5 things that help match a portafilter basket to your dose

The gap between your tamped puck and the shower screen tells you more about your basket's true capacity than the number etched into it ever will. A lot of baskets marked "18g" are cut generously enough to swallow 20g or more without complaint, because the same mould and tooling often gets reused across several machines and price points regardless of what's stamped on the metal. You can even feel the mismatch before you brew: if locking the portafilter in takes real force past where it normally clicks home, that's the puck fighting the shower screen for space, not the machine having an off day.

A domed puck, or a bright pressure-ring stamped into the top of the spent coffee, is overfill telling on itself before you've tasted a thing. When coffee sits proud of the basket rim, the shower screen can't seat flat against it, so it presses down harder on the centre than the edges and water bites into the middle first, channelling round it instead of through it. Check the group gasket too - stray grounds pressed into the rubber after you unlock the portafilter mean the puck had nowhere to go but up.

A shot that times normally - 25 to 30 seconds - but tastes hollow or sour is usually water finding a loose ring around the edge of an underfilled puck, not a grind problem. When there isn't enough coffee to fill the basket, the puck doesn't press fully against the wall all the way round, leaving a loosely packed moat that water rushes down in the first few seconds and dilutes the shot before the timer means anything. Knock the puck out and the same story shows up as texture: a puck that crumbles apart in the knock box didn't have enough mass to hold together under pressure in the first place.

A precision basket doesn't forgive a sloppy dose the way a stock basket does - it removes the slack, so buy one only once you already know your real number. Stock baskets are punched, not drilled, and their inconsistent hole sizes create their own turbulence that quietly smooths over a dose that's slightly off; a precision basket like VST or IMS laser-drills every hole to the same size and spacing, so that same half-gram error now shows up immediately as sour or bitter instead of getting masked. That's also why precision makers print a narrower working range than the vague nominal size stamped on a stock basket - buy to the dose you actually pull, not the label your old basket wore.

The same gram dose of a dark, oily roast and a dense light roast can fill a basket to two noticeably different heights, because roast level changes density far more than it changes weight. Beans lose mass as they roast, but their internal structure opens up and expands at the same time, so a dark, well-developed roast is bulkier per gram than a lighter, denser one even at an identical number on the scale. That's why a recipe that flowed beautifully on last week's bag can choke or run thin on this week's without you touching your technique - the bean changed the actual volume sitting in the basket, not you.