Marc Keys billed this as “the best fucking 12 weeks of powerlifting you’ll have all year,” and the structure backs the swagger up. It’s a 12 week meet prep run on 4 total-body days a week, where every session touches at least two of the three lifts and the squat spends most of the block parked at 85%.
Each lift gets its own logic rather than a shared wave. Squat runs preparation into a long accumulation block and then peaks. Bench alternates volume and intensity weeks because, in Keys’ words, it “hates accumulation training.” Deadlift pairs a deficit or pause variation with the competition pull every single session and gets a rep-out every third week.
This is an overreaching program and Keys says so himself. You need solid technique on all three lifts before you start, you should not run it back to back, and you should follow it with 4 to 6 lighter weeks. If you’re peaking for a meet and you’ve got 12 clean weeks to spend, it’s a good use of them. If you’re still adding weight session to session, run Candito’s 6 week program and come back to this one.
Table of Contents
Cast Iron Strength 12 Week Peaking Program Spreadsheet
Enter your squat, bench and deadlift maxes on the first tab and all 12 weeks fill in across 4 sessions each. There’s a rounding setting next to the maxes if you’re working in kilos. The second tab reads the same maxes, so you only type them once.
Program and original spreadsheet by Cast Iron Strength, where Keys writes up the reasoning behind each lift’s approach.
Why There Are Two Tabs
Keys published this program twice: as tables in his article, and as a spreadsheet linked from that article. The two disagree, and have since 2017. Squat and all the assistance work are identical in both. Bench differs across weeks 8 to 12, and deadlift differs in a handful of cells.
- Spreadsheet Version – an untouched copy of Keys’ own file, cell for cell.
- Article Version – the same program with bench and deadlift edited to match the tables in his write-up.
Neither is obviously the “real” one, so pick. The practical difference: the Article Version ends bench on a deload week, while the Spreadsheet Version ends it on the heaviest bench week of the whole block. If you’re peaking for a meet the week after, that matters. Arguing the other way, Keys writes that bench “just loves boring as fuck volume being recycled,” and the Spreadsheet Version is the one that recycles volume weeks. He also says the deload lands on “week 7 and 13,” and week 13 is past the end of the sheet, which fits a heavy week 12.
I’d take the Article Version for a meet prep and the Spreadsheet Version for an off-season block. If you have no strong feeling, use the Article Version.
Program Overview
- Length: 12 weeks, built for meet prep
- Frequency: 4 training days per week, plus an optional recovery day (pool or light cardio)
- Structure: total-body sessions, alternating a squat day and a deadlift/bench day
- Loading: percentages of 1RM for the main lifts, RPE and rough percentage ranges for assistance
- Deload: week 7
- Squat peak: ends with a double at 105%, or a max attempt
The two session templates repeat twice each week. Exercise selection inside each slot is yours.
| Squat + back day | Deadlift + bench day |
|---|---|
| Squat | Deficit or pause deadlift |
| Bench row or chest supported row | Bench press (pause the last rep) |
| Pause squat or front squat | Deadlift |
| Light bench assistance volume | RDL or touch-and-go deadlift |
| 2-3 upper back and shoulder movements | 2-3 upper back and shoulder movements |
Note on reading the tables below. The numbers are weekly totals. The spreadsheet splits each of them across the week’s two relevant sessions, so a weekly 4×10 shows up as 2×10 on each of two days. In the squat peak weeks the two squat sessions stop matching: the first carries the heavy work and the second drops to light volume.
Weekly Progression: Squat
Identical in both versions.
| Week | Phase | Sets x reps @ intensity |
|---|---|---|
| 1 | Preparation | 4×10 @ 60%, 4×8 @ 65%, 4×6 @ 70% |
| 2 | Preparation | 4×8 @ 65%, 4×6 @ 70%, 4×4 @ 75% |
| 3 | Preparation | 4×6 @ 70%, 4×5 @ 75%, 4×4 @ 80% |
| 4 | Accumulation | 10×3 @ 85% |
| 5 | Accumulation | 10×4 @ 85% |
| 6 | Accumulation | 10×5 @ 85% |
| 7 | Deload | 6×4 @ 80% |
| 8 | Accumulation | 10×6 @ 85% |
| 9 | Peak | 4×5 @ 90% |
| 10 | Peak | 3×4 @ 95% |
| 11 | Peak | 2×3 @ 100% |
| 12 | Peak | 1×2 @ 105%, or max out |
Week 8 is the one that gets people. It’s 60 reps at 85% in a single week, it lands right after the deload, and it’s the last hard week before the percentages climb. Keys calls accumulation blocks for squat “black magic” but only for 4 to 8 weeks depending on the lifter, which is exactly how long this one runs.
Weekly Progression: Bench Press
Bench alternates volume and intensity rather than accumulating. All reps have a pause on the last rep of the set. Weeks 1 through 7 are the same in both versions; the two columns only part company from week 8.
| Week | Spreadsheet Version | Article Version |
|---|---|---|
| 1 | 4×10 @ 62.5%, 4×8 @ 67.5%, 4×6 @ 72.5% | |
| 2 | 4×6 @ 72.5%, 4×5 @ 77.5%, 4×4 @ 82.5% | |
| 3 | 4×8 @ 70%, 4×6 @ 75%, 4×4 @ 80% | |
| 4 | 4×4 @ 80%, 4×3 @ 85%, 4×2 @ 90% | |
| 5 | Heavy single at RPE 8.5-9, then 5×3 at 20 kg below it | |
| 6 | Deload, 12×6 @ 60% | |
| 7 | 4×10 @ 65%, 4×8 @ 70%, 4×6 @ 75% | |
| 8 | 4×10 @ 65%, 4×8 @ 70%, 4×6 @ 75% | 4×6 @ 75%, 4×5 @ 80%, 4×4 @ 85% |
| 9 | 4×6 @ 75%, 4×5 @ 80%, 4×4 @ 85% | 4×8 @ 72.5%, 4×6 @ 77.5%, 4×4 @ 82.5% |
| 10 | 4×6 @ 75%, 4×5 @ 80%, 4×4 @ 85% | 4×4 @ 82.5%, 4×3 @ 87.5%, 4×2 @ 92.5% |
| 11 | 4×8 @ 72.5%, 4×6 @ 77.5%, 4×4 @ 82.5% | Heavy single at RPE 8.5-9, then 5×3 at 20 kg below it |
| 12 | 4×4 @ 82.5%, 4×3 @ 87.5%, 4×2 @ 92.5% | Deload, 12×6 @ 60% |
The Spreadsheet Version repeats week 7 into week 8 and week 9 into week 10. That’s what pushes the article’s last two weeks off the end.
Weekly Progression: Deadlift
Every deadlift day has three pulls: a deficit or pause variation, the competition deadlift, then RDLs. Every third week swaps the volume for a heavy double and a rep-out set at 85%, leaving one in the tank. The table below is the Article Version.
| Week | Deficit or pause | Deadlift | RDL |
|---|---|---|---|
| 1 | 10×4 @ 60% | 10×4 @ 65% | 4×12 @ 35% |
| 2 | 8×3 @ 70% | 8×3 @ 75% | 4×10 @ 40% |
| 3 | 2×2 @ 80% | Rep out @ 85% | 4×10 @ 45% |
| 4 | 10×4 @ 62.5% | 10×4 @ 67.5% | 4×12 @ 37.5% |
| 5 | 8×3 @ 72.5% | 8×3 @ 77.5% | 4×10 @ 42.5% |
| 6 | 2×2 @ 80% | Rep out @ 85% | 4×10 @ 45% |
| 7 | – | 12×4 @ 50%, sumo | – |
| 8 | 10×4 @ 60% | 10×4 @ 65% | 4×12 @ 35% |
| 9 | 8×3 @ 70% | 8×3 @ 75% | 4×10 @ 40% |
| 10 | 2×2 @ 80% | Rep out @ 85% | 4×10 @ 45% |
| 11 | 10×4 @ 62.5% | 10×4 @ 67.5% | 4×12 @ 37.5% |
| 12 | 8×3 @ 72.5% | 8×3 @ 77.5% | 4×10 @ 42.5% |
Deadlift is close between the versions. Five things differ in the Spreadsheet Version, and each one is a single cell rather than a different plan:
- Week 7 has no deadlift work at all. The article’s 12×4 @ 50% sumo is also where you recalculate your predicted max for the second half, which the back six weeks are loaded off, so this is the one worth knowing about.
- Week 2 pulls at 70% in the second deadlift session and 75% in the first.
- Week 8 drops the competition deadlift to 8×3 @ 65% and the RDL to 4×10.
- Week 11 runs the second deadlift session at 60% and 65% against 62.5% and 67.5% in the first.
- Weeks 3, 6 and 10 run the RDL as 4×8 instead of 4×10, and week 6 sits at 47.5% where the others are 45%.
Assistance Work
Assistance runs on a rolling 3 week cycle and the loading is deliberately vague. Keep the weight well inside what you can handle, cut the rest periods short, and let general fatigue take you close to failure rather than the load doing it. The spreadsheet starts you at 3 sets of 12 around 50-60% and tightens from there.
Keys wants these muscle groups hit 2 to 4 times a week:
- Lats – pulldowns or pull-ups, rotating grips
- Rhomboids, traps and lats – rows, rotating between machines, dumbbells and grips
- Shoulders – presses or raises
- Lower back – hyperextensions and reverse hypers
- Abs – bracing work: planks, weighted planks, rollouts, gymnastics progressions
Cast Iron Strength 12 Week Program FAQs
Which tab should I run?
Article Version if there’s a meet at the end of week 12, because it tapers bench into it. Spreadsheet Version if you’re using this as an off-season overreach and the last week doesn’t need to leave you fresh. Squat is the same either way, and squat is the lift this program is really built around.
Who is Marc Keys?
An Edinburgh-based strength coach who runs Cast Iron Strength. He was British and Commonwealth Senior Champion in 2011, sat on the 2012 Great Britain World Championship squad, and has held three British records. He’s coached athletes across more than 30 sports.
Can I run this twice in a row?
No, and Keys is blunt about it: “DO NOT RUN THIS PROGRAMME BACK TO BACK unless you want to end up injured.” It’s an overreach for squat and deadlift by design. Follow it with 4 to 6 lighter weeks before you go looking for the next block.
Is 105% realistic on the last week?
That’s the point of the 8 weeks of squat volume in front of it, and the spreadsheet gives you “or max out” as the alternative. Treat 105% as the target rather than the requirement. If your entered max was already a stretch, the double won’t be there and that’s information, not failure.
Does the 105% double apply to bench and deadlift too?
No. The 105% double is squat only. Bench finishes on either a deload or 4×2 @ 92.5% depending on the tab, and deadlift finishes on volume work in the 70s. Squat is the lift this program is really built to push.
I’m a bigger lifter. Do I need to change anything?
Keys flags this directly: if you’re a superheavyweight or sitting above a 460 Wilks, make the squat template less aggressive. The accumulation block at 85% scales badly at the top end because the absolute tonnage gets brutal.
What if I can only train 3 days a week?
The program leans on hitting each lift twice a week, so dropping to three days means halving one lift’s volume. If you have to, cut one of the two squat days and keep both deadlift/bench days, since the squat accumulation is the part most likely to bury you anyway. You’ll get less out of it, and I’d rather see you pick a block from the peaking programs that’s written for 3 days in the first place.
Updates
- Replaced four screenshots of the source site’s tables with real, readable tables, verified cell by cell against the spreadsheet. Added the frequency (4 days a week, which the post never stated), the two session templates, the assistance guidance and the author’s own warnings. Corrected “the program ends with a double @ 105% of 1RM” – that’s the squat only – and pinned the deload to week 7. The spreadsheet now carries two tabs, because Cast Iron Strength’s article and his own spreadsheet have disagreed on bench and deadlift since 2017; the original tab is untouched and a second tab follows the article’s tables.
