O'Conner made formulas for each lift that also included things like the lifter's body weight and muscle mass.
https://wikimedia.org/api/rest_v1/media/math/render/svg/031c2e1b81197b31ab6fd7a1267121918b5b169e
The O'Conner et al. 1RM prediction formula is another approach that focuses on lifting type and individual characteristics.
The formulas are:
For Back Squat: 1RM = Weight / (1.0703 - (0.0231 x Reps))
For Bench Press: 1RM = Weight / (1.0857 - (0.0516 x Reps))
For Deadlift: 1RM = Weight / (1.0827 - (0.0465 x Reps))
Where: Weight = Load lifted (kg) Reps = Number of reps completed
O'Conner et al. derived these formulas from analyzing strength data from 84 resistance-trained individuals performing the back squat, bench press, and deadlift exercises.
A key aspect of their approach was developing separate prediction equations for each lift, similar to Mayhew et al.'s method. This allowed accounting for different strength characteristics between lifts.
However, O'Conner et al. went a step further by also incorporating individual characteristics like body mass, lean body mass, and strength levels into their analysis.
Through multiple regression modeling, they determined that incorporating these individual factors alongwith reps and load improved the accuracy of 1RM predictions for each lift compared to just using reps and load alone.
The different coefficient values in each lift's formula (e.g. 1.0703, 0.0231 for back squat) represent the best-fit parameters when accounting for these additional individual characteristics.
While requiring more inputs beyond just reps and load, O'Conner's approach demonstrated potential increases in 1RM prediction accuracy, especially across individuals with different body compositions and strength levels.
This highlights incorporating additional individual factors, not just the strength curve characteristics, as another angle for improving 1RM prediction models.