Training Max: Use 85% or 90% of 1RM?
Set a Training Max that gives your program room to work: start with 85–90% of your 1RM, choose the more repeatable number, then use AMRAP results to adjust the next cycle.
Your Training Max is the number 5/3/1 uses to calculate your work sets. Enter a recent 1RM or estimate one from reps, then choose 85% or 90% as your TM. Use 85% for uncertain maxes. Use 90% only for recent, repeatable maxes.
Know your 1RM? Calculate your 5/3/1 work sets from it. Not sure? Estimate your 1RM from reps first.
Use 85% by default
Choose 85% if you are new to 5/3/1, returning after time off, using an estimated or old max, deadlifting, or recovering inconsistently.
Use 90% only for recent maxes
Your max is recent, reps stay clean under fatigue, recovery is reliable, and you have handled percentage-based strength work before.
What is a Training Max?
A Training Max is a deliberately conservative max used for program math. If your true squat 1RM is 140 kg, your starting TM is usually 119–126 kg. Your program percentages are then calculated from 119–126 kg, not from the full 140 kg.
That buffer is the feature: it keeps bar speed cleaner, protects technique on tired days, and gives you more cycles of progression before the weights force a reset.
Plain rule: 1RM is what you can test. Training Max is what you can train from repeatedly.
Why Not Use Your True 1RM?
Your true 1RM is fragile. It varies 2–5% between training days depending on sleep, nutrition, stress, time of day, caffeine, and about thirty other variables you can't control. Programming percentages from it means:
- On peak days, 85% feels like 82% (easy — you're leaving progress on the table)
- On bad days, 85% feels like 90% (grinder — your program just became too heavy)
- The tighter your programmed rep ranges, the bigger these misses hurt
A Training Max smooths this out. Because you're programming from a number that's 10–15% below your peak, "bad-day 85%" is really only ~72–77% of your max — still manageable, still clean form, still crisp bar speed. The buffer is what keeps the program honest week after week.
The second benefit: progression lives longer. Adding 2.5 kg to upper lifts and 5 kg to lower lifts every cycle works for a long time when you're starting from 85% of max. It stops working almost immediately when you're already at 100%.
The 85–90% Rule
The canonical Wendler recommendation is 90% of true 1RM. Practice has shown that for most lifters, especially newer ones or anyone returning from a layoff, 85% is the better starting point.
| TM % | Use Case | Progression Speed |
|---|---|---|
| 80% | Deep deconditioning, significant time off, injury return | Very sustainable, slow |
| 85% | Default for most lifters, especially first cycle | Recommended |
| 87.5% | Experienced lifters with good recovery | Aggressive but doable |
| 90% | Classic Wendler, advanced lifters in good health | Short runway before reset |
| 92.5%+ | Meet-prep peaking cycles only | Reliably stalls within 2–3 cycles |
If you're new to TM-based programming, start at 85%. If the first cycle's Week-3 AMRAP comes in strong (8+ reps), that's a sign the starting point was conservative — adjust up for the next cycle through the standard increment, not a big jump (see below). Starting too low is easy to fix. Moving down after a missed rep is demoralizing.
How to Calculate Your Training Max
Simple formula: TM = 1RM × 0.85 (or × 0.90 if you're confident).
But that assumes you actually know your 1RM. Most lifters don't — they know their last heavy PR, which is something else.
Three ways to find your starting TM
1. From a recent max attempt (within 4 weeks):
If you've actually tested a true 1RM recently, multiply by 0.85 or 0.90. Done.
2. From a heavy rep set (within 2 weeks):
Use a 1RM calculator to estimate your max from a clean set of 3–5 reps. Then apply the 85–90% rule.
3. From gym experience:
If you haven't maxed out or repped heavy recently, pick a weight you're confident you could hit for 3 clean singles today. That's a rough approximation of 90% of max. Multiply that number by ~0.95 to get a sensible starting TM.
Rule of thumb: if you can't decide between two numbers, pick the lower one. The cost of starting too light is one easy cycle. The cost of starting too heavy is a missed rep in Week 2.
Adjusting Your TM Between Cycles
Standard Progression (Additive)
After every completed 4-week cycle where you hit all the reps:
- Upper body lifts (bench press, shoulder press): +2.5 kg (+5 lbs)
- Lower body lifts (squat, deadlift): +5 kg (+10 lbs)
Small, steady jumps are the point. Trying to add 5 kg to bench or 10 kg to squat every cycle will stall within 8–12 weeks even for beginners.
When to Reset Down
If any of these happen, reset your TM by 10%:
- You miss required reps on the Week-3 AMRAP (can't hit the 1+)
- Your AMRAP reps drop to weak-range numbers for 2+ cycles in a row
- Form breaks down on working sets that used to be clean
- You've been running the same TM for 4+ cycles and progress has stalled
A reset feels like losing ground. It isn't — you'll blow through the old stall weight 3–4 cycles later, usually with more reps than before. The buffer is why it works.
When to Recheck the Starting Point
If you hit 8+ reps on the Week-3 AMRAP (1+ set at 95% TM), your TM is probably conservative. That is a good signal, not an automatic reason to double the next jump. Read AMRAPs as a trend across the same lift, not as one-set proof. Stop when reps stop being technically consistent. Keep the standard increment by default. If the first cycle was clearly set too low, recalculate from a recent clean estimated 1RM and start the next block conservatively.
Training Max vs Estimated 1RM
These are two different numbers with two different jobs:
| Training Max (TM) | Estimated 1RM (E1RM) | |
|---|---|---|
| Purpose | Program percentages from | Track strength progress |
| Source | 85–90% of your actual max | Formula applied to a heavy set |
| Changes | Every cycle (+2.5 or +5 kg) | Continuously — every session produces a new data point |
| When to trust it | Always — it's chosen deliberately | Most when rep range is 1–5 |
The E1RM is a snapshot of what you could likely lift today. The TM is the conservative estimate you'll program from for the next 4 weeks. When your E1RM climbs significantly past what your TM assumes, that's a signal to review your setup — not to rush the progression. Most cycles should still use the standard TM increments.
Our 1RM calculator produces E1RMs from rep sets. Multiply that by 0.85–0.90 for a starting TM.
Common Mistakes
Setting TM = 95%+ of 1RM. Feels ambitious. Stalls in 2 cycles. The whole point is the buffer — without it, you're just doing a stiff percentage program with an extra name.
Setting TM from an old PR. If you haven't touched that weight in 6 months, it's not your 1RM anymore. Estimate from a recent rep set instead.
Never resetting. Some lifters keep bumping until they miss reps three cycles in a row. The reset feels like failure, so they avoid it. Net result: six stalled cycles instead of one clean reset.
Using one TM for all lifts based on ratios. Your squat, bench, deadlift, and press each have their own 1RM and their own TM. "My squat is usually 2× my bench" is a guideline, not a way to set your bench TM.
Bumping the TM weekly. The TM is a 4-week fixed reference. Changing it mid-cycle defeats the programming math and destroys the AMRAP signal.
Confusing TM and 1RM in reports. When telling a coach "my squat is 140", be specific: is that TM, true 1RM, or E1RM? These diverge by 15–20% and getting them mixed up derails programming.
Treating TM as a cap. The TM is the reference for programmed work, not a ceiling on your strength. The Week-3 AMRAP is only 95% of your TM, but a strong rep-out there can imply an estimated 1RM well above the TM — by design, it's a window into your actual current max.
Frequently Asked Questions
What is a Training Max?
A Training Max (TM) is an intentionally lower reference number used to calculate all program percentages — typically 85 to 90 percent of your true 1RM. Every set, every week, every cycle is scaled from the TM rather than your actual maximum, which creates a built-in safety buffer and enables long-term progression.
Why not just use my true 1RM?
Your true 1RM is a single, fragile number that varies 2 to 5 percent between training days. Programming from it means your working sets will be too heavy on bad days and perfectly calibrated only on peak days. The Training Max fixes this by leaving a buffer so programmed reps stay crisp year-round.
How do I calculate my Training Max?
Multiply your estimated 1RM by 0.85 to 0.90. Use 90 percent if you're confident in your 1RM estimate and recover well. Use 85 percent if you're a beginner, recovering from a layoff, or running a program for the first time. The lower you start, the longer you can progress before stalling.
When should I reset my Training Max?
Reset when you miss reps across multiple cycles or when your AMRAP numbers trend weak for 2 or more cycles in a row. Drop your TM by 10 percent and restart the progression. It feels like going backward, but you'll pass your stalled number faster than you would by grinding through.
Should my Training Max match my competition total?
No. Competition maxes test what you can do on one perfect day with full adrenaline, a spotter, and a peak. Training Max is what you can program from reliably on a regular Tuesday. They are different numbers, serving different purposes — and the TM is always lower.
Can I use different TM percentages for different lifts?
Yes. A common pattern: 85% TM on deadlift (where systemic fatigue is higher), 90% on bench (where recovery is usually faster). Match the buffer to where each lift's recovery runs. Just be consistent within each lift across cycles.
Try a Training Max in Action
The 5/3/1 Wendler calculator shows you exact weights for every set across a 4-week cycle — scaled from your Training Max, not your true 1RM.
Open the 5/3/1 Wendler Calculator → Need a 1RM first? Calculate it here →- Wendler, J. (2009). 5/3/1: The Simplest and Most Effective Training System to Increase Raw Strength. Jim Wendler LLC.
- Tuchscherer, M. (2008). The Reactive Training Manual. Reactive Training Systems.
- Zourdos, M.C., et al. (2016). Novel Resistance Training–Specific Rating of Perceived Exertion Scale Measuring Repetitions in Reserve. JSCR 30(1): 267–275.
- Brzycki, M. (1993). Strength Testing — Predicting a One-Rep Max from Reps-to-Fatigue. JOPERD 64(1): 88–90.
- Prilepin, A.S. (1978). Training Methods of Weightlifters. Soviet Sports Review 13(2): 76–80.