How to Pick Your Next Weight

Every lifter meets the same small decision, over and over: you did your sets last time, you're back under the bar, and something has to go on it. Add weight? Keep it? How much? Most people either guess, or add a plate because the calendar said so — and then wonder why the lift stalls or the joints complain. There's a better answer, and it isn't a fixed schedule. The next weight is a read on the last set, not a number you decided a month ago.

The one question that decides it

Before you touch a plate, answer this: how did the last set actually go? Not "did I finish it" — how. Three pieces:

• The reps — did you hit the target on every work set, or fade on the last one?
• The effort — how many clean reps did you have left? (That's RPE / reps in reserve — see how to rate a set honestly.)
• The bar speed — did it move, or did it grind?

Those three tell you almost everything. A set you finished at target reps with two in the tank and good speed is a green light. The same reps, same weight, but the last one ground to a crawl at zero in reserve — that's the same number on paper and a completely different signal. Progression is just reading that signal correctly.

Three ways lifters answer it

1. Linear — add a fixed amount

The beginner's gift: add a small, fixed jump every session (say 2.5 kg on the big lifts) and ride it as long as it works. When you're new, your body adapts fast enough that this just keeps paying off for months. Its whole virtue is that it needs no judgment — which is also why it eventually breaks. When the fixed jump starts costing you reps two sessions out of three, you've outgrown it.

2. Double progression — the workhorse

The method most intermediate lifters should live on. Pick a rep range instead of a single number — say 6–8. Keep the weight and add reps session to session until you hit the top of the range on every work set. Only then do you add weight, and drop back to the bottom of the range. Three sets of 6 → … → three sets of 8 → add weight → three sets of 6 again.

The discipline is in the words every work set. If your target is 3×8 and you get 8, 8, 7, you have not earned the increase — you repeat and try to turn that 7 into an 8 first. Double progression works because it makes you prove you own a weight before it lets you move on. Most reputable strength writing lands here; it's the core of, for instance, Stronger by Science's take on progression.

3. Autoregulation — adjust load to a target effort

The most responsive method, and the one that handles bad days gracefully. Instead of a fixed weight, your plan targets an effort: "top set at RPE 8." Then you let the day set the load. If the prescribed weight comes up at an RPE 7, add a little; if it's already a 9, you've hit your 8 early — stop there. A useful rule of thumb from Reactive Training Systems is that roughly 2% of load moves the RPE about half a point — so being one full point off target is a ~4% adjustment, not a wild guess. The deep dive on effort lives in MASS's guide to RPE and RIR.

These aren't rival religions. Most good training is double progression inside an effort cap: work your rep range, but never grind past the RPE you set.

A rule you can run in ten seconds

At the end of a lift, look at the last session and pick a lane:

• All sets at target reps, effort at or below target, bar moved well → add the smallest useful jump. You earned it.
• Hit the reps, but it ground (effort above target) → repeat the weight. The goal next time is to make the same number feel lighter. That's progress too.
• Missed reps, or speed fell off a cliff → hold, or back off ~5–10% and rebuild. One bad session is noise; two in a row is a message.

How much to add

Less than you want to. The smallest jump that still challenges you is almost always the right one, because it keeps the rep quality you built the progress on.

• Big lower-body lifts (squat, deadlift) tolerate bigger jumps — 2.5–5 kg / 5–10 lb.
• Upper-body and isolation work wants smaller — often 1–2.5 kg / 2.5–5 lb. Microplates exist for exactly this; a 20 kg curl doesn't want a 2.5 kg jump to 22.5.
• Add weight or reps, not both at once — change one variable so you can read what worked.

Common ways it goes wrong

• Adding weight you haven't earned — the 8, 8, 7 trap. Own every set first.
• Ego jumps — a big plate that buys you worse reps isn't progress, it's a worse version of last week.
• Changing everything at once — new weight, new rep target, new tempo. Now you've no idea which move helped.
• Never backing off — a weight that grinds three sessions running isn't asking for more grit; it's asking for a small deload so you can climb again.

How this works in Katado

This decision is the one thing Katado quietly makes for you between sessions. Finish a workout and the stat footer already shows a "Next time" weight for each lift, with a small ▲, →, or ▼ beside it. Bring that lift back — add it to today's session, or repeat the workout — and its first set is pre-filled as a dashed "planned" chip carrying that suggestion, read from your last top set (your heaviest working set) and how hard it felt. It's a starting point, not a command: if you disagree, spin the wheel and put your own number in. Each set still owns its own weight, reps, and effort, so the suggestion is a nudge, never a lock.

Two deliberate choices sit underneath it. The engine runs between sessions only — it never drops your weight mid-workout because one set felt heavy; that's your call in the moment. And the suggestion is built on what you actually did — the reps you hit and the effort they cost — not on a formula's estimate of a one-rep max you never lifted. When you beat your heaviest on a lift, a gold ★ PR quietly marks that session in your history.

The point isn't to automate the thinking away. It's to put the last session's signal right where the next decision happens, so picking your next weight is a two-second read instead of a guess. Read the last set honestly, add the smallest jump you've earned, and let it compound.