You did 5. Six formulas disagree about your single.
From —, taken to the point where one more was not happening. Six published formulas put it between —; the figure above is the middle of them. On this lift, —.
The six, with your numbers in them
The panel above prints six results. Here each formula is written out in full with your weight and reps substituted, so the number can be checked on a calculator instead of taken on trust.
Where they stop agreeing
At one rep the six formulas nearly agree. Twenty reps later the highest of them reads more than half again what the lowest does. That gap is the whole argument for printing a band.
Every percentage, loaded
Nine working percentages off that estimate, each rounded to a weight the bar can actually hold. The highlighted row is about where the set you typed in sits.
Each row draws one side of the bar: a 20 bar with a single 20 on each end reads 60.
If the set was not to failure
Every number on this page assumes the rep you stopped on was the last one available. If it was not, here is what the same set is actually worth.
Why six numbers and not one
- These are regressions, not physics. Each formula is a curve somebody fitted to a room full of lifters doing sets to failure and then testing a true single. Different rooms, different decades, different curves. None of them knows anything about you.
- Most were fitted on the bench press. That is the easiest lift to take to honest failure without getting hurt, so it is the one the data exists for. Applying the same curve to a deadlift is an extrapolation, and the page says which lift you picked for exactly that reason.
- They agree at low reps and fall apart at high ones. A long set measures how long you can keep going as much as how much you can lift, and the six curves disagree about how to convert one into the other. Where they stop agreeing plots all six against rep count; the band widening from left to right is that disagreement.
- The set has to have been a real one. These curves assume the rep you stopped at was the last one available. The arithmetic cannot tell the difference and will not warn you — If the set was not to failure is the same set recomputed with one, two and three reps left in the tank.
- Reps do not transfer between lifts. A squatter who gets ten at 70% will often get five at the same percentage on a deadlift, where grip and lower back quit before the legs do. Percentages are a starting point for the lift they came from.
- The rep count beside each percentage runs the same six formulas backwards. There is no closed-form inverse, so the page tries one to twenty reps and keeps whichever lands closest to that percentage. It reads off the middle of the six, the same rule the headline follows — not one formula quietly picked for you.
- The bar cannot make every number. With a 20 kg bar and 1.25 kg as the smallest plate, loadable weights come in 2.5 kg steps — so 82% of your max may simply not exist. Each row shows what you actually get and how far off the target it lands.
- An estimate is not a test. The one number nobody on this page has is the one you get by actually loading the bar and finding out. Everything here is what the last twenty-five years of published regressions predict, which is a good place to set next week's working weights and a poor place to set a personal record.