Progression

When your second set beats your first

You bench 80 for 7, rack it, sit down, and something feels off. Not bad, just light. Second set, same 80, and you get 9 out of it without much argument.

That feels like a good day. Most of the time it is telling you something less flattering: your first set wasn't ready.

WhyRep flags this one, and the flag has a consequence that catches people out. Here's what it means and what happens to your numbers afterwards.

The short version

• Set one is the set your progress is judged on.
• If any later set matches or beats it, that's flagged as a warm-up problem.
• On that day, your best set becomes the number used for the verdict.
• That better number then becomes your baseline, and it stays your baseline.
• So a normal set one next week can read as a regression against it.

Why set one carries the whole verdict

Progress gets measured by comparing set one of today against set one of the last comparable session on the same exercise. Not your average across sets, not your total, not your best. The first one.

The reason is that everything after set one is measuring something else. Set three is downstream of how long you rested, how hard set two was, and whether you were pacing yourself for a fourth. Those are real things, and none of them are the question "am I stronger than last week".

Set one is the cleanest read you get. It's the one set in the session that isn't contaminated by the rest of the session.

The number being compared is effective reps, which is reps plus reps in reserve. Seven reps with one left in the tank is eight effective reps, the same as eight reps taken to failure. That's covered properly in what counts as progressive overload.

So what does a later set beating it actually mean

When a correct warm-up has happened, later sets come out lower. That's the expected shape of a session. Fatigue accumulates, and the first set is the strongest one.

When that shape inverts, one of two things usually happened, and the methodology names both.

The first is that you didn't warm up enough, so your first working set was doing the job of a last warm-up set. The second is that set one was sandbagged, held back a little, with the real effort going into set two.

WhyRep doesn't try to tell those apart. It can't, from a logbook. It says a later set matched or beat your first and that this usually means set one wasn't warmed up. Which of the two it was is yours to know.

In the app the card reads "Warm up before your first working set", and it explains that set one is the one your progress is judged on. It doesn't change anything in your program. It's a note about how the session was run.

The day gets judged on your best set instead

Here's the part that surprises people. The flag doesn't void the session or ask you to log it again. Your best set of the day substitutes in as that day's comparison number.

If your baseline was 10 effective reps, and today you put up 8 on set one and 11 on set two, the day is judged on the 11. That's a progress verdict, not a flat one and not a discarded one.

Which is arguably generous. You got credit for the best work you actually did, on a day the app just told you was run badly.

And then it keeps that number

The substituted number becomes your new baseline. Permanently. There's a phrase for this in the methodology: numbers are numbers.

Follow the same example into next week. Your baseline is now 11, because that's what you did. You warm up properly this time and set one comes in at 10, which is a perfectly good set and better than the 8 you opened with last week.

That reads as a regression. Against 11, it is one.

This is deliberate and it's worth understanding rather than working around. Baselines don't get asterisks for the circumstances that produced them. The alternative is an app that quietly decides some of your best numbers didn't count, and once it can do that for one reason it can do it for others.

The practical version: an unusually good second set today raises the bar you're measured against from then on. That's not a punishment. It's just the bar being where you put it.

Where this actually sits today

This one ships. It isn't a documented rule waiting on an engine.

The comparison, the flag and the best-set substitution all run in the progression engine, which is live. The suggestion card is one of the ones the app can raise on a finished session, alongside the rest-period note and the plateau fixes.

One honest limit. The flag tells you a later set beat your first. It doesn't tell you which of the two causes it was, and it can't, because nothing in a logbook distinguishes an under-warmed set from a held-back one. Anything more confident than that would be the app guessing at your intent.

Common questions

Why is my second set stronger than my first?

Usually one of two things. Either set one wasn't warmed up, so your first working set was really your last warm-up set, or set one was held back and the effort went into set two instead. WhyRep doesn't try to tell the two apart. It flags that a later set matched or beat set one and leaves the diagnosis to you.

Which set does WhyRep judge my progress on?

Set one, compared against set one of the previous comparable session of the same exercise. Later sets carry fatigue, rest length and pacing, so they measure how the session went rather than whether you got stronger.

Does the warm-up flag change my verdict?

It changes which number is used. On a day a later set beats set one, that better set becomes the day's comparison number, so the verdict is judged on your best set rather than your first. The separate short-rest flag never changes a verdict, because rest periods happen after set one.

Does the better number get thrown away afterwards?

No. It becomes your new baseline and it stays. If set one comes in lower next session, that reads as a regression against the substituted number. Baselines don't get grace in WhyRep, including ones that arrived on an unusual day.

A note on sources

Everything on this page comes from WhyRep's progression detection methodology document, signed before it was implemented, and every rule described here is covered by a worked example in that document. The two causes named for the flag are the two the document names. It offers no third, and neither does this page.

Related reading

The metric this page keeps referring to is defined in what counts as progressive overload, and what happens when set one stops improving for weeks at a time is how many stalled sessions actually counts as a plateau.