Deep divefrom Episode 4Training3 October 202616 min

Hold It Until It Wins — Slow Yielding vs Lifting

A deep dive into one paper: what was known before, what the researchers did, what the numbers say and what they don't.

The paper: Comparing six weeks of full-body eccentric quasi-isometric and dynamic resistance training on muscular size, physical performance, and serum hormones in untrained men (Scientific Reports)

MP3

Chapters

Transcript

The script the two AI voices read from. The times are worked out from the script, not heard, so a line can start a few seconds early or late.

Read the whole episode

Cold open

Host 1You heard the trailer — this is the whole story. Welcome to the Hastrup Review of Science — where artificial intelligence reads real science, so you don't have to pretend you understood the paper. One paper today: thirty young men, six weeks, and a style of training where you hold a weight until you can no longer hold it, and then keep resisting as it slowly wins.

Host 2The paper is Comparing six weeks of full-body eccentric quasi-isometric and dynamic resistance training on muscular size, physical performance, and serum hormones in untrained men. The authors are Reza Nemati, Tohid Romian, Kazem Khodaei, Mohamadreza Zolfaghar Didani and Dustin Jay Oranchuk, and it was published in Scientific Reports in September twenty twenty-six.

Host 2One thing to say up front. We could not retrieve the full text, so this episode is built on the published abstract plus established background from the field, and I will flag which is which.

Host 1So where the abstract is silent, we say so.

Host 2Yes. That matters here, because several of the details you would want — which exercises, how long each hold lasted, how muscle size was measured — are not in the abstract.

Why this field exists

Host 1Start with the name. What is an eccentric quasi-isometric?

Host 2This is background, not from the paper. An isometric contraction is one where the muscle produces force but the joint does not move, like holding a half squat. An eccentric contraction is one where the muscle produces force while lengthening, like lowering a weight under control.

Host 2An eccentric quasi-isometric, usually shortened to E-Q-I, joins the two. You hold a position for as long as you can, and as fatigue sets in you are not allowed to drop the load — you resist while the joint slowly gives way, until you reach a deep, lengthened position.

Host 1So a set is really one very long repetition.

Host 2Typically, yes. A single effort that can last well over a minute, often with bodyweight or a light load, ending with the muscle stretched and fully fatigued.

Host 1Why would anyone train that way instead of just lifting?

Host 2The abstract gives the main reason: E-Q-I has gained popularity as a way to promote muscle adaptation when traditional resistance training is contraindicated. Think of someone coming back from injury who cannot tolerate fast or heavy loading, or someone with no equipment.

Host 2In the background literature, the appeal is that the load stays low, the movement is slow and controlled, and the muscle still spends a long time under tension and ends in a lengthened position. All of those are features researchers associate with strength and size gains.

What was known before

Host 1What did we already know before this study?

Host 2Background again. Ordinary dynamic resistance training — lifting and lowering a load through a range of motion — is the most studied intervention in exercise science, and in untrained people it reliably increases strength and muscle size within weeks.

Host 2Isometric training is also well studied. It increases strength, but the gains tend to be largest near the joint angle you trained at, and training at longer muscle lengths seems to transfer more broadly than training at short lengths.

Host 1And eccentric training?

Host 2Eccentric training is known to produce strength and size gains, and lengthening work has been associated with increases in the length of muscle fibre bundles. E-Q-I borrows from both lines of evidence, but it is a specific hybrid, and direct training studies on it have been scarce.

Host 1The senior author's name has come up before in this area.

Host 2Dustin Oranchuk has written several earlier papers on isometric and eccentric quasi-isometric training, including concept papers and short-term studies. The abstract itself puts it plainly: evidence on the chronic effects of E-Q-I remains limited.

The open question

Host 1So what is the open question?

Host 2Whether a programme built entirely on E-Q-I, across the whole body, produces the same adaptations over weeks as conventional lifting. Not one muscle in a lab, but a full training plan.

Host 2And the authors asked about more than muscle. They looked at size, strength, power, endurance and blood hormones together, which is a broad net for a small trial.

Host 1Who would that answer matter to?

Host 2Physiotherapists and coaches choosing tools for people who cannot lift normally, and anyone training with little equipment. If E-Q-I matched lifting, it would be a reasonable substitute in those situations.

Methods: the trial

Host 1Walk me through the design.

Host 2Thirty untrained men, aged eighteen to twenty-six, were randomized into three groups of ten. One group did full-body E-Q-I training, one did dynamic resistance training, and one was a control group that did not train.

Host 2Both training groups trained three times a week for six weeks. That is eighteen sessions in total, assuming full attendance, which the abstract does not report.

Host 1Having a no-training control is useful.

Host 2It is. Untrained people improve on retesting simply from getting familiar with the tests, so the control group shows how much change happens without any training at all.

Host 1What did they measure?

Host 2Before and after the six weeks: muscle size, strength, power, endurance, and serum hormone concentrations. The results mention upper-body and lower-body strength and endurance, jump power, and the hormones growth hormone, follistatin and cortisol.

Host 1Follistatin is not a household name.

Host 2Background: follistatin is a protein that binds and inhibits myostatin, which is a signal that limits muscle growth. So more follistatin is generally read as a pro-growth signal, though a resting blood level is a very indirect window into what is happening inside the muscle.

Host 1And what is not in the abstract?

Host 2The specific exercises, how the two programmes were matched for effort or volume, the duration of the holds, how muscle size was measured, and what time of day the blood was drawn. Those choices strongly shape results like these, and we cannot evaluate them from here.

Methods: the statistics

Host 1How did they analyse it?

Host 2Two layers. Within each group, they compared before and after using paired Wilcoxon signed-rank tests, which are non-parametric tests suited to small samples that may not be normally distributed.

Host 2Between groups, they used analysis of covariance, A-N-C-O-V-A, with each person's baseline value as a covariate. That adjusts for groups that start at slightly different levels. They then applied a Bonferroni correction for comparing three groups against each other.

Host 1And the effect sizes?

Host 2They report effect sizes as d, a standardized difference. As a rough guide, zero point two is called small, zero point five moderate, and zero point eight large.

A word from our sponsor

Host 2Do you have problems with thinning on top? Then you are in luck.

Host 2The Hastrup Review of Science is brought to you by hair-loss shampoo. Caffeine shampoo for men, used daily, as part of a routine for anyone who would like to keep what they have.

Host 1I use it after every session. I hold it on the scalp as long as I can, and then I let it go very slowly.

Host 2It comes in a range of formulas and bottle sizes, and it is very reasonably priced. It is widely available and in stock right now.

Host 1Six weeks in, I am not saying it grew back, but I am responding.

Host 2Hair-loss shampoo. Our referral link is in the show notes.

Results: both groups improved

Host 1Back to the trial. What happened?

Host 2Both training groups improved, and the changes were broadly similar between them. In the E-Q-I group, the within-group effect sizes ranged from zero point two three to three point zero three. In the dynamic group, from zero point two two to four point three three.

Host 1Those top numbers are very large.

Host 2They are, and some caution is warranted. Within-group effect sizes in small samples can be inflated when the spread of baseline scores is narrow, so a d of four is less dramatic than it sounds. Untrained people also respond quickly, especially in strength, much of which early on is the nervous system learning the movement.

Host 1What about the p-values within groups?

Host 2The abstract reports Wilcoxon p-values of zero point zero six nine or less for E-Q-I, and zero point one zero five or less for the dynamic group. Read literally, that means at least some outcomes in each training group did not reach the conventional zero point zero five threshold.

Host 1And the control group?

Host 2The control group's effect sizes ran from minus zero point seven nine to zero point one four, with Wilcoxon p-values of zero point one zero three or higher. So no meaningful improvement without training, which is what you would hope to see.

Results: head to head

Host 1Then the main question. E-Q-I against lifting.

Host 2No significant differences between the two training groups. The Bonferroni-corrected p-values ranged from zero point four zero zero to one point zero zero.

Host 2Both training groups generally improved more than control. The abstract says generally, so not every outcome separated clearly.

Host 1But there were differences in direction.

Host 2There were. E-Q-I tended toward small-to-moderately greater gains in upper-body strength and upper-body endurance, and larger increases in growth hormone and follistatin, with effect sizes of zero point three five to zero point seven zero.

Host 2Dynamic training went the other way for muscle size, lower-body strength and endurance, and jump power, and for cortisol, where the change was more favourable after dynamic training. Those effect sizes were zero point two four to zero point seven zero.

Host 1So each method had its corner.

Host 2In point estimates, yes, but none of those between-group differences were statistically significant. With ten people per group, they are best treated as directions for a future study rather than findings.

Host 1Is there a plausible reason lifting would favour jumping and leg strength?

Host 2This is speculation from background principles, not the paper. Jumping is fast and explosive, and dynamic lifting at least involves moving a load through a range with intent, while E-Q-I is slow by design. Specificity — you get better at what you practise — would predict that pattern.

Results: the hormones

Host 1Now the hormones. What should we make of them?

Host 2Cautiously. Background: large studies of resistance training have found that short-term changes in growth hormone and similar hormones are poor predictors of who builds more muscle.

Host 2A single resting blood sample before and after six weeks is also sensitive to sleep, time of day, recent meals and the last training session. So the hormone differences here are a weak signal, and at effect sizes below zero point seven with ten per group, an uncertain one.

Host 1So greater follistatin with E-Q-I does not mean more growth.

Host 2Correct, and the muscle size results lean slightly toward dynamic training anyway. If anything, the hormone and size data pointing in different directions is a reminder not to read hormones as a proxy for muscle.

Why it matters

Host 1Why does this matter?

Host 2The principle is that, for untrained young men over six weeks, a slow, low-load, hold-until-you-yield programme produced broadly comparable gains to conventional lifting. That fits a wider pattern in the background literature: early on, effort and time under meaningful tension seem to matter more than the specific way the load is applied.

Host 1What changes, and for whom?

Host 2For clinicians and coaches working with people who cannot load normally, this adds some controlled evidence that E-Q-I is a reasonable tool rather than an internet fad. For someone travelling with no equipment, it is an option with at least some supporting data.

Host 1And what does it not mean?

Host 2It does not show that E-Q-I is equivalent to lifting. Failing to find a difference in a small trial is not the same as showing there is none, and the study was not designed as an equivalence trial, as far as the abstract shows.

Host 2It also says nothing yet about trained lifters, women, older adults, injured patients — the group the method is pitched at — or athletes who need power. A sprinter, for example, would note that jump power and lower-body strength leaned toward conventional training.

Caveats

Host 1Let us go through the caveats properly.

Host 2First, sample size. Ten per group gives low statistical power. Background estimate: to reliably detect a moderate difference of about zero point five between two groups, you would typically need somewhere around sixty or more people per group.

Host 2Second, duration. Six weeks is short, especially for muscle size, and in untrained people much of the early strength gain is neural. Whether the two methods diverge after three or six months is unknown.

Host 1Third?

Host 2Population. Untrained men aged eighteen to twenty-six respond to almost any reasonable programme, which makes it easier for two different methods to look similar.

Host 2Fourth, the analysis. Within-group tests ask whether each group changed, but the comparison that matters is between groups, and that one found no significant differences. A mix of within-group significance and between-group non-significance is easy to over-read in either direction.

Host 1And the missing details.

Host 2Fifth, the programme matching. Without the full text we cannot say whether the groups were matched for effort, sets or time, and that determines what a comparison like this means. Sixth, the hormone measures carry all the noise of single resting samples.

Host 1Anything positive on rigour?

Host 2Yes. It is randomized, has a no-training control, adjusts for baseline values, corrects for multiple comparisons, and reports effect sizes alongside p-values. For a small training study, those are sensible choices.

What to watch

Host 1What should we watch for next?

Host 2Longer trials, ideally three months or more, with larger samples and a pre-planned equivalence or non-inferiority design. That is the design that could actually show E-Q-I is close enough to lifting.

Host 2Trials in the people it is meant for: patients in rehabilitation, older adults, and trained athletes. And imaging of muscle architecture, such as fibre-bundle length, since the long, lengthened holds are where E-Q-I might plausibly differ from conventional lifting.

Host 1And the upper-body hint?

Host 2Worth testing directly. If the small upper-body advantage holds up in a larger study, it would be a modest, specific result, not a general one.

Sign-off

Host 1That is the deep dive: Comparing six weeks of full-body eccentric quasi-isometric and dynamic resistance training, by Nemati, Romian, Khodaei, Zolfaghar Didani and Oranchuk, published in Scientific Reports. In short: both methods worked about equally well for six weeks in untrained young men, the differences between them were small and not significant, and the sample was too small to call them equivalent. The link is in the show notes.

Host 2Every Friday this same feed carries the roundup episode: five new papers as trailers, and you pick which become deep dives. This has been the Hastrup Review of Science.