Skip to content
All posts
Journal9 min read

GZCL method with an AI coach: how Yuge programs T1, T2, and T3 work

The tiered structure is simple. Running it well isn't. Most GZCL apps treat the tiers as three independent programs bolted together. They're not. Here's how Yuge handles GZCL as a system — T1 progression, T2 volume management, T3 autoregulation, and what happens when any tier stops working.

GlenBuilding Nahyeh
GZCL method with an AI coach: how Yuge programs T1, T2, and T3 work

If you've run GZCL for any length of time, you already know the structure. T1 for competition lifts, heavy and low-rep. T2 for close variations, moderate volume. T3 for accessories, high-rep, autoregulated. The theory is clean: each tier serves a different purpose, and together they build strength and hypertrophy in proportion.

The practice is harder. T1 progression seems straightforward until your 5RM stalls and you're not sure whether to reset or keep pushing. T2 volume is easy to miscalibrate — too little and it doesn't do much, too much and it eats into T1 recovery. T3 selection is often an afterthought, rotated manually only when someone mentions it on Reddit.

Most apps get the percentages right. Few get the system right. Here's how Yuge handles it.

T1: competition lifts, wave loading, and when to reset

T1 is where GZCL starts — your heaviest compound movement, trained in the 1-5 rep range, with loads derived from a tested 5RM. The standard prescription is 85%+ for clusters of 2-3 reps across multiple sets, working toward a heavy single or double as the top set.

When you start with Yuge, you don't input a 5RM estimate from memory. You start with a conversation. The coach asks about your recent performance on that lift, what the last few training cycles looked like, and how the weight has been feeling. From that, it establishes a working 5RM and sets the T1 loading accordingly. Conservative start is the default — it's easier to increase than to recover from an overcalibrated opener.

Progression. Yuge runs T1 on a percentage-based wave. After each T1 session, it reads the performance data and RPE. If your top set hit the target with something left in the tank, loads advance next session. If you hit it but RPE was near ceiling, it holds the load and watches the next session before advancing. If you missed reps, it tracks whether that was a one-off or a pattern.

The progression is modest by design. GZCL T1 is strength expression, not daily maxing. The goal is to move steadily upward without ever grinding. When lifters run GZCL in a spreadsheet they often push T1 progression too hard after a good session. Yuge doesn't respond to individual sessions that way — it responds to trends.

Resets. The most common T1 mistake is running loads too high for too long. RPE starts creeping up. Rep targets become hard to hit. Rather than resetting at the first sign of struggle, the lifter keeps grinding. Eventually they miss, stall, and reset from a worse position.

Yuge resets proactively. When RPE trends upward across two consecutive cycles, or when rep performance is consistently below projection, it proposes a reset at 90% of the current 5RM estimate and explains why. You're not blindsided. You can push back if you think the data is wrong.

T2: close variations, volume bridge, and the hypertrophy layer

T2 is where most GZCL programs live or die. The prescription — moderate intensity, 3x10 or 4x8, close movement variations — sounds straightforward. The execution has more variables than it looks like.

Which movements go here? How much total volume? How do loads progress? What happens when T2 starts interfering with T1 recovery?

Movement selection. Yuge matches T2 movements to T1 patterns. If your T1 is squat, T2 pulls from a pool of close squat variations — pause squats, front squats, SSB squats, high-bar vs low-bar if you usually train one style. The coach picks based on what you have access to, what your training history shows, and whether any movements have been flagged as problematic. It asks if you have preferences before finalising, but it doesn't ask you to pick from scratch.

The T2 movement for upper days follows the same logic. T1 bench gets a close variation like incline press, close-grip bench, or floor press. T1 overhead press gets a seated press or dumbbell variation. The T2 movement is a training tool, not a competition exercise, so variation within the pattern is fine.

Volume management. T2 volume is the knob most people get wrong. Jacked and Tan 2.0 prescribes generous T2 volume. That works well when recovery is solid. When T1 intensity is high or life stress is elevated, it's often too much — the T2 work doesn't recover in time and T1 performance suffers.

Yuge tracks the relationship between T2 volume and T1 performance over time. If T1 RPE has been trending upward alongside high T2 volume, it adjusts. It might pull a set off T2, drop the rep range, or hold volume flat while T1 intensity climbs through a peak block. The adjustment is small and explained, not a wholesale program overhaul.

T2 progression. T2 loading follows a slower arc than T1. The goal is volume accumulation and hypertrophy stimulus, not strength expression. Loads advance when rep targets are hit cleanly across multiple sessions at the same weight. RPE targets are kept moderate — 7-8 is the range, not 9+. If T2 sessions are regularly hitting RPE 9, either the weight is too high or recovery is compromised. Yuge flags this and adjusts.

T3: accessories, autoregulation, and rotation

T3 is the flexibility tier. High reps, lower loads, as many movements as you want to include. The tradeoff is that T3 is also the tier that gets neglected or run on autopilot.

The typical approach: pick some accessories, run them at the same weight indefinitely, rotate when they feel stale. That works. It's also leaving something on the table.

Autoregulation. Yuge runs T3 on RPE, not fixed loads. You log the weight and reps, report how it felt, and the coach adjusts load for the next session accordingly. The target RPE for T3 work is 8-9 — you should be working, not grinding. If you're consistently hitting RPE 7 at the same weight, load goes up. If you're hitting RPE 9+ and form is breaking down, it holds or drops.

This sounds like extra overhead, but in practice it's a quick check-in. Voice logging makes it faster — you say "12 reps, felt like an 8" and it's logged in seconds.

Rotation. T3 movements should rotate when they stop producing stimulus. That usually means 4-8 weeks before switching. Most lifters do this too infrequently because remembering what you ran 6 weeks ago is hard without a record.

Yuge tracks T3 movement history automatically. When a movement has been in rotation for 6+ weeks without clear progression, it flags it and suggests a swap from a pool of appropriate alternatives. If your T3 curl variation has been in place for 8 weeks and the loads haven't moved in three sessions, the coach asks whether you want to cycle it out. It keeps a record so you're not running the same movements back-to-back.

Selection logic. T3 selections aren't arbitrary. The coach cross-references T1 and T2 movement patterns when building the T3 pool. If squat is T1 and front squat is T2, the T3 lower movements skew toward hip hinge and posterior chain rather than adding a third quad-dominant movement. The total training stimulus across all three tiers should be balanced, not accidentally redundant.

The thing most GZCL apps get wrong

Most apps that support GZCL treat the tiers as independent programs. T1 has its own progression logic. T2 has its own volume rules. T3 is whatever you add. Nothing talks to anything else.

The methodology doesn't work that way. The tiers are load-sharing across a system. When T1 stalls, T2 volume is often the culprit. When T3 stops producing stimulus, the issue is sometimes that T2 has drifted into the same patterns rather than complementing them. When a T1 movement gets substituted because of pain, the T2 and T3 selections need to shift with it — you can't keep running close-grip bench in T2 if your T1 bench is out because of shoulder impingement.

Yuge treats program changes as system changes. If your T1 squat stalls, the coach looks at whether T2 squat variation volume is interfering with T1 recovery before assuming the issue is T1 programming. If T1 bench gets subbed out for an injury, T2 and T3 upper movements reconfigure accordingly. The tiers stay coherent.

This is the main place where running GZCL in a spreadsheet breaks down. You can copy-paste a T2 substitution without noticing that it's now duplicating the T1 stimulus. Yuge doesn't let that happen silently. See how Yuge handles training interventions for more detail on how substitutions propagate through the program.

Spreadsheets and Boostcamp

The Cody Lefever GZCL spreadsheets are good. They implement the percentage logic correctly, they have sensible T2 volume prescriptions, and if you understand the methodology you can run them effectively. What they can't do is monitor the relationship between tiers, notice when T2 volume is eating T1 recovery, or update T3 selections when the T1 movement changes. You handle all of that manually.

Boostcamp's GZCL implementation deserves credit — the 5RM percentages are right and the structure is accurate. It's the best passive implementation available and it's free. Where it stops is adaptation. The program runs as written. Miss two sessions, the program doesn't know. T2 volume becomes problematic, the program doesn't adjust. You make those calls manually or you don't.

The difference with Yuge isn't a better version of the spreadsheet. It's a coach watching the system over time and flagging when the relationship between tiers is breaking down — before you've been grinding at an overcooked T1 for six weeks and wondering why your recovery is shot.

GZCLP, Jacked and Tan 2.0, and variants

GZCL isn't one program — it's a methodology that people have applied in different ways.

GZCLP is the linear progression variant designed for newer lifters. T1 work starts with 5x3 and progresses to 6x2 and 10x1 before resetting. It's simpler than standard GZCL and more appropriate for someone still making weekly strength gains. Yuge handles GZCLP as a distinct entry point — if the conversation suggests you're earlier in your training career, it'll propose GZCLP rather than assuming standard GZCL, and it explains why linear progression still makes sense for your stage.

The transition from GZCLP to standard GZCL happens when linear progression stalls — when you're resetting regularly and not recovering the weight within a few cycles. Yuge monitors this and can flag when it's time to move to percentage-based periodisation.

Standard GZCL is what most intermediate lifters are running. Percentage-based T1, volume-managed T2, autoregulated T3. This is the main implementation described above.

Jacked and Tan 2.0 is Cody Lefever's specific program template built on the methodology. Higher T2 volume than many standard implementations, specific movement pairings, more prescriptive structure. Yuge can run J&T 2.0 as a template and adapts it the same way — T2 volume adjusts based on T1 performance, T3 rotates on schedule. If you want to follow J&T 2.0 closely because you've run it before and know it works for you, that's a valid starting point. If you want the structure of J&T 2.0 with adaptive management on top, that's what Yuge does.

You can also ask for a hybrid — GZCL structure with DUP loading on T1, or block periodisation layered over the T2 volume prescription. The methodology is a framework. How rigid you want the implementation to be is up to you.

If you're weighing GZCL against 5/3/1, we've made the same case there — different methodology, the same place where a spreadsheet stops being enough. And for the within-week version of structured loading variation, see daily undulating periodization.

If you want to run GZCL properly — with the tiers actually talking to each other — Yuge is in early access.

References

  1. 01Helms ER, Morgan A, Valdez A (2019), "The Muscle and Strength Pyramid: Training", 2nd ed. Open paper

Early access

Want early access?

If you're a serious lifter who wants AI that understands your training methodology, we'd like to hear from you. Gym owners should check out Hoist. Early adopters get first access.

Early access. We'll let you know when it's ready.