Recovery Signal Quality: 3 Ways to Decide Whether to Feed, Train, or Wait

Recovery Signal Quality: 3 Ways to Decide Whether to Feed, Train, or Wait

AI coaching is only useful when it tells you whether fatigue is a nutrition problem, a training problem, or a patience problem. The signal gets clearer when you stop treating appetite, blood sugar, and recovery as the same thing.

The clearest signal in the Kahunas coaching logs is blunt: Justin Harris told a client that retatrutide at 2 mg was “definitely lowering my appetite” and that he had been “a bit more fatigued than normal” within days, while also warning that if food intake needed to rise, he might pause or reduce the dose. That is a mechanism-level clue, not a vibe: appetite suppression can change intake fast, and the first thing that shows up may be fatigue, not a dramatic body-comp swing. The practical thesis is simple and falsifiable: when recovery worsens, AI coaching should first ask whether the next lever is food, training, or time, because appetite-shifting interventions can make fatigue look like under-recovery even when the real problem is intake mismatch.

Recovery is not one signal

Most coaching software wants to collapse recovery into a single score. That is attractive and often wrong. Recovery is a bundle of at least three separable questions:

1. Did the athlete actually eat enough to support the current workload?
2. Did the current workload exceed what the athlete can presently recover from?
3. Is the problem just lag — the body has not had enough time to adapt yet?

AI is useful only if it helps separate those. Otherwise it turns every complaint into either “push harder” or “rest more,” which is how coaches miss the real fix.

The strongest example from the source material is nutritional, not mystical. Justin’s off-season framing is that the body should learn to digest and assimilate a massive amount of clean food, because the more good bodybuilding food someone can handle, the better the chance of muscle growth and the better the metabolism for contest prep. In plain English: if food tolerance rises, downstream recovery and output improve because intake can keep up with demand. That is why a sudden appetite drop matters. If the athlete is no longer willing or able to eat the planned calories, performance can fall before the scale or the mirror shows anything useful.

When the next change is nutrition

The Rory Lazowski exchange gives the cleanest coaching pattern. Justin noted that retatrutide was lowering appetite and making him more fatigued than normal. He did not jump straight to a training overhaul. He looked at the direction of the next phase — adding food — and said he might pause or reduce the dose if calories were going up. That sequence matters.

If appetite is down, recovery signal quality is degraded because the athlete’s report of “I feel flat” may really mean “I’m under-eating relative to the plan.” In that case, the next move is not more coaching intensity; it is nutrition reconciliation. The question becomes:

• Has food intake actually fallen below the intended target?
• Has meal timing drifted so far that training quality is getting hit?
• Did the intervention intended to help body comp also create a fatigue signal that clouds everything else?

This is where AI can be practical. A good system should flag a mismatch between planned intake and realized intake before it flags a mysterious recovery crash. If appetite is suppressed, the most likely fix is not a new supplement, not a new “recovery routine,” and not a dramatic reduction in training volume. It is usually: restore intake, simplify food choices, or reduce appetite-suppressing inputs if they are no longer serving the phase.

The food-volume point from Justin’s off-season language is useful here too. He emphasizes being able to digest and assimilate more clean food over time. That is a long-run adaptation goal. But on a day-to-day basis, the coach still has to answer the short-run question: did the athlete get enough food today to recover from today’s work? Those are not the same problem.

When the next change is training

Not every fatigue report is a nutrition problem. Joe Webb’s insulin note is a good reminder that systems can drift even when the athlete is trying to execute correctly. He reported improved insulin sensitivity on a high day: the same insulin dose as the prior week brought blood sugar down more than expected, forcing meal timing to move earlier. Justin’s response was operational, not dramatic: bring the meals closer together and adjust the dose downward on the next high day.

Why does this matter for recovery? Because training and nutrition are linked by timing. If meal timing shifts to correct a blood sugar issue, training quality may change even if total calories stay the same. In that case, the signal is not “the athlete is overreached” and not “the athlete needs more food overall.” The signal is “the execution schedule has changed.”

AI coaching should catch this distinction. A client can feel worse because:

• meal spacing is off,
• pre-training fuel timing is off,
• insulin or carb handling changed,
• or the session itself is simply too expensive for the current state.

If the issue is training dose, the right move is usually a workload change: reduce volume, trim intensity, or adjust exercise selection. But you do that only after you’ve ruled out a nutrition/timing error. Otherwise you end up shrinking the training plan to solve a fuel problem.

That is the key practical hierarchy:

1. Fix food availability and timing.
2. Then adjust training cost.
3. Only then decide whether you need patience.

When the next change is patience

Patience is the most underused intervention in AI coaching because it is hard to quantify and easy to market around. But the sources include a very human version of the problem: Justin says he is always wary that clients will think he is soft-pedaling issues, yet he also repeatedly distinguishes expected adaptation from meaningful trouble. That distinction is the heart of recovery signal quality.

If a client has just changed an intervention — less appetite, different carb timing, new loading pattern, more food, less food — the first several days can be noisy. A single bad session or one off day is not enough to rewrite the plan. Sometimes the best decision is to hold the line and let the system settle, especially when the athlete is not showing a clear performance collapse or a persistent intake failure.

Patience is not passivity. It is a rule: do not overreact to a signal until you know what domain generated it. If the athlete is eating what was prescribed, training is stable, and the fatigue is mild and transient, the right move may be time. If the athlete is missing meals, appetite is down, or blood sugar handling is altered, the right move is nutrition. If food is stable but workload is too costly, the right move is training.

What good AI should actually say

A useful recovery model does not hand out generic advice like “recover more.” It asks sharper questions:

• Did appetite or satiety change after a new intervention?
• Did meal timing drift enough to change training quality?
• Is fatigue proportional to workload, or is it out of proportion to what changed?
• Is this a transient adaptation signal or a real mismatch?

The best systems will not pretend those are the same. They will prioritize the simplest explanation that fits the data.

That is what makes the recovery signal high quality: the coach can identify whether the athlete needs more food, less training, or more time, and can do it without turning every fatigue report into a mystery. In the examples above, the answers were concrete. Retatrutide lowered appetite and increased fatigue; the likely next move, if calories were rising, was to reduce or pause it. Improved insulin sensitivity changed meal spacing; the fix was to adjust the dose and timing. Off-season food tolerance is a long-term adaptation; patience matters when the body is still building that capacity. Recovery coaching is not about one score. It is about choosing the next lever correctly.

Sources Used

• raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w13-18m/clients/rory_lazowski___members-c5balaovjbdoeefqmfuqdhh2tbpmfdu16lnf0tnrtmw.json
• raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w13-18m/transcripts/rory_lazowski___members-c5balaovjbdoeefqmfuqdhh2tbpmfdu16lnf0tnrtmw.md
• raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w19-24m/clients/rory_lazowski___members-c5balaovjbdoeefqmfuqdhh2tbpmfdu16lnf0tnrtmw.json
• raw/Justin_TT1.txt
• modules/03-knowledge/kahunas-coaching-deep-nutrition.md