Nutrition Timing and 3 Adjustment Rules for Prep

Nutrition Timing and 3 Adjustment Rules for Prep

The evidence favors waiting for the signal, then making the smallest effective change; the fastest way to derail a physique plan is to keep “fixing” it before the last fix has had time to work.

Justin Harris’ coaching note is blunt: when a client got backed up after drinking slightly less water on a drive back from Kansas City, the issue wasn’t solved by immediately chasing a new strategy; the underlying pattern was constipation-driven distension, i.e. gut motility slowing when prep gets drier and food load stays high. That mechanism matters because it points to a simple thesis for AI coaching: most nutrition errors in cutting are not solved by faster adjustments, but by better timing of the adjustment and less panic between checkpoints. If you change calories, fiber, fluid, sodium, or meal spacing before you’ve seen the prior change settle, you create noise faster than you create signal.

The core problem: coaches overreact to transient noise

In physique prep, the body does not update on command. Water intake varies day to day. Travel changes routine. Stress changes GI function. Meals can sit differently after one low-fluid day or one lower-fiber day. The result is easy to misread: a flatter look, a heavier scale reading, more abdominal distension, or slower bathroom output gets treated like proof that the plan is failing.

Justin’s note gives the practical heuristic: he saw a big weight drop and a leaner look, then separately observed constipation and distension earlier than usual. His explanation was not “increase everything immediately.” It was that prep makes the gut drier, food backs up, and distension builds. That is a timing problem first, and a macro problem second.

Why AI coaches should slow the feedback loop

AI is very good at spotting short-term deviations. It is less good at knowing whether those deviations matter. If a model reacts to every bad morning weigh-in, every missed bathroom trip, or every puffier midsection by changing the plan, it can become an over-adjustment engine.

That is the failure mode to avoid:

• one low-water day becomes a new hydration protocol
• one heavy meal becomes a new carb target
• one distended morning becomes a new fiber strategy
• one flat look becomes a new refeed

Each of those may be valid changes eventually. The mistake is making them all at once, or making them before the prior variable has had time to stabilize.

The coaching rule should be: change one lever, wait long enough to observe it, then decide whether the result is a real trend or just a transient artifact. In practice, that means AI should treat nutrition changes like controlled experiments, not emergency response.

The mechanism is simple: gut load and water shift together

The useful part of Justin’s comment is not the joke about sensitivity. It is the causal chain. Slightly less water over a weekend led to being backed up for a couple of days. In a prep context, that matters because less fluid often means less comfortable digestion, slower clearance, and more distension. If the athlete is already in a depleted, high-stress state, small deviations are amplified.

That same logic applies to food changes. If you abruptly push fiber up to “fix” digestion, you may worsen abdominal volume before you improve stool output. If you yank fiber down too hard, you may get the opposite problem. If you swing sodium or water in response to a single day’s look, you can create the very puffiness you were trying to avoid.

The mechanism here is not magic. It is simply the interaction of intake, gut transit, and time.

What “timing nutrition changes” should mean in an AI workflow

For coaches using AI, the first job is not to be clever. It is to be disciplined.

A practical workflow looks like this:

1. Separate signal from noise. A single distended morning is not a conclusion. A pattern across several days may be.
2. Respect latency. Nutrition changes often need time before their effect is visible. If you change multiple things in 24 hours, you won’t know which one mattered.
3. Make the smallest effective change. If the issue is likely mild constipation or travel-related distension, a massive plan rewrite is usually worse than a modest correction.
4. Keep the review window stable. Don’t reopen the case every few hours because the mirror or scale changed.

This is where AI coaching can actually be useful: it can remember what changed, when it changed, and whether the athlete has had enough time to show an adapted response. But only if the coach programs restraint into the system.

The final 8-10 weeks are not the time for drama

Justin’s note also mentions planning on adding a serving of dulcolax daily during the final 8-10 weeks of any prep, though he explicitly says he has no evidence beyond experience. That detail is important for a different reason: even experienced coaches are not immune to heuristic bias. The right takeaway is not to copy the intervention blindly; it is to notice the mindset behind it.

The coach saw a predictable late-prep pattern and reached for a repeatable structure. That is good coaching behavior when it is applied cautiously: anticipate the recurring issue, pre-plan a response, and avoid inventing new fixes every time the athlete has a rough week.

The bad version is when the coach uses every rough week as proof that the whole strategy is wrong.

A falsifiable thesis for AI coaching

Here is the testable claim: if your AI coach makes nutrition edits only after a defined observation window, it will make fewer bad corrections than an AI coach that reacts to every daily fluctuation.

That’s a falsifiable thesis because you can measure the output. Compare two systems over a prep block:

• System A changes macros, water, fiber, sodium, or meal timing after one bad day.
• System B waits for a pattern, changes one variable, and holds the line long enough to see the effect.

The second system should produce fewer false alarms, fewer unnecessary reversals, and fewer “I think we fixed it yesterday, why is it worse today?” conversations.

That is the real daily lesson. Nutrition adjustment timing is not a small detail. It is the control system.

What to tell athletes today

Keep the message simple:

• Don’t panic at one off day.
• Don’t stack fixes.
• Don’t confuse travel, hydration changes, and constipation with a failed diet.
• Make the minimum change that fits the evidence.
• Wait long enough to know if it worked.

If AI coaching is going to be worth using in physique prep, it has to help athletes do exactly that: adjust less often, adjust later when appropriate, and adjust with less drama.

Sources Used

• raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w19-24m/clients/david_lamartina___members-tlssnsjthkmnhfqcscszce25acz_vhdm_x2_xdlpx_i.json
• raw/_consumed/2026-05-26/troponin_supplements_kb.md
• raw/Justin_TT1.txt
• raw/_consumed/2026-06-02/_GRAS/gras_strategy_training.md

Sources Used

• /Users/justinharris/TroponinIQ/kb/supertrop/raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w19-24m/clients/david_lamartina___members-tlssnsjthkmnhfqcscszce25acz_vhdm_x2_xdlpx_i.json
• /Users/justinharris/TroponinIQ/kb/supertrop/raw/_consumed/2026-05-26/troponin_supplements_kb.md
• /Users/justinharris/TroponinIQ/kb/supertrop/raw/_consumed/2026-06-02/_GRAS/gras_strategy_training.md
• /Users/justinharris/TroponinIQ/kb/supertrop/raw/Justin_TT1.txt