Trio Starting Point Estimator

{prob.questions[qIdx].text}

{item.setting}

The Mathematical Cascade

Every estimate in this tool flows from a single starting input — hypothetical body weight — through a chain of standard clinical formulas. Each step depends on the one before it:

Two Categories of Settings

The estimator distinguishes between settings that can be reasonably estimated from weight-based formulas and those that require real-world outcome data. This distinction is fundamental and is clearly marked throughout the tool:

Estimated settings (Basal Rate, ICR, ISF, Max IOB, Threshold) are derived from the mathematical cascade. They provide a legitimate starting line based on the same population-level formulas endocrinologists use, but they are not personalized and must be verified through testing.

Outcome-based settings (Adjustment Factor tuning, Dynamic CR, Adjust Basal, Weight Percentage, Sigmoid) cannot be derived from weight alone because they describe how your specific physiology responds to insulin at different blood glucose levels, across different days, and under different conditions. These require 1–4 weeks of observed data.

The Connected Troubleshooter

The Pattern Troubleshooter is connected to the Starting Estimates. When you adjust your weight, factor, or other inputs, the troubleshooter's recommendations update to reference your specific estimated values — showing actual calculated numbers rather than generic ranges. It can tell you that your estimated max SMB of 0.39U is likely too small for a 60-gram meal, not just that "Max SMB Minutes might need to be raised." This connection makes the troubleshooter's guidance concrete and actionable.

Source Attribution

Every claim in this estimator is tagged with its source. Source Code labels indicate information derived from Trio's actual algorithm at github.com/nightscout/Trio — the real formulas, decision trees, and constants the software uses. Trio Docs labels indicate information from the official Trio documentation at triodocs.org. Clinical labels indicate established diabetes management and pharmacological knowledge. Strategy labels indicate patterns observed in real-world community usage. These tags exist so you can evaluate each piece of information on its own merits and verify it independently.

Why These Are Estimates — Not Prescriptions

The Rules of 500 and 1800 are derived from population averages. They are standard in clinical practice as starting points, but human biology is highly variable. Insulin resistance changes with time of day, exercise, stress, illness, hormonal cycles, and dozens of other factors this estimator cannot account for. The 50/50 basal-to-bolus split is a simplification — many people require a different ratio. The flat hourly basal rate this estimator produces does not reflect the time-of-day variation most people need.

These estimates are a starting line, not a finish line. They must be verified through methodical testing — basal testing during fasting periods, ICR testing with well-counted meals, and ISF testing with correction doses — before they can be trusted for real-world use.

How Estimated Values Connect to Trio's Algorithm

The three core settings this estimator produces — Basal Rate, ICR, and ISF — are the exact values Trio asks for during onboarding. They become your profile settings, and the algorithm's dynamic features build directly on top of them. Dynamic ISF divides your profile ISF by a sensitivity ratio. Dynamic CR divides your profile CR by the same ratio. Adjust Basal multiplies your scheduled basal by a TDD trend ratio. If the profile values are inaccurate, every dynamic adjustment amplifies the inaccuracy. Getting these starting values close is therefore essential to good algorithmic performance.