless another joint moves or a bone breaks, the number of degrees of freedom is an absolute constant of the motion just as energy is, and conclusions solely from DoF considerations are true regardless of just about everything. It's truly a somewhat magical (so it seemed when I learned it) "cuz math" kind of thing that pulls answers seemingly out of thin air, much like dimensional analysis occasionally does (that's a third tool in the toolbox for avoiding having to try to solve F = m*a or make it yield general answers).

This illustrates why it's not entirely fair of me to use degrees of freedom analysis in a forum post. The argument is hard for anyone who hasn't had quite a bit of mathematics to interact with or evaluate. I did so anyway because it is such a powerful way of finding the correct answer, with the proviso that it might have to be re-formulated for easier consumption after the correct answer is known. Apologies if it muddles the waters instead.

About efficiency and the mid-foot... the combined CG of the lifter and bar need to be between the toe and the heel in the hole and at lockout (for standing lifts) -- Newton and all that. The mid-foot argument is describing a potential location for the CG to target during the movement. This description is an excellent tool for visualization and teaching, but (and correct me if I am miss-characterizing) the 'actual' location of optimal efficiency / stability will vary based on anthropometry. (aka not 0.5*LengthOfFoot) Not that that distinction gets you anything in real-world application.

The above is true, and actually is fundamental to the argument that really convinced me that stability, not efficiency is the real answer. Essentially, if the answer were really efficiency then it would depend to a degree on individual lever lengths and muscle strengths, and so *someone* would end up lifting with the bar displaced a bit. See below.

I believe a solid argument implying the link between stability and strength output / the human brain is the lifting belt and its impact on the squat. (Re-you squat more with a belt... but your quads / glutes didn't get stronger... there is a protective / limiting mechanism in there somewhere)

I completely agree that there is a limiting mechanism, and in fact that's the answer I alluded to about why going off-track kills the drive if it isn't leverage. My believe is that the brain shuts it down in an attempt to save stability. I guess this is a reasonable time to make that argument:

I was put on this trail by Rip's explanation of why you cannot necessarily tell when grip strength is limiting your deadlift--the brain refuses to let you tear a weight out of your grasp. I didn't know that, and it seemed likely to be one aspect of a more general rule. For example, I suspect that the reason the biceps are much less active in a pull-up than in a chin is more or less the same thing--the brain will not let you twist your support out of your grasp. In any event, I realized that it's likely that the reason you lose strength on a lift that goes too far off-track is not actually loss of leverage (though of course that can happen as an independent factor, especially to the front), but rather than the brain immediately shuts down the contraction to avoid further going out of balance. I think this is corroborated by the fact that, if anything, leverage really ought to improve as the weight moves back toward the heels, certainly Rip's moment-arm argument implies this. But we know that the brain doesn't really let you do this (to a reasonable error, perhaps to within at least a cm or two), and in fact it seems to try very hard to make you fail forward rather than backward. It wants to be maximally stable, but also tries to bias any errors forward. It doesn't make sense for it to prefer erring in the weaker direction on leverage grounds.

Having gotten a clue as to the direction of the answer, I thought it through to the degree of freedom analysis just given. At that point, I had a genuine testable prediction--if efficiency is the right answer, there is bound to be *someone* who doesn't fit the rule (this is in addition to the fact that I rather suspect that everyone actually has slightly better leverage with the CoM over the heel and doesn't depend on whether that happens to be correct). But if stability is the answer then there will not be, because the DoF analysis shows that it doesn't depend on anthropometry in any way. I of course have no data set to test this, but I didn't need one, just a coach with wide experience. So I asked Rip if anyone ever has anthropometry so strange that they hold the bar anywhere but the mid-foot position, whether gibbons or T-Rexes, dachshunds or giraffes. His answer was basically that this never happens--gibbons, T-Rexes, dachshunds, and giraffes all lift over the mid-foot. (Admittedly he didn't actually tell me about his experience training zoo animals and extinct creatures, but I could read between the lines. :-) That would be less weird than the typical gym clientèle I saw before I bought my own rack, and I'm sure everyone here is too polite to suggest that my private little home gym sees a similar parade of weirdness every time I use it.) I think that's a reasonable test of a reasonable prediction, and it supports the stability theory.

That may be a better explanation than I managed back when Rip and I were tossing this back and forth, we'll see.

If we want to look at what the human brain is optimizing for in a movement I would argue that you need to look at what happens with things go sideways. (failed reps, 1RMs, form breakdowns) The body moves from position of stability to position of stability until it either becomes completely unstable or it sorts out the problem. Muscles that too weak for a given position get compensated for by other muscle groups which may end up in an out of control system / a viral video on instagram.

Yes, you can see me using lift failure to advantage in the argument just made. You can take it further--when form breaks down the lifter may no longer have a symmetrical posture (I fight not doing this in the hole) and now he'd be modeled as an 8-DoF system, but you can still get some useful analysis. For example, when the squat is heavy and I get sloppy I sometimes lose symmetry by having my knees both move in the same direction, but the bar doesn't really move at all. Granted that it's got a lot of inertia, I believe what is happening is that my brain is attempting to use the extra hip degrees of freedom that I ignored earlier (if the knees stay out the hip is constrained to a single DoF even though in general it has two) to optimize force production--but it only does so under the constraint that the bar stays in balance, so it doesn't (intentionally) break the mid-foot constraint. If I kept trying to go up instead of staying or resetting back a bit to force the form to stay correct then I imagine eventually that would indeed get out of control. I think that shows that even when form breaks down as the brain hunts for the optimal solution, it really doesn't want to break the mid-foot constraint. It's determined not to be fall over holding heavy weights, and we're all grateful that this is so.

Dustin