Resistance Bands for Home Workouts: Beginner Guide

“What weight is this band?” has no answer, and that is a property of the equipment rather than a gap in the packaging. A band is a spring. It gives almost nothing near its resting length and a lot near full stretch, so any single number printed on it describes one moment in a rep and says nothing about the rest.

Which leaves a practical question the colour system does not answer: at the point in the movement where it feels hard, how many kilograms is the band giving you? Two groups have measured that, and the useful part is that they do not agree.

How far you stretch it sets the load

Load rises with elongation, measured as a percentage of the band’s resting length. Stretch a band to double its length and that is 100 percent elongation.

This is why the colour is a slippery unit. Two people using the same green band get different loads because one is a foot further into the stretch. The same person gets different loads from the same band on a row and on an overhead press. And a band anchored short gives more at the same range of motion than the same band anchored long, because the same movement is a larger percentage of a smaller resting length.

So the colour is a step size rather than a weight, and to use it you need a starting point in kilograms. Two papers provide one.

Two measurements, and they differ by up to 65 percent

Uchida and colleagues measured Thera-Band tension directly, on eight half-metre bands, at ten elongations from 25 to 250 percent, with two observers taking five readings each. Their figures at 100 percent elongation, in kilograms-force:

Note the second column against the first. Doubling the stretch again, from 100 to 200 percent, adds about 40 to 60 percent, not another 100. A band you have already pulled to double length is giving most of what it has.

The AGUEDA equations, published in 2024, take a different approach and give you something you can compute with. Load in kilograms equals a per-colour constant times the percentage of elongation, and the fit is tight, with R-squared between 0.977 and 0.995:

Set the two right-hand columns beside each other and they do not match. Green is 1.82 in one and 3.0 in the other. Blue is 2.51 against 3.5, black 3.28 against 4.2, gold 5.66 against 6.9. AGUEDA runs between 22 and 65 percent higher across the range, on the same nominal product.

Both are peer-reviewed and both state their method, so this is not a case of one being wrong. The most obvious difference is band length: Uchida measured 0.5 m bands and AGUEDA measured 200 cm ones. That would fit the physics, since percentage elongation on a short band and a long band are not the same absolute stretch. Neither paper claims to settle the other, and I am not going to pick one for you.

What follows from the disagreement is the actual advice on this page. A chart gets you an order of magnitude and not a number. Use either table to know that green is roughly two to three kilograms at double length rather than roughly ten, then set the load by how the last two reps feel, because that is the only measurement taken on your band at your anchor length.

The manufacturer’s ladder does not match either

TheraBand publishes its own progression: “a 25% increase in pull force in colors Yellow through Black and a 40% increase in pull force in colors Silver and Gold.” That reads as a specification and it is a design intent.

Work the steps out of the measurements instead. From Uchida’s 100 percent column, yellow to red is +37 percent, red to green +14, green to blue +38, blue to black +31. From the AGUEDA constants, the same four steps are +24, +43, +17 and +20.

Neither sequence is a flat 25 percent, and the irregularity has a practical cost. Moving up a colour is sometimes a small step and sometimes a large one, and which is which depends on where you are in the ladder and whose data you use. If a jump feels impossible, that is worth knowing before you conclude something about yourself.

Loops, tubes, therapy bands, figure-8s and hip loops are not substitutes

The bands themselves come in five forms that are not substitutes for each other.

• Loop bands, the long continuous ones with no handles. Lower body, pull-up assistance, accommodating resistance on a barbell.
• Tube bands with handles. The most general-purpose shape. With an anchor they cover rows, presses, pulldowns and curls, and many stack.
• Flat therapy bands, the ones the two papers above measured. Thin, light, cut to length, made for rehabilitation and shoulder work.
• Figure-8 bands. Chest and rear-delt work, and not much beyond it.
• Fabric hip loops. Glute work and warm-ups, and they do not slide the way latex does.

The common mistake is buying one kit and expecting it to cover everything. For most people in a flat, one tube set plus either a hip-loop three-pack or a flat therapy set, chosen by what you are training, is the whole kit.

How to progress, without adding a pound

You cannot add 2.5 kg to a band, so progression works differently, and there are two levers.

Step up a colour, accepting that the step is somewhere between about 15 and 45 percent depending on where you are in the ladder. Or keep the colour and shorten the working length, by choking up on the band or moving closer to the anchor, which raises the percentage elongation at the same range of motion and therefore raises the load. The second lever is finer, free, and the one I would use first when a colour step feels like too much.

Because the load sits at the top of the stretch and nearly disappears at the bottom, bands suit higher repetition ranges than iron does. That is a consequence of the curve rather than a rule anyone has tested, so treat it as a starting point and let the last two reps of a set tell you the rest.

Bands win four jobs and lose three

They win on joint-friendly accessory work, where the load is lightest at the bottom of the range; on travel, where a full kit packs into a jacket pocket; on micro-sessions, where setup is close to zero; and on pull-up assistance, where the curve offloads exactly the hardest part of the movement. They lose on maximal low-rep strength, on heavy compound lifts above bodyweight, and on any movement whose safe stretch is shorter than its range of motion. If the choice is between bands and a first set of weights, that decision is worked through in adjustable dumbbells vs regular, and the wider kit in the small home gym setup guide.

Three facts from the safety page that bear on load

Nothing on this page is a safety instruction, and the safety material for bands is a separate subject with its own sources. It is all in anchoring resistance bands in a rental, which carries the anchor manufacturers’ own published conditions of use, the instruction to inspect a band before every session and discard it on any damage, the published limit on how far a band may be stretched, the manufacturers’ recommendation of eye protection, and both published papers on injuries, including what a case series can and cannot establish about how likely any of it is.

Three things from there bear directly on load, so they belong here too.

The stretch limit is published and it bounds every elongation figure on this page. HCI Fitness states it plainly: “do not stretch product more than 2 times its resting length. This includes each loop individually and/or combined.” That is 100 percent elongation, which is the left-hand column of both tables above, so the 200 percent column describes measurement conditions rather than a target to train at.

A replacement interval, second: the one manufacturers publish is scoped to heavy clinical use and is not a general figure for a band used twice a week.

And third, there are two documented failure modes, the band leaving its anchor and the band itself letting go under tension, so inspecting the band matters as much as inspecting what it is hooked to.

If you have an injury or a condition, or any reason a recoil would matter more in your case, that belongs with a clinician rather than with a chart.