Every method below has a real argument behind it and a real limit on it. Both are on the page.
Training methodology has a credibility problem, and it is not that the ideas are wrong. It is that they arrive stripped of the thing that made them worth believing. A finding from forty-five untrained women on a leg extension machine becomes "the research says" inside two reposts, and by the fourth it is a rule about how you should bench.
So each page here does the same three things. It says what the method claims. It names the study the claim rests on, with a link. And it states the population that study used, because a result from untrained beginners on one machine is a different kind of evidence from a meta-analysis across fifteen trials, and you deserve to know which one you are being sold.
None of these methods is a religion, and you do not have to pick one. Most well-built blocks borrow from several: volume landmarks to decide how much, proximity to failure to decide how hard, long-muscle-length work to decide how a set ends, polarized distribution to decide what the easy days look like. What matters is knowing which decision each one is actually helping you make.
One more thing the table below makes obvious, and it is the reason this cluster exists at all. The strength of the evidence and the popularity of the method are close to unrelated. The best-supported idea here is the dullest one, and the idea with the most reposts is standing on forty-five beginners and a leg extension machine.
"Research-backed" covers everything from a meta-analysis across fifteen trials to one machine and forty-five beginners. Both are real. They do not deserve the same confidence, and no page that lumps them together is telling you the truth.
| Method | The decision it helps you make | What it rests on | Who was studied |
|---|---|---|---|
| Lengthened partials | How a set should end | One randomised ROM trial, plus an 8-week pilot a decade earlier | 45 untrained women, knee extension only |
| Volume landmarks | How much to do in a week | A meta-regression over 34 treatment groups from 15 studies | Broad, but it gives no landmark numbers |
| Hybrid training | Whether the two goals fit together | A meta-analysis of 21 studies and 422 effect sizes | Broad; the figures are from 2012 |
| Polarized endurance | What the easy days should look like | A randomised crossover where each rider was his own control | 12 trained male cyclists |
| Training to failure | How close to take a set to the limit | Four meta-analyses and meta-regressions, 2021 to 2024 | Broad; results shift with matched volume, and highly trained lifters are under-studied |
The strongest evidence on this page supports the least fashionable method, and the most fashionable method has the narrowest base. That is worth sitting with before rebuilding a training year around a graph you saw reposted.
Fire Your Coach stores the intensifier, the volume and the intensity as structured data rather than as notes, which is what lets a plan still be recognisable in week four.