Theory of constraints (training the limiter)
Evidence: limited
Borrowed from systems management: a system’s output is capped by its single biggest constraint, so improving anything else changes nothing until that constraint is addressed. Applied to training, the fastest gains come from finding your actual limiter, the determinant holding your racing back, and training it, rather than polishing a strength you already have. The framework is sound and useful; the hard part, and where the evidence thins, is correctly identifying the limiter.
The theory of constraints is a management idea from Eliyahu Goldratt: a system’s total output is set by its single biggest bottleneck, and improving any other part does nothing for the whole until that bottleneck is fixed (Goldratt 1984). A chain is only as strong as its weakest link. It transfers cleanly to endurance training, where it is the antidote to the natural habit of training what you are already good at and enjoy.
The limiter, not the strength
Running performance rests on several determinants: VO₂max, the fraction of it you can sustain, running economy, durability, plus the supporting cast of strength, fuelling, gut tolerance, pacing, the mind, and the plain ability to stay uninjured (Joyner & Coyle 2008). For any given runner at any given time, one of these is the binding constraint, the thing that gives way first in a race. The theory of constraints says the highest-return training is whatever raises that, and that work spent on a non-limiting quality is largely wasted until the limiter moves.
This is why two runners with the same race time can need opposite training. One has a huge engine but folds late in long races: their constraint is durability and strength, not VO₂max, so more intervals do little. Another has ground out a big aerobic base but has no gear change: their constraint is speed and economy, and yet more easy volume is the wrong lever. The plan follows the limiter, which is why it is individual and why copying an elite’s programme usually misfires.
The loop
Goldratt’s method is a repeating cycle, and it maps onto a training block: identify the constraint, get the most out of it, organise the rest of training around it, raise it, then re-test and move to whatever is now the weakest link (Goldratt 1984). The “re-test and move on” step matters: fix one limiter and a different one becomes binding, so the target rotates over a career. This is the same logic that underpins sound periodisation, developing qualities in sequence rather than chasing all of them at once.
Where it gets hard, and where it misleads
The framework is only as good as the diagnosis, and that is the catch.
- The limiter is hard to identify. It takes honest testing and pattern-reading (where do races fall apart, and why?), and runners reliably misdiagnose, naming the quality they enjoy training rather than the one holding them back.
- Running is not a single clean bottleneck. Unlike a factory line, the determinants interact, and more than one can be limiting at once, so “train only the constraint” is a guide, not a literal rule.
- For most runners the constraint is the basics. Before any of this gets subtle, the binding constraint for the great majority is simply doing enough consistent training and staying healthy. Hunting for an exotic limiter while under-trained or frequently injured is solving the wrong problem.
Used with those caveats, it is one of the more useful mental models in training: spend your limited training stress where it actually changes the result.