Why Consistency Beats Intensity After 40
A 2-year, 811-person trial found diet composition barely predicted weight loss — but showing up to sessions did. Here's why the plan you can actually run beats the optimal one you can't.
I lost 12 pounds over about twelve weeks this winter. No hyper-precise macros. A few beers most weeks. Fried cheese more than once.
A year earlier, that would've been the reason I failed. This time, it was part of the plan.
Why did going all-in the first time backfire?
The first time I tried to cut weight, I did what a lot of motivated people do: went harder. Lifted more. Walked more. Told myself I was "eating clean" — without measuring anything.
Fat loss stalled. I panicked. My response wasn't strategic, it was reactive — I started weighing and logging everything to "tighten up," and that's when I found the actual problem: my protein intake was way lower than I thought. I'd probably been losing some muscle along the way, which meant a lower daily calorie burn, which meant fat loss got harder, which made me push harder, which made things worse.
It wasn't a motivation problem. It was a design problem. I had no guardrails — just effort.
Does the diet you pick actually matter that much?
Less than the fitness industry wants you to believe.
Sacks et al. (2009) ran a 2-year randomized trial — 811 overweight adults split into four diets with deliberately different macronutrient ratios: 15% vs. 25% protein, 20% vs. 40% fat, 65% vs. 35% carbohydrate. Every diet was calorie-matched. If diet composition mattered the way most advice implies, these groups should have diverged hard.
They didn't. At 2 years, weight loss ranged from 2.9 to 3.6 kg across every comparison — protein, fat, and carb ratio alike — with P > 0.20 for every single comparison. Statistically, the diets were indistinguishable.
Weight loss at 2 years, by diet composition
| 15% protein | 3.0 kg |
|---|---|
| 25% protein | 3.6 kg |
| 20% fat | 3.3 kg |
| 40% fat | 3.3 kg |
| 65% carbohydrate | 2.9 kg |
| 35% carbohydrate | 3.4 kg |
What did move the needle: attendance at group counseling sessions. It strongly predicted weight loss at 2 years — 0.2 kg lost per session attended — and that relationship held equally across all four diets. The researchers called attendance a proxy for commitment and engagement. Not what the plan said. Whether the person kept showing up to it.
That's the whole argument for tracking adherence instead of obsessing over the "optimal" protocol. The protocol barely moved the outcome. Showing up did.
What did "selective discipline" actually look like?
I stopped asking "what's the most optimal fat-loss protocol?" and started asking "what plan would I actually run for three months?"
Here's what I know about myself: I don't enjoy hyper-precise macro tracking. I'm not giving up drinks with friends, though I'll reduce them. I love certain foods too much to pretend I'll never eat them again. But I can train consistently, prioritize protein, and be patient if I trust the process is working.
So instead of overriding those traits, I built a system around them:
- A moderate deficit, not an aggressive one
- Protein as a daily anchor (eyeballed, not weighed)
- Strength training 5x/week, zone 2 cardio after lifting
- Weekly average weigh-ins only
- One hard rule: no changes unless the trend is flat for two full weeks straight
That last rule mattered more than any of the others. There was a week the scale didn't move at all. Old me would've slashed calories or piled on cardio. Instead, I waited. The following week, the trend resumed. No drama, just physics on a longer timescale than I wanted to admit.
I was strict about training, protein, trend tracking, and patience. I was flexible about exact macros, social plans, and the occasional plate of fried cheese. That split reduced friction — and less friction is what let the plan survive twelve straight weeks instead of collapsing in three.
Why not just do the faster, more aggressive protocol?
There's a version of this that would've moved faster on paper — something like a Protein Sparing Modified Fast: very low calories, very high protein, rapid loss. Physiologically, it works.
Behaviorally, it wouldn't have worked for me. It demands rigid food control and leaves almost no room for a social life, on top of a training and cardio load that already asks a lot of recovery. Could I have lost weight faster? Maybe. Would I have run it for twelve weeks while lifting, training, and having an actual life? No chance.
A plan that's better on paper is worthless if it collapses under the life you actually have to live during it. The Sacks trial is the population-level version of the same finding: the "best" diet on a spec sheet loses to the one somebody keeps attending.
So how do you actually pick the plan, not just the goal?
Before starting the next attempt, ask honestly:
- What do I consistently quit, historically?
- What foods or habits am I not realistically giving up?
- Where am I already naturally consistent?
- What's the one metric I'll actually track every week?
- How long am I genuinely willing to sustain this?
Then design around those answers — not the aspirational version of yourself, the actual one. The plan you can run for twelve weeks beats the perfect plan you abandon in three, every time. (For the full data from this stretch — the day-by-day trend, the flat week, the recovery — see why one bad week isn't the failure mode.)
This is the same reasoning behind Fortiv's Adherence Engine: it tracks whether you showed up — training, zone 2, overall consistency — as a rolling score, not whether you hit a "perfect" macro target. Miss a session because of a work trip or a dinner out and the score dips; come back and it recovers. It's not measuring the optimal plan. It's measuring the one you're actually running.