Pacing from your own numbers

Almost every training-zone chart in existence is built on a maximum heart rate nobody measured, and almost every race prediction on a rule of thumb nobody checked. Both can be replaced with something derived from efforts you have actually done.

One race gives you the rest

Daniels and Gilbert's VDOT combines the oxygen cost of running at a given speed with the fraction of maximum you can sustain for a given duration. From one honest race it produces an equivalent time at every other distance and a full set of training paces. It is decades old, published rather than proprietary, and still the standard — a 20:00 5 km is a VDOT just under 50, and the table has said so since long before watches had opinions.

Two efforts give you the line

Critical speed needs two maximal efforts of different durations — say three to six minutes and twelve to twenty, on different days. The slope through them is the pace you can hold more or less indefinitely; the intercept, D′, is the fixed distance you can cover above it before you must slow down. D′ behaves like a battery: you can spend it quickly or slowly, but you cannot spend more than you have, and it does not refill mid-race.

Which is why going out too fast costs so much

Thirty seconds a kilometre too quick for the first third does not cost you thirty seconds. It empties D′ early, after which you are held below critical speed for the remainder and lose several times what the fast start gained. The model makes that concrete: it is the difference between a pacing plan and a hope.

Zones from something you can measure

The 220-minus-age rule is a line through a very scattered cloud — the standard deviation is about eleven beats a minute, so for a large minority it is out by more than ten. Every zone built on it inherits that error, and an easy run can quietly be a moderate one, which is the most common way endurance training goes wrong. Threshold heart rate is the average you can hold for about thirty minutes flat out. It takes one hard session to find, it rises as you get fitter, and zones anchored to it are the zones a coach means.

Load, and what the ratio does not tell you

Comparing this week's training load with the last month's is a reasonable way to notice that you have ramped up sharply. It is not a probability of injury: the original findings have been criticised over how the two windows share data, and later work has not consistently reproduced them. Treat a big jump as worth noticing and a number inside a band as no kind of permission.

Frequently asked questions

What counts as an honest effort?

One where you finished unable to have gone meaningfully faster, on a flat course, not into a headwind, and not after a hard week. A race is ideal. A time trial where you slowed in the middle and sprinted at the end will fit a line to something that is not your fitness.

Does this work for cycling?

Critical speed and the load ratio carry over directly — in cycling the same model is normally expressed as critical power from two maximal efforts. VDOT and the race-equivalence table are running-specific, because the oxygen cost of a pace is.

How often should I retest?

Every six to twelve weeks is plenty. Thresholds move slowly, and testing often enough to notice is testing often enough to interfere with the training that moves them.