Coming back to lifting with a pair of PowerBlocks

April 24, 2026 - Fitness

Three years ago I was a Starting Strength guy. I had finally clawed my way to a 240lb squat for 5, a 315 deadlift for 5, 135 on the bench, 125 overhead, and 110 power cleans — numbers that are not impressive in any absolute sense, but were hard-earned for me. Then life happened. A new job, a move, a stretch of travel, a long stretch of “I’ll start again next Monday.” Next Monday turned into next year, and next year turned into three.

The body remembers, but it does not forgive. I am 6’3” and currently 218lbs. Three years ago at the same height I was 195lbs and a lot of that was muscle; now a lot of it is not. I am softer in the middle, weaker everywhere, and I have lost the easy mobility I used to take for granted. I want to get back to 190lbs — leaner, more durable, more athletic. So this time the plan is different, and this time I am going to build it on what the actual research says, not on what some YouTube guy with abs and a sleeveless hoodie says.

This post is mostly for me — a contract I am signing with myself in public so I cannot quietly back out of it. But maybe it is also useful to anyone else who is trying to come back to lifting in a small space, with limited equipment, and with goals that have shifted from “max out the bar” to “fit into clothes I like and not hurt my back picking up a kid.”

What I have to work with
What the science actually says
Honest baseline
Goals, in order
The split
The exercises
Progression
Conditioning and weight loss
Nutrition
Sleep
Alcohol
The first 12 weeks
What I am measuring
The point
Sources

What I have to work with

The big change since the last time I wrote about a home gym is that I no longer have a barbell, a squat stand, or a bench. What I do have is a pair of PowerBlocks that go up to 90lbs each in 5lb increments (2.5lbs with the adder kit). That is 180lbs of total load in two hands, which sounds like a lot until you remember I used to put 240 on my back for sets of five.

That constraint actually drives most of the program design. Without a barbell I cannot meaningfully load a back squat or a conventional deadlift. Without a bench I cannot meaningfully load a flat bench press. Everything has to be built around movements that two 90lb dumbbells can challenge — which turns out to be mostly unilateral work, plus a lot of high-rep accessory work for the bigger muscle groups where the load cap is the real bottleneck.

Importantly, this is not a meaningful disadvantage for the goals I now have. A 2025 meta-analysis in Sports Medicine compared unilateral and bilateral resistance training and found no significant difference in muscle hypertrophy between the two (Kassiano et al., 2025). Strength gains follow the principle of specificity — train one limb at a time and unilateral strength wins; train two at a time and bilateral wins — but for size, it does not matter. So building a hypertrophy program around dumbbells and split squats is not a compromise; it is a perfectly valid choice.

What the science actually says

Before designing the program I want to be explicit about the principles I am building on. None of these are controversial; they are what the meta-analyses converge on as of 2024.

Variable	What the evidence says	Source
Frequency	Each muscle group at least 2×/week beats 1×/week on a volume-equated basis	[Schoenfeld et al., 2016][freq-2016]
Volume	~10+ hard sets per muscle per week to maximize hypertrophy; dose-response continues at higher volumes for many trainees	[Schoenfeld et al., 2017][volume-2017]
Load / rep range	Hypertrophy is similar across ~5–30 reps if sets are taken close to failure; heavier loads are needed for max strength	[Currier et al., 2023][currier-2023]
Proximity to failure	Going all the way to failure is not necessary for hypertrophy; stopping 1–3 reps short works just as well	[Refalo et al., 2023][refalo-2023]
Progression	Adding load and adding reps both produce similar hypertrophy over an 8-week block	[Plotkin et al., 2022][plotkin-2022]
Unilateral vs bilateral	No significant difference in hypertrophy between unilateral and bilateral exercises	[Kassiano et al., 2025][unilateral-meta]
Protein	~1.6 g/kg/day saturates muscle gain in normal training; 2.3–3.1 g/kg of LBM during a cut to preserve muscle	[Morton et al., 2018][morton-2018]; [Helms et al., 2014][helms-2014]
Rate of fat loss	0.5–1% of bodyweight per week preserves (and can even add) lean mass; faster loss costs lean mass	[Garthe et al., 2011][garthe-2011]; [Helms et al., 2014][helms-2014]
Sleep	Sleeping 5.5h vs 8.5h while in a deficit cut fat loss by 55% and tripled fat-free mass loss	[Nedeltcheva et al., 2010][sleep-2010]
Daily steps	8,000–10,000 steps/day is associated with the lowest all-cause mortality risk for adults under 60	[Paluch et al., 2022][paluch-2022]

If you want to skip everything else in this post, the one-line summary of the science is: train each muscle 2× per week with at least ~10 hard sets, leave a couple of reps in reserve on most sets, eat enough protein, lose weight slowly, and sleep. Everything below is just my attempt to operationalize that with two dumbbells.

Honest baseline

A few cautious sessions back in the basement, plus stepping on the scale and pulling out a tape measure, gave me roughly this picture:

Metric	Today	Goal
Bodyweight	218 lbs	190 lbs
Height	6’3” (75 in)	—
Goblet squat 5RM	~50 lbs	90 lbs (PowerBlock cap)
DB Romanian deadlift 5RM	~2 × 50 lbs	2 × 90 lbs (cap)
DB floor press 5RM	~2 × 40 lbs	2 × 75 lbs
DB seated overhead press	~2 × 35 lbs	2 × 60 lbs
Single-arm DB row	~50 lbs	90 lbs (cap)

Twenty-eight pounds is the headline number. The old barbell numbers are not really useful targets anymore — the equipment is different and the goal is different. What matters is whether the dumbbell numbers above creep up week over week, and whether the scale number creeps down.

Goals, in order

When goals conflict you have to know which one wins. Mine, in order:

Lose body fat, 28 lbs of it (218 → 190), at 0.5–1% of bodyweight per week — the rate the Garthe et al. (2011) data show preserves (and can even add) lean mass, and that the Helms et al. (2014) natural-bodybuilding review recommends.
Build muscle. Especially in the back, glutes, and posterior chain — the things that age the worst when you sit at a desk.
Rebuild work capacity. Be able to do real physical work without gassing out.
Reclaim strength. Not as a primary driver, but as a happy by-product.

When (1) and (2) conflict, fat loss wins. The recomposition fantasy (lots of muscle gained while lots of fat is lost) is rare in trained people; for someone three years detrained it is more plausible, but I am not going to bet the program on it.

The split

Four days a week, upper/lower, with cardio on the off-days. The frequency comes straight from the Schoenfeld et al. (2016) meta-analysis showing each muscle should be hit at least twice a week on a volume-equated basis to maximize hypertrophy. Four days is also the highest frequency I can recover from while in a calorie deficit; the Currier et al. (2023) network meta-analysis specifically found that “higher-load, multiset, twice-weekly” was the highest-ranked prescription for hypertrophy.

Day	Session
Monday	Lower A (squat-focused)
Tuesday	Upper A (push-focused)
Wednesday	Zone 2 cardio, 30–45 min
Thursday	Lower B (hinge-focused)
Friday	Upper B (pull-focused)
Saturday	Long walk or hike, 60+ min
Sunday	Off

The exercises

A note on rep ranges before the tables. The Currier et al. (2023) meta found that all rep ranges from low to high produce comparable hypertrophy as long as sets are taken with sufficient effort. So I am not going to be religious about hitting 8–12. For movements where the PowerBlock cap limits load (anything where my legs are doing most of the work), I will let the rep range drift up to 15–20+. For movements where 90lbs in each hand is genuinely heavy (overhead pressing, single-arm rowing, biceps work), I will stay closer to 6–10. Same hypertrophic stimulus either way.

All sets are taken to about 2 reps in reserve (RIR 2). The Refalo et al. (2023) meta-analysis found no meaningful hypertrophy advantage from training all the way to momentary muscular failure, and stopping a couple reps short lets me hit the next set fresher and recover better between sessions — which matters more in a calorie deficit.

Lower A — squat-focused

Exercise	Sets x Reps	Notes
Goblet squat	4 × 8–12	Hold one PowerBlock at chest
DB Bulgarian split squat	3 × 8–10/leg	Rear foot on couch or step
DB step-up	3 × 10/leg	Knee-height step
DB walking lunge	3 × 10/leg	Across the room and back
Standing DB calf raise	4 × 12–15	One foot at a time on a plate or step
Hanging knee raise or DB sit-up	3 × 12	Whichever I have the equipment for

Upper A — push-focused

Exercise	Sets x Reps	Notes
DB floor press	4 × 6–10	Pause briefly at the floor, no bouncing
DB seated overhead press	4 × 6–10	On the floor, back against a wall
DB Z-press	3 × 8–12	Seated on floor, legs straight
DB lateral raise	4 × 12–15	Strict, no swinging
DB skullcrusher (floor)	3 × 10–12	Lower behind the head
Push-up (deficit if easy)	3 × AMRAP–2	Hands on PowerBlocks for extra range; stop ~2 short of failure

Lower B — hinge-focused

Exercise	Sets x Reps	Notes
DB Romanian deadlift	4 × 8–10	The closest thing I have to a real deadlift
DB single-leg RDL	3 × 8/leg	Slow, controlled, light
DB hip thrust (off floor)	4 × 10–12	Shoulders on couch, PowerBlock on hips
DB front-foot elevated split squat	3 × 10/leg	Front foot on a plate
DB suitcase carry	3 × 30s	Heavy, one side at a time
Plank	3 × 45–60s

Upper B — pull-focused

Exercise	Sets x Reps	Notes
Single-arm DB row	4 × 8–10	Free hand on a couch or chair
Chest-supported DB row	3 × 10–12	Lying chest-down on an incline pillow setup
DB shrug	3 × 12–15	Slow, hold the top
DB rear-delt fly	4 × 12–15	Bent over, strict
DB hammer curl	3 × 10–12
DB curl	3 × 10–12	Supinated

Volume check

A quick sanity check against Schoenfeld et al. (2017), which found a graded dose-response relationship and recommended at least 10+ hard sets per muscle per week for hypertrophy:

Muscle group	Direct sets/week	Indirect sets/week	Total
Quads	13	4 (RDL/hinge)	17
Hamstrings/glutes	14	4 (squat work)	18
Chest	7	3 (push-up)	10
Shoulders	11	7 (press carryover)	18
Triceps	7	7 (press carryover)	14
Back	14	0	14
Biceps	6	4 (row carryover)	10
Calves	4	0	4

This sits comfortably above the 10-set minimum threshold for every muscle except calves, which I will probably bump up by adding a second calf raise on Lower B.

Progression

The PowerBlock has fixed weight jumps — 5lbs (or 2.5lbs with the adder), which is fine for upper body but a big jump for some accessory work. So progression is double progression:

Hit the top of the rep range on every set with good form (~RIR 2). If the prescription is 3 × 8–10 and I get 10, 10, 10 — graduate.
Add the smallest available increment next session. 2.5lbs with the adder, otherwise 5lbs.
Restart at the bottom of the rep range and climb back up over the following sessions.
If I miss reps two sessions in a row at the same weight, drop 5lbs and rebuild.

This works because — per Plotkin et al. (2022) — adding reps and adding load produce essentially the same hypertrophy over a training block. When the PowerBlock cap is the limit, I just keep adding reps until the muscle adapts; that is a perfectly valid form of progressive overload.

Conditioning and weight loss

Lifting will not, by itself, get the weight off. I have made that mistake before. Resistance training in a deficit preserves muscle but it is the deficit that drives the fat loss.

Day	Cardio
Mon–Fri	8,000+ steps minimum, every day
Wednesday	30–45 min zone 2 (treadmill incline walk or stationary bike — conversational pace)
Saturday	60+ min walk or hike outdoors

The 8,000-step floor is not arbitrary. The Paluch et al. (2022) dose-response meta-analysis of objectively measured step counts found that for adults under 60, all-cause mortality risk dropped progressively up to about 8,000–10,000 steps per day, after which it plateaued. So 8,000 is roughly the point of diminishing returns; I will aim higher when I can.

Zone 2 means I can hold a conversation but it would be slightly annoying to do so. I am deliberately not doing HIIT or metcons — those are great for someone who is already lean and fit, and a great way to interfere with lifting recovery for someone who is not.

Nutrition

This is the section that actually decides whether I hit 190lbs or not. Lifting and cardio matter, but you cannot out-train a bad diet — especially not at thirty-something with a desk job.

The numbers I’m working with

Step one is figuring out roughly how many calories my current body burns in a day, then eating less than that. There are a few ways to estimate this. The simplest practical heuristic that lines up well with the Mifflin-St Jeor equation for moderately active men is bodyweight in pounds × 14–15 for someone who lifts 4×/week and walks a lot.

Variable	Value
Bodyweight	218 lbs (98.9 kg)
Estimated maintenance	218 × 14.5 ≈ 3,160 kcal/day
Target rate of fat loss	1.0–1.5 lbs/week (~0.5–0.7% of bodyweight)
Required daily deficit	500–750 kcal
Starting cutting target	2,400–2,500 kcal/day

The 1.0–1.5 lb/week target sits right in the sweet spot from Garthe et al. (2011): their 0.7%/week group not only preserved lean mass but gained 2.1% LBM and increased 1RM strength, while their 1.4%/week group did neither. For me at 218lbs, 0.7%/week is exactly 1.5 lbs.

This is an estimate, not a prescription from God. After two weeks I will look at the trend and adjust:

Losing more than 2 lbs/week: add 200 kcal/day. Too fast costs lean mass.
Losing 1–1.5 lbs/week: stay the course.
Losing less than 0.5 lbs/week (and not just water fluctuation): drop another 200 kcal/day, or add 1,500 steps.

Macros

With ~2,500 kcal as the starting point, the macro split that the literature supports is:

Macro	Target	Calories	% of total	Rationale
Protein	200 g	800	32%	~0.9 g/lb of bodyweight; in the upper end of [Helms et al. (2014)][helms-2014]
Fat	70 g	630	25%	Middle of Helms’s 15–30% range; supports hormones and satiety
Carbs	270 g	1,080	43%	Remainder; fuels training and recovery
Fiber	35–40 g	—	—	~14 g per 1,000 kcal, the standard satiety/gut-health target
Water	3.5–4 L/day	—	—	Practical heuristic for a 100kg active adult

The protein number is the one that does the most work. Morton et al. (2018) put the plateau for muscle gain at ~1.6 g/kg/day in non-cutting trainees. Helms et al. (2014) recommend going higher — 2.3–3.1 g/kg of lean body mass — when in a deficit, specifically to protect muscle. For me at ~218 lbs and roughly 85% LBM, that works out to 195–270g/day. Two hundred grams is the floor; on harder training days I might hit 220+.

The other two macros are more flexible. Fat below ~0.3 g/lb starts to mess with hormones and satiety; carbs are mostly a training-fuel and adherence variable. If you genuinely prefer more fat and less carbs, swap them around — the meta-analyses on macronutrient ratios consistently find no difference in fat loss as long as protein and total calories are controlled.

A default day

Decision fatigue is the silent killer of diets. The single best adherence hack I know is to build a “default day” that I eat unless something specific overrides it, then only think about food when something is actually different.

Meal	Food	kcal	P	F	C
Breakfast	4 whole eggs + 1 cup oats (dry) + 1 banana + black coffee	620	35	22	75
Mid-morning snack	1 cup non-fat Greek yogurt + 1 cup berries + 1 tbsp honey	250	25	1	35
Lunch	6 oz grilled chicken + 1 cup cooked rice + 2 cups roasted veg + 1 tbsp olive oil	660	50	18	65
Pre-workout	1 scoop whey + 1 apple + 1 rice cake	280	28	2	40
Dinner	6 oz 93% lean ground beef + 1 medium sweet potato + large salad with vinaigrette	690	50	27	50
Total		2,500	188	70	265

That gets me within shouting distance of all the targets without having to think hard. If protein is short for the day, I add another scoop of whey before bed (cottage cheese works too). If I miss a meal because of a meeting or a flight, I do not “make it up” later — I just eat the next default meal and move on.

Adherence, the actual lever

The literature on which diet works “best” is mostly noise once protein and calories are controlled. The literature on adherence is much clearer: people who follow any reasonable diet consistently do better than people who follow the optimal diet inconsistently. So the rules I am giving myself are about removing friction, not optimizing micronutrients:

Track for 30 days, then stop. Long enough to recalibrate my eyeballs on portion sizes; short enough not to turn me into the guy who weighs his almonds at a dinner party.
Default day unless something is different. Restaurants, travel, social events get their own rules.
Protein at every meal. Easiest way to hit 200g without thinking about it.
Eat the carbs around training, not at midnight on the couch. This is a satiety and adherence point, not a metabolic one.
No drinking calories. Black coffee, water, sparkling water, plain tea. Juice and soda are a near-zero-effort way to evaporate the entire daily deficit.
One “off” meal per week, planned. Built-in releases beat unplanned binges every time.

Refeeds and diet breaks

Twenty-eight pounds at ~1.25 lbs/week is roughly 22–24 weeks of dieting, which is a long time. Continuous deficits drive metabolic adaptation (lower NEAT, lower resting metabolic rate, higher hunger hormones), so structured breaks help both physiologically and psychologically.

The plan:

Weekly refeed day. One day per week (usually Saturday, after the long walk) at maintenance calories — about 3,100 kcal, mostly extra carbs. Helps glycogen, training quality, and sanity.
Diet break every 8–12 weeks. Two consecutive weeks at maintenance. Not a “cheat fortnight” — still 200g protein, still mostly whole foods, just no deficit. This is what Helms et al. (2014) recommend for natural bodybuilders on long contest preps and what the Helms group later validated in the MATADOR-style intermittent dieting literature.

Supplements with evidence

Most supplements are a tax on people who do not read papers. The short list of things that actually have human RCT evidence behind them:

Supplement	Dose	Evidence
Whey protein	25–50 g/serving	Hits the protein target. [Morton et al. (2018)][morton-2018] meta-analysis.
Creatine monohydrate	5 g/day, every day	Single best-supported supplement for resistance-trained adults; [Helms et al. (2014)][helms-2014] include it as one of three with strong evidence.
Caffeine	100–200 mg pre-workout	Improves training output; same Helms et al. review.
Vitamin D3	1,000–2,000 IU/day if deficient	Most desk-bound adults at 40°N latitude are deficient; cheap insurance.
Fish oil (EPA/DHA)	1–2 g/day combined	Modest but real anti-inflammatory and cardiovascular benefits.

Notably absent: BCAAs (redundant if you eat enough total protein), pre-workout blends (caffeine plus marketing), fat burners (basically caffeine plus risk), testosterone boosters (do not work in eugonadal men).

Timeline

At a 1.0–1.5 lb/week rate, with a weekly refeed and one two-week diet break around the halfway point, the realistic timeline looks like this:

Phase	Weeks	Bodyweight	Notes
Cut block 1	1–10	218 → ~205 lbs	Easiest pounds; rate may even start a bit fast
Diet break	11–12	~205 lbs (hold)	Maintenance, same protein, no deficit
Cut block 2	13–22	205 → ~195 lbs	Slower; expect a real plateau or two
Final push	23–28	195 → 190 lbs	Last 5 lbs always slowest
Maintain at 190	29+	190 lbs	Eat at maintenance, keep training, reassess

Six to seven months. That is a long time, which is exactly why the adherence rules matter more than the macro precision.

Sleep

This is the part of the plan most people skip and it is the one with the most embarrassing evidence base. Nedeltcheva et al. (2010) put overweight adults through 14 days of moderate caloric restriction with either 8.5 or 5.5 hours of nighttime sleep opportunity, in a within-subject crossover design. Both conditions produced about 3 kg of total weight loss. But the proportion of that weight that was fat dropped by 55% in the short-sleep condition (1.4 kg of fat lost vs 0.6 kg), and the loss of fat-free body mass increased by 60% (1.5 kg vs 2.4 kg). Short sleep also increased hunger.

In other words: sleeping 5.5 hours instead of 8.5 hours during a diet means you lose less than half as much fat and almost twice as much muscle, for the same total weight on the scale. That is a brutal trade-off for the sake of a few more hours of doomscrolling.

So 7.5–8.5 hours of sleep, every night, in a dark cool room, with the phone in another room, is a non-negotiable input to this program. It is honestly more important than any individual training session.

Alcohol

Mostly out. Alcohol acutely impairs muscle protein synthesis and recovery, wrecks deep sleep, and adds calories that do not fill you up. I do not need a study to know that the post-bourbon version of me is not going to PR a goblet squat the next morning.

The first 12 weeks

Three four-week blocks:

Block	Weeks	Focus
Reintroduce	1–4	Just show up. Conservative weights. Learn the new movements. RIR 3–4 instead of 2. No soreness chasing.
Build	5–8	Push the top of every rep range. Add weight whenever I can. Cardio stays steady.
Consolidate	9–12	Ratchet up cardio slightly. Lock in the new bodyweight. Reassess and rewrite the plan based on actual data.

At the end of week 12 I will write a follow-up post with what actually happened versus what I planned. That public reckoning is half the point of writing this in the first place.

What I am measuring

I am only tracking five things, because tracking everything is the same as tracking nothing.

Bodyweight, weekly average — not daily reading. Starting at 218, target 190, at 1.0–1.5 lbs per week, per Garthe et al. (2011).
Waist circumference, weekly. A better fat-loss proxy than scale weight alone.
Top-set load × reps for every main lift, every session. Written down so progressive overload is auditable.
Sleep, average hours per night. If this drifts under 7, the rest of the plan is wasted effort.
Steps per day, because I cannot lie to my phone.

The point

I am not chasing a number on a barbell anymore. I am chasing the version of me that can pick up his kids without his back complaining, hike a real trail without getting winded, and look at himself in the mirror without flinching. The PowerBlocks are not the ideal tool for that, but the literature is pretty clear that they are enough — and the tool I have, used consistently for a year, beats the tool I want, used twice and abandoned, every single time.

If this works, the next post in the fitness category will be a follow-up in three months. If it doesn’t, it will be a post about why it didn’t, which is arguably more interesting anyway.

Sources

Currier et al. (2023). Resistance training prescription for muscle strength and hypertrophy in healthy adults: a systematic review and Bayesian network meta-analysis. British Journal of Sports Medicine 57(18):1211–1220. The largest network meta-analysis of resistance training to date (178 strength studies, 119 hypertrophy studies). Found that all rep/set/frequency combinations beat no exercise; higher-load multi-set thrice-weekly maximized strength; higher-load multi-set twice-weekly maximized hypertrophy.
Schoenfeld et al. (2016). Effects of Resistance Training Frequency on Measures of Muscle Hypertrophy: A Systematic Review and Meta-Analysis. Sports Medicine 46:1689–1697. Volume-equated, training each muscle 2×/week beat 1×/week (effect size 0.49 vs 0.30, p=0.002).
Schoenfeld, Ogborn & Krieger (2017). Dose–response relationship between weekly resistance training volume and increases in muscle mass: A systematic review and meta-analysis. Journal of Sports Sciences 35(11):1073–1082. Each additional weekly set per muscle increased hypertrophy effect size by 0.023; recommends 10+ sets/muscle/week as a minimum.
Refalo et al. (2023). Influence of Resistance Training Proximity-to-Failure on Skeletal Muscle Hypertrophy: A Systematic Review with Meta-analysis. Sports Medicine 53(3):649–665. No significant advantage for training to momentary muscular failure vs non-failure for hypertrophy. Stopping 1–3 reps short is fine.
Plotkin et al. (2022). Progressive overload without progressing load? The effects of load or repetition progression on muscular adaptations. PeerJ 10:e14142. Eight-week study; load progression and rep progression produced similar hypertrophy. Load progression slightly favored 1RM strength.
Kassiano et al. (2025). Comparison of Muscle Growth and Dynamic Strength Adaptations Induced by Unilateral and Bilateral Resistance Training: A Systematic Review and Meta-analysis. Sports Medicine. Nine studies, 703 screened. No significant hypertrophy difference between unilateral and bilateral resistance training. Strength gains follow the principle of specificity.
Morton et al. (2018). A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults. British Journal of Sports Medicine 52(6):376–384. Pooled 49 RCTs with 1,863 participants; protein supplementation augmented gains, with the benefit plateauing at total intakes above ~1.62 g/kg/day.
Helms, Aragon & Fitschen (2014). Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation. Journal of the International Society of Sports Nutrition 11:20. Recommendations: 0.5–1%/wk weight loss to preserve muscle; 2.3–3.1 g/kg of LBM in protein per day during a cut.
Garthe et al. (2011). Effect of Two Different Weight-Loss Rates on Body Composition and Strength and Power-Related Performance in Elite Athletes. International Journal of Sport Nutrition and Exercise Metabolism 21(2):97–104. Slow weight loss (0.7%/wk) increased LBM by 2.1% and 1RM strength; fast weight loss (1.4%/wk) did neither.
Nedeltcheva et al. (2010). Insufficient Sleep Undermines Dietary Efforts to Reduce Adiposity. Annals of Internal Medicine 153(7):435–441. Crossover trial: 14 days of caloric restriction with 5.5 vs 8.5 h sleep opportunity. Short sleep cut fat loss by 55% and increased fat-free mass loss by 60% for the same total weight loss.
Paluch et al. (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. The Lancet Public Health 7(3):e219–e228. For adults under 60, mortality risk dropped progressively up to ~8,000–10,000 steps/day, then plateaued.