Deck railing in Idaho

A guard is not trim. It is a structural assembly that has to stay put when somebody stumbles into it, and the hardware behind it costs more than the part you can see. That is why rail is routinely a fifth to a third of a deck budget, and why it is the line that most often separates two quotes on the same yard.

A guard is a lever

That multiplication is the entire engineering problem, and it is invisible in a photograph. Two rails that look identical can have completely different connections underneath, and the one you can feel the difference in is the one you lean on.

It also explains something people find counterintuitive on a quote. The infill, the part you chose, is often the cheaper half. The posts, the ties, the blocking and the fixings are the half that decides whether the assembly does anything.

Whether you need one at all

A guard becomes necessary once the walking surface is far enough above the ground that leaving the edge open would be a fall rather than a step. Below that, it is a design choice.

On flat ground this is the most useful thing to know at the sketch stage, because it is often available. A deck set a step or two down from the back door in a level Meridian yard can frequently stay beneath the line and do without a rail entirely. That removes a large cost, and it leaves an open edge, which is a different and usually better relationship between the deck and the lawn. The layout guide covers how the door height drives it.

Infill: the gap is the specification

Whatever fills the space between the posts, it is spaced closely enough that a small child cannot get through it. That single requirement drives baluster count, cable count, and every gap at the end of a run where the infill meets a post.

Cable is the system where this is hardest to hold. Cables are flexible by nature, so they need enough tension and enough intermediate support that a push does not spread them apart, and both of those are things a cheap installation economizes on. A cable rail that has gone slack within its first season did not have enough of either.

Handrails, and what they are for

A guard stops you leaving the deck. A handrail is a different object with a different job: it is something to hold while going down, which is a thing you reach for without looking, usually in the dark, sometimes while carrying something.

That job sets its shape. It has to be graspable, which a wide flat board on top of a post run is not. It runs continuously the length of the flight, because a handrail that stops partway is worse than none at the moment your hand finds air. And it returns into the guard or the wall at each end rather than terminating in open space, so a sleeve or a bag strap cannot hook on it.

Stair stringers bear on their own footings, on solid ground, exactly as the deck posts do. What they must not do is land on an existing patio slab, and the footings guide covers why that particular shortcut ends in a racked stair.

Choosing a rail system

Once the structure is settled, the rest is preference and money. Installed rates in the Treasure Valley, per linear foot:

• Wood, $45 to $80. Cheapest, warmest looking, and the only one that needs re-coating on the same cycle as a wood deck. Cedar holds up better than treated pine on a rail because it moves less.
• Composite, $75 to $130. Matches composite decking, needs no finishing, visually the heaviest of the systems because the components are chunky.
• Aluminum picket, $80 to $140. Slim profile, powder coated, essentially maintenance free. The most common upgrade on a composite deck.
• Stainless cable, $110 to $190. The clearest view. Needs terminal posts anchored for cable tension, which is a real structural difference rather than a hardware swap.
• Glass panel, $180 to $300. Fully transparent, blocks wind, and shows every water spot. In a valley with hard water that is a maintenance consideration people underestimate.

Rail is priced on the open perimeter, so the number that decides the cost on your own deck is linear feet of open edge rather than square footage. A wide shallow deck against the back of a house rails cheaply. A deck that stands clear on three sides does not.

Retrofitting rail onto an existing deck

The limiting factor is almost always the posts, not the infill. Many decks built in the 2000s Meridian subdivision wave have rail posts lag screwed to the outside face of the rim joist, which was normal practice at the time. Go back to the lever: that connection resists the pull at the base of the post only through the withdrawal strength of a few screws, and withdrawal is the weakest thing you can ask of a fastener in wood.

Swapping the infill on those posts therefore buys appearance and nothing else. A retrofit that changes anything replaces the posts and their anchoring first, then fits the new system. The infill is the visible cost and the post work is the one that disappears out of a cheap quote. The repair page covers what else is worth looking at while the rail is off.