I ate chicken and rice every day for three years. Weighed every gram. 98% compliance, and the 2% was mostly skipped cheat meals, not missed ones.
I'm not doing that anymore, and that's the point of this post.
That version of training was optimized for a number on a scale and a look in a mirror. This version is optimized for something else entirely: how much useful thinking I can do at 9pm on a Wednesday after a full workday, when the actual work -- the stuff I own -- finally gets its turn.
The body is infrastructure for the brain. I train the infrastructure.
The Asset You're Actually Protecting
If your income comes from thinking, your training goal is not aesthetics. It's not longevity in the abstract either. It's cognitive output per day, sustained over decades.
That reframe changes every decision downstream.
A two-hour leg session that leaves you useless for the rest of the night is a loss under this objective, even though it's a win under a bodybuilding objective. Same work, same soreness, opposite verdict -- because the metric moved from "did I stimulate hypertrophy" to "how many good hours did I have afterward."
I stopped asking "was that a good workout" and started asking "how was the block of work after it."
Those are different questions and they have different answers more often than you'd think.
The Minimum Effective Dose
Three sessions a week. 45 minutes each. Full body, compound movements, stop two reps short of failure.
That's it. That's the program.
- Squat or deadlift pattern -- one per session, alternating
- Push -- bench, overhead press, dips
- Pull -- rows, chin-ups
- Carry or hinge accessory -- 5 minutes, whatever's weakest
- Walk daily -- 30+ minutes, not optional, not counted as training
Three sessions is where the evidence says you capture the overwhelming majority of the strength and health return. Four and five add real muscle if that's the goal. They also add recovery cost, time cost, and -- the one that actually kills programs -- decision cost.
Stopping short of failure is the specific lever for mental output. Training to failure spikes systemic fatigue way out of proportion to the extra stimulus. You buy a slightly better set and pay for it with the rest of your evening. Bad trade when the evening is the product.
I'm leaving gains on the table. Deliberately. The table is not where I'm trying to win.
Why Lifting, Specifically
Cardio is great for the brain and I do it -- that's the daily walk, plus stairs, plus not sitting for nine straight hours.
But resistance training is doing something the walk isn't: it's defending the floor.
Muscle mass and strength are among the better predictors of how well you function in your seventies and eighties. Grip strength alone tracks with all-cause mortality well enough that it's used as a clinical proxy. Bone density responds to load and basically nothing else. And insulin sensitivity -- which is a cognition issue as much as a metabolic one -- improves more from having muscle to put glucose into than from almost anything else you can do.
None of that shows up next week. It shows up in decade four, which is exactly the problem: the returns are invisible right up until they're unrecoverable.
That's the same structure as compound interest, and I treat it identically. It gets automated and scheduled rather than left to how I feel about it on a given Tuesday, because a decision that only pays off in 30 years will lose every single negotiation against a decision that pays off tonight.
Sleep Is The Actual Multiplier
Uncomfortable truth: if you're sleeping six hours, your training program doesn't matter very much.
Sleep deprivation hits exactly the functions I'm training to protect -- working memory, attention, emotional regulation, the ability to hold a complex system in your head. One bad night measurably drops cognitive performance. A chronic deficit just moves your baseline down and then hides the fact that it did, because the thing you'd use to notice is the thing that's impaired.
So the honest hierarchy:
- Sleep 7-8 hours. Everything else is downstream.
- Walk daily. Cheapest return available.
- Lift 3x/week. The long-horizon insurance policy.
- Eat enough protein. Roughly 0.7g per pound of target bodyweight. This is the only nutrition rule I still track.
- Everything else. Supplements, protocols, optimization content. Noise until 1-4 are boringly consistent.
Most people work this list backwards. They'll read about cold plunges while sleeping five and a half hours. The order isn't a preference -- it's a dependency chain, and you can't optimize a layer you haven't built yet.
The Part I Get Wrong
I'm not going to write a fitness post pretending I'm compliant.
The three-year chicken-and-rice run was real, and so is the fact that I can't currently claim anything close to that consistency. Training is the first thing that falls off when a build sprint takes over -- the same months that produced 499 commits produced approximately zero sessions, and the months where nothing shipped weren't exactly disciplined in the gym either.
That's not a humility bit. It's a diagnosis. The system I built for money runs without me and the system I built for training doesn't exist, so it loses every negotiation against a deadline. Same failure mode, different domain: I relied on wanting it instead of building the mechanic.
Which means the fix isn't a better program. The program above is fine. The fix is the scheduling layer -- fixed slots, same three days, treated like a meeting with someone I can't reschedule on.
I'm writing that one down publicly on purpose. A public deadline is a different animal than a private intention.
Practice How You Play
Three sessions. 45 minutes. Compounds, short of failure, protein hit, walk daily, sleep like it's load-bearing.
Not because it's optimal for muscle. Because it's optimal for having a functioning brain at 9pm, and for still having one in forty years.
The body is the machine the work runs on. You don't get a second one, there's no respawn, and by the time the maintenance debt comes due it isn't payable.
Protect the asset.