You Ran the Bucket Test. Now What? Reading Pool Loss Like a Pro

Somewhere on a pool forum right now, a homeowner is posting the same story: chemicals drifting strangely, the autofill running more than it should, a bucket test showing the pool dropping past the bucket by a quarter inch a day, and a pool builder shrugging that the difference is "normal." One widely read thread on Trouble Free Pool documented exactly that: a week-long test where the bucket dropped less than half an inch while the pool lost more than two and a half, followed by a dive inspection that found nothing and a homeowner asking the internet what to do next. If that is you, this post is the answer. The bucket test told you that you have a leak. Two more observations tell you roughly where it is, before any professional equipment arrives.

First, make sure the test itself was clean

The bucket test works by comparison: a bucket on the pool step, filled to match the pool's level, loses water only to evaporation, while the pool loses water to evaporation plus any leak. For the comparison to hold, the conditions must match. The bucket sits in the water so both surfaces share the same sun and air. No swimming, no backwash, no rain, no autofill during the window. Twenty-four hours minimum; forty-eight gives a clearer signal on slow leaks. In Compton's marine-moderated climate, summer evaporation runs roughly a quarter inch a day, less under humidity or a cover, so a pool outpacing its bucket by a quarter inch or more per day has confessed.

The two observations that localize the leak

Observation one: pump on versus pump off. Run the test twice, once with the pump on its normal schedule, once with it off. The comparison sorts the plumbing from the structure:

Observation two: the stabilization level. If the loss looks structural, let the level fall unattended, with the pump off and equipment protected, and note where it stops. A leak stops leaking once the water drops below it, so the stabilization point is the leak's elevation, testifying for free. Water stopping at the skimmer's bottom lip convicts the skimmer throat. Stopping at the return fittings names them. A level that keeps falling toward the main drain says the failure lives deep, in the drain, the floor plumbing, or low shell.

What the professional visit adds

Your observations shrink the search; instruments finish it. Each plumbing circuit gets isolated and pressure-tested to prove which line, if any, will not hold. Dye tracing at the usual structural suspects, skimmer throat, returns, light niche, visible cracks, shows exactly where water is being pulled through the shell. Buried return and suction runs that fail their pressure tests get traced to the break point with listening gear or tracer gas rising through soil and deck, so the surface opens at one marked spot rather than along the whole run. The methodology scales to the deep end too; how drains, floor returns, and shell questions get handled at depth is covered on our inground pool page, including why a pool on this water table should never be casually drained.

Compton context: why local pools leak the way they do

The backyard pools across this city mostly date to the same postwar decades as their houses, gunite shells that have spent sixty years in soft alluvial soil that never fully holds still. That age and ground movement concentrate failures at the penetrations, skimmer-to-shell joints above all, and in the buried plumbing loops of the era. The flat low ground adds a diagnostic gift and a cost: escaped pool water does not run off, it soaks in beside the shell, so a persistently damp strip between pool and equipment pad is real evidence, and the same saturation quietly undermines decks when leaks run long. Pool-dense neighborhoods like South Compton generate a steady spring caseload of exactly these patterns, and the households that arrive with bucket numbers and a stabilization level get answers measurably faster.

The equipment pad, before anyone gets wet

One more stop costs nothing: inspect the equipment pad with the pump running and again at rest. Pump shaft seals weep only under operation; filter tank o-rings mist so finely the loss reads as mystery; a multiport valve can pass water silently to waste for a season. Pad failures are the cheapest finds in pool work and they mimic everything else, so a dry-towel inspection there belongs in every homeowner's follow-up to a positive bucket test.

Your next move, in one paragraph

Write down four numbers: the pool's daily loss, the bucket's daily loss, whether the pump changed the rate, and where the level stabilized if you let it fall. Photograph any damp deck or lawn. Then make the call with that file in hand. On the forums, the thread you started ends with strangers guessing. With the numbers, a leak visit ends with a mark, a quote, and a pool that holds its water again. The metered water you stop pouring into the ground pays for the diagnosis faster than most owners expect.

Bucket-test mistakes that produce false verdicts

The test is simple, which is exactly why its failure modes are worth naming. The most common false positive is the autofill left active: a fill valve topping the pool overnight erases the loss the test exists to measure, or, misreading in the other direction, a float set low lets the pool drop to its setpoint and no further, mimicking stabilization. Disable autofill for the full window, physically if the controller cannot be trusted. The second is splash and use: one enthusiastic swim session moves more water than a day of modest leaking, and a dog with pool privileges is a walking measurement error. The third is weather: rain obviously voids the window, but wind matters more than most owners expect, since a dry Santa Ana day can double evaporation and make an innocent pool look guilty. Run the test in settled weather or extend it to forty-eight hours so the signal outweighs the noise.

Measurement technique adds its own traps. Mark levels with tape at the waterline rather than trusting memory or photographs taken at different angles; a quarter inch is the whole verdict, and parallax eats quarter inches. Keep the bucket weighted so it neither floats nor tips, and set it where pets and sprinklers cannot amend the data. If the first result lands ambiguous, within an eighth of an inch either way, resist the urge to declare victory or defeat; repeat the test rather than interpreting noise.

Finally, the false negative worth knowing: a pool that only leaks under pump pressure can pass a pump-off bucket test entirely, because the failing return line never sees pressure during the window. That is precisely why the two-run protocol exists, and why a pool with symptoms, chemistry drift, autofill working overtime, a damp pad, deserves both runs before anyone concludes the bucket cleared it. The test is a fine instrument. Like every instrument, it only answers the question it was actually asked.