Why the Four-to-One Rule Exists, and What It Protects Against

Why this matters

The 4:1 setting angle is not a convention somebody picked for tidiness. It sits between two opposite failure modes, and it is the angle at which the demand each one places on the setup is small at the same time. Set it shallower and the base wants to slide out from under the climber. Set it steeper and the ladder wants to rotate backward off the wall. That is why the rule has no safe side to err toward, and why the useful thing to know is not the ratio but what the ratio is doing to the friction your ladder feet have to produce.

What the rule says, and what it is measured on

29 CFR 1926.1053(b)(5)(i) requires a non-self-supporting ladder to be used at an angle such that the horizontal distance from the top support to the foot is approximately one-quarter of the working length, which the same paragraph defines as the distance along the ladder between the foot and the top support. The general-industry counterpart sits in 29 CFR 1910.23. A base offset of one quarter of the working length puts the ladder at about 75.5 degrees from horizontal.

The words that get lost are working length and top support.

        rails extend above the landing; this
        part is NOT in the working length
                   ___
                  |   |
   ---------------|---|------- upper landing and
                 /|              top support
                / |
   working     /  |
   length     /   |  vertical rise
             /    |
            /     |
           /______|
        foot
        |<------->|
         base offset = working length / 4

The working length runs from the foot to where the ladder bears, not to the tip. When a ladder is used for access to an upper landing, its rails have to extend at least 3 ft above that landing, and none of that 3 ft is in the working length. Neither is the extra length of a ladder that reaches past the point it touches. Getting this wrong is the single most common way a ladder ends up set shallow while the person who set it believes they followed the rule.

This is also a different four from the one in 29 CFR 1926.1053(a)(1), which requires a ladder to support at least four times its maximum intended load. That is a strength factor covered in the duty-rating card. The two fours have nothing to do with each other, and a ladder set at 4:1 does not thereby have a four-times margin in its setting.

The two failure modes it sits between

Too shallow: the base slides out. With the top resting against a wall, the wall pushes the ladder horizontally away from itself. The only thing resisting that push at the bottom is friction between the ladder feet and the ground. Lower the angle and the horizontal push grows while the vertical load on the feet stays the same, so the friction the feet must produce goes up. When the ground cannot produce it, the base goes out and the climber comes down with the ladder.

Too steep: it rotates backward. As the angle approaches vertical, the horizontal push at the top falls toward zero, which means the wall is barely holding the ladder against anything. Any backward force at all, a climber reaching, a door opening into the ladder, wind on a sheet of material, has almost nothing resisting it, and the ladder rotates about its feet away from the wall. There is no clean published number for where this becomes governing, which is exactly why the standard says "approximately" rather than naming a tolerance band.

Because the two modes run in opposite directions, there is no conservative direction to round in. Round the offset up and you buy tip-back margin by spending slide-out margin. The conservative act is hitting the target, not biasing it.

The friction demand, and what it holds constant

Take moments about the foot of the ladder. Let W be the climber and everything they carry, w the ladder's own weight acting at its mid-length, L the working length, d the climber's distance up the ladder from the foot, and theta the angle from horizontal. Treat the wall as frictionless, which is the conservative assumption because a real wall contributes some friction and reduces the demand.

The horizontal reaction at the top, which is also the friction force the feet must produce:

N = [ W x (d/L) + w/2 ] / tan(theta)

And the coefficient of friction the feet and the ground must produce between them:

mu = N / (W + w)

Both expressions hold constant: a rigid ladder, a frictionless wall, a static climber, no side load and no wind, and the ladder's weight distributed evenly so its centre sits at mid-length. Change any of those and the numbers move.

Run it for a climber and load of 250 lb on a 40 lb ladder, with the climber three quarters of the way up, which is the position held constant across every line below:

Going from 4:1 to 3:1 raises the friction demand by about 37%. Going to 2:1 raises it by a factor of 2.2. Those are demands, not capacities.

What is available is not a ladder property. The coefficient a ladder foot can develop is a property of the shoe and the surface together, and the ladder manufacturer publishes performance for their shoe on defined surfaces. Nobody can hand you one number that covers a rubber pad on dry swept concrete, the same pad on concrete with a film of dust, and the same pad on a frosted membrane or a dropcloth. What the table above tells you is how quickly a shallow set consumes whatever you have, on a surface you did not test.

What the number does as you climb

Hold the setting at 4:1 and move the climber from three quarters of the way up to the top rung. The numerator gains the last quarter of the climber's weight:

Three quarters up:  250 x 0.75 + 20 = 207.5 -> mu 0.185, push 54 lb
At the top:         250 x 1.00 + 20 = 270.0 -> mu 0.241, push 70 lb

The friction demand rises by about 30% over the last quarter of the climb, at the same angle, on the same ladder. That is worth knowing on its own, because it means a ladder that felt solid at the working height is at its highest demand at the moment the climber transitions to the landing, which is also the moment they are least able to react.

Worked example: one 20 ft eave, set three ways

Rise from the ground to the eave where the ladder will bear: 20 ft. Same climber and load as above, same three-quarters position for every comparison so the readings are like for like.

Setting it correctly. The base offset for a rise h at a true 4:1 comes out of the geometry. With base b and working length L, b equals L over 4 and L squared equals h squared plus b squared, so:

L^2 = h^2 + (L/4)^2
L^2 x (15/16) = h^2
L = h x 4 / sqrt(15)
b = L / 4 = h / sqrt(15) = h / 3.873
b = 20 / 3.873 = 5.16 ft

Set the base 5 ft 2 in out. Working length is 20.66 ft. Friction demand 0.185, horizontal push 54 lb.

The harmless shortcut. Setting one foot out for every four feet of rise gives 5.00 ft rather than 5.16 ft, about 2 in steeper. That lands at 1:4.12 and drops the friction demand slightly. It is inside "approximately" and it is not worth correcting on a roof.

The shortcut that hurts. Setting one quarter of the ladder's extended length instead of the working length. Note first that a ladder's labelled size is not its extended length, because the sections must overlap by the amount the manufacturer specifies, and the extended length is on the maker's label. Say this one measures 25 ft extended.

Base offset used: 25 / 4                         6.25 ft
Distance along the ladder to the eave:
  sqrt(20^2 + 6.25^2)                           20.95 ft
Actual setting: 6.25 / 20.95                     1 : 3.35
Angle                                            72.6 deg
Friction demand, same climber at three
  quarters, same formula                         0.224
Horizontal push at the top                       65 lb

Compared with a correct 4:1, that is 21% more friction demanded of the feet and 11 lb more push at the top, from a person who measured carefully and used the wrong length. The ladder looks fine. Nothing about it looks shallow to the eye at 72.6 degrees against 75.5. The only visible difference is 13 in of base position.

The failure mode. It does not fail on the way up, when the demand is 0.224 and the concrete is dry. It fails on the way down, after two hours, when the tech has tracked roof gravel onto the pad under one foot and the available coefficient has quietly moved while the demand did not. A 21% overdraft on a resource you never measured is spent long before you notice it is gone.

What changes the answer

If the top cannot resist sideways, the angle is not the governing question. A ladder bearing on a gutter, a smooth column, a corner or a single point can move laterally, and the plane geometry above says nothing about that mode. The answer there is securing, covered in the sibling how-to.

If the feet are not both bearing flat, none of these numbers apply. The friction demand is computed for a foot pad in full contact. A shoe on one corner, on gravel, on a slope, or propped on a board has an available coefficient nobody can state, and the correct response is to fix the base rather than to reason about the angle.

A self-supporting ladder is a different structure. A stepladder stands on its own spreaders and the 4:1 rule does not apply to it at all.

How to verify you got this right

Every check below happens from the ground, before anyone climbs, because a check performed from the ladder is a check performed after the risk was taken.

The tape check is the accurate one. Measure the rise to the point where the ladder bears, divide by 3.873, and set the base at that. Dividing by 4 instead is 3% steep at any rise and does not matter; dividing the ladder's length instead of the rise does matter, by the margin computed above.

The arms check is the fast one. Stand with your toes against the ladder feet and your arms straight out at shoulder height. At roughly 4:1 your palms meet a rung. It is a rough check and it depends on your own arm length, so calibrate it once against a tape rather than trusting it blind.

The built-in indicator on the rail is the maker's, and it references the rail rather than the ground, so it reads true only when both feet are bearing flat. On uneven ground it will report a correct angle for a ladder that is not correctly set.

Then push the ladder horizontally at chest height from the ground and watch the feet. Any slip at all, at a force nowhere near the 54 lb the top will actually see, means the surface will not deliver, and the answer is a different surface or a secured base, not a more careful climb.

References

• 29 CFR 1926.1053(b)(5)(i), which sets the horizontal distance from the top support to the foot at approximately one quarter of the working length, and defines working length as the distance along the ladder between the foot and the top support.
• 29 CFR 1926.1053(a)(1) for the separate four-times-maximum-intended-load strength requirement, and 29 CFR 1910.23 for the general-industry portable ladder requirements.
• Manufacturer instructions for ladder shoe performance on defined surfaces; the available coefficient is a property of the shoe and the surface together and comes from them, not from this article.
• See related: universal-what-a-ladder-duty-rating-actually-means and universal-how-to-set-and-tie-off-an-extension-ladder-so-it-cannot-move.