Does elevation gain equal elevation loss on a loop hike?

The short answer

Elevation gain and elevation loss are both running totals built the same way: walk from one recorded point to the next, compare the two altitude readings, and add the difference to gain if the second is higher or to loss if it is lower. Add up every point in the loop this way and, because you end at the same elevation you started from, the two totals should in theory cancel out to the same number. They come close in practice, but a small gap of a few percent between gain and loss is completely normal, and it comes from the same source as most of the small oddities in a recorded track: altitude is measured with some noise, one reading at a time, rather than looked up from a fixed height map of the trail.

What should be true, in theory

A loop is a closed shape. If you leave a trailhead at some elevation and eventually walk back to that same trailhead, your net change in height over the whole route is zero — every climb has a matching descent somewhere else on the loop, because you cannot end up back where you started any other way. Add up every uphill metre and every downhill metre across the entire route and, mathematically, they have to balance. That is the reasoning that makes gain and loss on a loop feel like they ought to match exactly, and it is correct as a statement about the real hillside you walked.

Why a loop is actually a weaker test of this than it looks

An out-and-back hike gives its own gain and loss two full legs to agree with each other, because the same ground gets measured twice — once walking out, once walking back over it. A loop never repeats any ground. It only touches the question of whether gain equals loss at a single moment: the instant the very last recorded point compares itself back against the very first one. Everything in between is free to wander, because the running total does not check itself against the starting elevation until the walk is already over. A loop's gain-versus-loss match rests on far less evidence than an out-and-back's does — it is really a comparison between two points, not two legs.

The start and end readings are two separate measurements

Even standing on the exact same patch of ground, an altitude reading taken at the start of a three-hour loop and one taken at the end are two independent measurements, not two copies of the same number. GPS altitude carries a few metres of ordinary noise on any given reading, and that noise does not know or care that you happen to be standing where you began.

Barometric drift moves the goalposts over time

iPhones from the 6 onward read relative altitude change from a barometric pressure sensor, which is smoother than GPS altitude alone but is calibrated against the current air pressure — and air pressure drifts with the weather over the course of an afternoon, independently of how high you actually are. A trailhead that reads at one altitude when you set out can read a few metres higher or lower by the time you get back, purely because the air pressure shifted while you were out on the trail, with nothing to do with the ground itself.

The noise filter runs point by point, not leg by leg

Before small jitters get added to either total, an app filters out up-and-down noise below some threshold so a flat stretch does not quietly rack up gain or loss on its own. That filter is applied to each comparison as the track is walked, with no memory of the running gain-versus-loss balance, so it has no way to trim gain and loss back toward matching each other at the end.

What a small gap does and doesn't mean

A gap of a few percent between total gain and total loss on a loop is the ordinary result of measuring a real, noisy quantity along a route, not a sign the app miscounted or that you veered off the trail. It is small precisely because the two totals really do describe the same closed shape — the physics is on their side — even though nothing in the measurement forces them to land on an identical figure. A much larger gap is a different story, and points more often to a longer detour partway round, a stretch under heavy tree cover that degraded one section's readings, or a pause that quietly dropped part of the climb, than to elevation gain being unreliable in general.

Things that trip people up

• Expecting a loop to close its gain-versus-loss balance the way an out-and-back closes its outbound-versus-return one — a loop gets one comparison at the very end, not two full legs of overlap, so a bit more disagreement is normal.
• Reading any mismatch as proof the trailhead elevation was recorded wrong — the two readings are separate measurements taken hours apart, and a small difference between them is ordinary sensor noise, not an error at either end.
• Assuming a longer loop should produce a wider gap than a short one — the size of the gap depends on conditions along the way and how much the weather shifted, not on how much distance or time the loop covered.
• Adding gain and loss together to sanity-check "how much climbing" a loop involved — the two numbers already describe the same climbing from two directions; neither one is the odd one out to be corrected against the other.

Where TrailCam fits in

TrailCam reports elevation gain alongside distance, moving time and pace for every recorded route, built from the altitude readings as they were actually logged rather than adjusted to force a loop's gain and loss into agreement. Replaying a saved route shows the same elevation profile underneath the map, so a small gap between the two totals on a loop is something you can look at directly. The app is free to download.