Estimating your 1RM without maxing out
Your one-rep max is the heaviest weight you can lift once with good form. Testing it directly is tiring and carries injury risk, so lifters estimate it from a sub-maximal set instead. Several formulas do this; the four most cited are Epley (w × (1 + reps ÷ 30)), Brzycki (w × 36 ÷ (37 − reps)), Lombardi (w × reps0.1) and O'Conner (w × (1 + reps ÷ 40)). For 225 lb × 5 they give 262.5, 253.1, 264.3 and 253.1 lb — an average of about 258 lb. The spread between them is the honest error bar: a few percent at low reps, more as reps climb.
Use sets of 1–6 reps for the best estimate. The formulas were fitted to data in that range; beyond about 10 reps, muscular endurance matters more than strength and the estimates drift high. A set of 12 taken to failure will overestimate what you can actually lift once. If you have several recent sets, the one with the fewest reps at the heaviest weight usually gives the most reliable figure.
Programming from percentages. Most strength programs prescribe loads as a percentage of 1RM: 65–75% for hypertrophy sets of 8–12, 80–90% for strength sets of 3–6, 90%+ for peaking singles and doubles. The table converts your estimate into those working weights and rounds them to a loadable increment — 5 lb or 2.5 kg by default — so you can walk to the rack with a number.
Accuracy varies by lift. Epley and Brzycki were developed on the bench press; deadlift and squat estimates tend to run a little lower relative to a true max, and machine or isolation exercises are less predictable. Re-estimate every few weeks as you get stronger rather than training off a stale number.