Insulin 1 IU: Timing Nutrition Changes Without Chasing Noise

Why the best adjustment is often the next one, not the next three

The clearest coaching note in the corpus is Joe Webb’s: the same insulin dose that worked one week started dipping blood sugar the next, so he simply brought meal 2 forward by 30 minutes and then reduced the dose by 1 IU on subsequent high days. That is the mechanism in one phrase: changing nutrient handling faster than your plan changes around it. The falsifiable thesis is simple: when appetite, insulin sensitivity, or food volume shifts, the winning move is a single small nutrition adjustment with enough time to show effect, not a cascade of reactive edits that blur cause and outcome.

In practice, that means timing matters as much as the change itself. Joe did not rewrite the whole day because one meal landed differently. He noticed a repeatable pattern, moved the next meal closer, and only then trimmed the dose further on the following high day. That sequence matters because it preserves signal. If you change food, timing, and dose all at once, you lose the ability to tell whether the improvement came from the first tweak, the second, or just a random swing in the week.

That same logic shows up in Justin’s broader nutrition coaching. In the deep nutrition cases, he treats the big rocks as macro compliance first, then meal-by-meal details only where they actually matter. His fruit guidance is a good example: pre- and post-workout, fruit can make up about 50% of the carbs in those meals with no downside; on medium days it is fine in other meals too as long as it is not every meal and at least half the carbs per meal come from more complex sources; on high days, when carbs are very high and insulin is elevated all day, fruit matters even less because it helps keep food volume down. The lesson is not “fruit is magic” or “fruit is dangerous.” The lesson is that the same food can be a non-issue or a useful tool depending on where it sits in the day’s structure.

That is the right frame for AI coaching, because AI is very good at generating options and very bad at respecting lag. A model will happily propose three carb swaps, a new meal schedule, and a supplement tweak after one check-in. The human error is to treat every suggestion as equally urgent. The coaching error is worse: to make every change immediately and then call the resulting mess “responsive programming.” Responsive programming is not the same as over-adjustment.

The anti-hype version of coaching is boring in a useful way. First, identify the variable that changed. Second, make one edit that addresses it. Third, wait long enough for that edit to express itself. Joe’s log shows the sequence cleanly: improved sensitivity at a high day led to a meal timing change and a smaller insulin dose, not a wholesale rewrite of the diet. Justin’s response in the same thread was equally pragmatic: because the athlete had become leaner, he wanted to run with the current setup and continue leaning out while it was easy, then use that state as a cleaner example of how the approach behaves during a mild rebound later. That is periodization logic, not improvisation.

This is where nutrition timing earns its keep. People often want to “fix” a problem by shifting calories immediately, but many issues are really timing issues wearing a calorie costume. A meal that needs to come 30 minutes sooner is not evidence that the whole intake level is wrong. A day where fruit improves adherence is not evidence that fruit should displace all slower carbs. A week where appetite is lower is not proof that you need a new diet; it may just mean the current intake now requires less forcing. If you keep changing the whole plan every time one variable moves, you make the plan untestable.

The useful rule for coaches is to respect the delay between cause and effect. Let the first adjustment have a chance to work. If the problem is acute and concrete, fix the acute and concrete thing: move the meal, trim the dose, reduce the food volume, change the carb source in the meal that is actually causing friction. If the problem is not acute, do less. A stable plan with one measurable change beats a clever plan with five moving parts.

AI can help here if it is used as a constraint checker rather than an adrenaline source. It can flag that a repeated low at the same meal means timing needs attention. It can remind you that fruit on high days is a volume tool, not a moral event. It can organize the log so you see what changed first. But it should not be the thing that keeps asking for another modification before the last one has had time to show up. The more uncertain the signal, the smaller the change should be.

That principle generalizes well beyond the specific insulin example. In almost any physique or performance setup, there is a temptation to interpret every small fluctuation as a mandate to act. The better pattern is to ask: did the system actually change, or did the data just wiggle? If the answer is “the system changed,” make one adjustment and hold it. If the answer is “the data wiggled,” stay put. Over-adjustment usually looks smart on day one and sloppy by week three.

The strongest coaching edge, then, is not faster reaction. It is better timing of reaction. Give the change time to declare itself. Keep the next adjustment small enough that you can still learn from it. And do not confuse activity with precision.

Sources Used

modules/03-knowledge/kahunas-coaching-deep-nutrition.md
raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w13-18m/clients/joe_webb___members-rksigkykimaxwmo_t4_e8nwvbtc2j0etleutkyysads.json
raw/_consumed/2026-05-31/kahunas-export/2026-05-31-w13-18m/transcripts/joe_webb___members-rksigkykimaxwmo_t4_e8nwvbtc2j0etleutkyysads.md