Building the Engine: Run Economy to Race Day

Speed didn't show up when I "tried harder." It showed up when I got cheaper -- less energy per kilometer. For months I thought running faster meant pushing harder. Turns out it meant wasting less.

This is the story of building a running engine from a 7:11/km baseline, and then what happened when HYROX Mumbai stress-tested every piece of it. The engine-building part is methodical and boring. The race part is not.

For most of my life, I avoided running. When I finally started, I made the mistake everyone makes: I tried to run faster by running harder. More effort, more suffering, more gasping. The result was predictable -- I got tired faster, not faster. The breakthrough came when I understood that running economy is the game. You don't need a bigger engine if you stop burning fuel on friction. Smoother, cheaper, more repeatable. That's the whole thesis.

Baseline to target

Why this matters for HYROX: every station gets "easier" when the run is cheaper. Lower HR coming into sleds/lunges = free time. You don't win HYROX by running fast. You win by arriving at stations with something left.

The 2-lever plan

I didn't overcomplicate this. Two sessions a week. That's it.

Lever 1: Threshold intervals (raise the ceiling) -- Tuesdays

Lever 2: Zone 2 long run (build the floor) -- Fridays

Everything else is optional. These two levers move the curve.

8-week progression

Down-week tip: if fatigue climbs, drop 1 rep on Tuesday and 10-15 min on Friday -- keep the pattern.

Form and cadence notes

Posture: Tall, crown of head up, ribs stacked. Land under center of mass; let speed (not reach) lengthen stride.

Cadence: Aim >170 spm on easy runs. Don't force huge jumps -- add +3-5 spm over 2-3 weeks.

Breathing: 2-in / 3-out rhythm for Z2; 2/2 at threshold if needed.

Strides (technique, not sprints): Relax shoulders/jaw, quick feet, pop off the ground. If you can't smile while striding, you're straining -- back off.

Hill touches (low dose, high return): Once per week after an easy day, 6-8x10 s up a gentle hill, walk down fully. Cue: tall + quick + light. These hardwire mechanics without wrecking you.

What the numbers started showing

By week 4, the Z2 drift was visible. Same route, same effort, lower heart rate. Nothing dramatic -- maybe 3-4 bpm at the same pace. But that's the kind of signal that tells you the aerobic system is adapting. The threshold splits were getting more consistent too. Early on, my last rep would be 8-10 seconds slower than my first. By week 6, the spread was under 4 seconds. That's not fitness showing up as speed. That's fitness showing up as repeatability, which is what actually matters in HYROX.

The strides were the most fun part. There's something about touching 5:xx/km pace at the end of an easy run that rewires your brain. You realize the speed exists inside you -- you just can't access it yet when it costs too much. The whole project was about making it cheaper to access.

Brick lite: the cramp insurance

Every two weeks, I'd run a simple brick: 800 m steady, row 500 m at target split, 2:00 easy walk/jog, repeat three times. This teaches breathing reset + bracing so cramp risk drops. After what happened in Mumbai, this became non-negotiable.

Then came race day.

All the engine-building, the threshold reps, the Z2 volume, the cadence work -- it all meant nothing until HYROX Mumbai tested it under real conditions. Eight 1 km runs wrapped around eight stations. Simple on paper; savage in execution.

We crossed in 1:37:23. Not pretty. Precise. Now we know.

HYROX Mumbai: lap by lap

Warm-up and start

Easy jogs, mobility, a few wall-ball sets. Ate light; electrolytes in. Goal: hold ~7:00-7:10/km, arrive at each station breathing but calm. Reality: ~7:11/km average. HR climbed earlier than planned.

Lap 1 -- SkiErg (1,000 m)

Felt okay. Settled too conservatively on the ski to "save legs." Technique was fine; could have pressed 2-3 s/500 m faster without spiking HR.

Lap 2 -- Sled Push

The truth serum. Friction + fatigue lit the quads. Cue drifted (hips behind pad, long steps). Lost time. Economy before load. Short piston steps, hips under pad.

Lap 3 -- Sled Pull

Harness posture inconsistent; too much upper trap. Sit into it, long pulls, quick rebounds, stop shrugging.

Lap 4 -- Burpee Broad Jumps (80 m)

Pleasant surprise. Rhythm clicked. Ranked well here (top-third range). Cadence > jump distance; keep hands/feet pattern identical.

Lap 5 -- Row (1,000 m)

Cramp struck mid-piece. Breathing scrambled, split slipped. Pre-race sodium was fine; this was a run-to-row transition + bracing weakness. Need core + breathing resets before handle-down.

Lap 6 -- Farmers Carry (200 m)

Posture solid, grip okay, pace steady. Carries training transferred perfectly. Keep them.

Lap 7 -- Walking Lunges (100 m)

Form drifted under breathing; knee travel forward, torso pitched. Tempo lunges + reverse lunges in training; preserve shin angle, trunk tall.

Lap 8 -- Wall Balls (100 reps)

Sets broke earlier than planned but no misses. Set pattern (20-20-15-15-10-10-10) and own the rests.

Finish: 1:37:23. Spent, not shattered. Grit intact. Weak links circled in red.

The split board told the full story: run pace averaged ~7:11/km. Strengths were Farmers Carry and Burpees. Limiters were Sled Push/Pull, Row (cramp), and Lunges under fatigue. Every number was a receipt for something the training either built or missed.

What actually went right

What broke (and why)

Each of these failures had a signature. The sleds were a technique failure -- I knew the cues but couldn't hold them under fatigue. The row was a transition failure -- I'd never specifically trained the run-to-row gear change under race conditions. The lunges were a durability failure -- good form in training means nothing if it evaporates at minute 70.

The good news: these are all solvable problems. The bad news: I'd been training around them instead of through them.

The race is a mirror

Race day reflected exactly what the training missed -- and confirmed what carries over.

The engine was real but young. The threshold and Z2 work had moved the needle, but 7:11/km was still too expensive for a race that demands you arrive at every station ready to work. The form cues that held up in training dissolved under race-day breathing. The row cramp wasn't random -- it was a gap in transition fitness that I'd been training around instead of through.

But the carry was clean. The burpees were strong. The pacing was disciplined. Those are the sections where the system did its job.

That's the thing about racing: it doesn't lie. It doesn't round up. It gives you a split sheet and says, "Here's what you actually built."

The plan from here is surgical, not heroic. Run economy keeps climbing (threshold + Z2, every week). Sled technique gets fixed under fatigue, not fresh. Row-to-run transitions get drilled in bricks. Lunges get rebuilt with strict posture before adding distance.

The race proved I could suffer well. The engine proved it could hold together for 97 minutes. Now the job is making it cheaper, cleaner, and harder to break.

1:37:23 is not the finish. It's the first honest measurement.