TRAINING SAFETY

Your judgment outranks the algorithm.

Optional adaptive-energy observations ask you to review suitability and confirm complete, closed food-log days. They are not offered for unsupported ages, medical-caution contexts, pregnancy or breastfeeding. Missing food is not zero intake, and confirming a day cannot verify portions, omissions or the cause of weight changes. Collection counts describe data quantity, not metabolic accuracy or clinical confidence. With enough suitable observations, an experimental model can estimate average daily expenditure over the dated observation period. It uses logged food and a smoothed recorded-weight trend, not measured metabolism or a forecast. Confidence remains limited and assumption based: food-reporting errors and fluid changes can shift the result, including outside the displayed scenario range. That range illustrates selected assumptions, not a statistical confidence interval or guaranteed physical limits. Unstable or out-of-scope observations withhold the number rather than being silently removed or corrected. The result is not an eating target and never feeds training recommendations or rewards. Turning off energy features or hiding calories suppresses it without deleting saved records. This is not medical advice.

Before training

• Use equipment you understand and inspect its setup.
• Choose a conservative starting load when uncertain.
• Use safeties, a spotter or an appropriate machine for difficult sets.
• Tell LiftNext about relevant limitations, but do not treat its response as a diagnosis.

During training

You can edit a warm-up or working target, defer occupied equipment, replace an exercise, or stop it for discomfort. A recommendation is never an instruction to train through pain or unsafe fatigue.

Special contexts

Enhanced/TRT status is optional context and never permission for unsafe volume. Calorie estimates, resting-energy estimates and recovery scores are approximate and are not eating targets, medical tests or proof of readiness. Optional protein ranges and calorie targets are user-chosen tracking settings, not medical advice or verified provider recommendations. Steps and activity never earn back calories or automatically raise or lower food or protein targets. An optional starting step goal can use the average of seven past manual days you confirm as complete, rounded down with no automatic increase. This is a tracking reference you choose, not a health prescription; incomplete days and partial Health Connect readings are not assumed to be complete.

Food and activity estimates

Optional energy review compares seven confirmed closed food days with a separately dated recent expenditure estimate. These averages share food and weight records: their difference is a model comparison, not a measured weekly calorie surplus or deficit or a prediction of fat change. Optional target previews require a separate choice and current body/context review; they never set a Food goal automatically. Adjustment caps, a 1500 kcal feature-scope floor and the current resting-estimate floor are product safeguards, not proof of a safe or adequate intake. Many people need more. Previews are withheld when safeguards or source checks fail, rather than being clipped to a plausible number. Pregnancy, breastfeeding, medical caution, unsupported ages, hidden calories and protein-only modes exclude numeric previews. No result is medical advice, a promised weight-change rate or a reward for eating less.

Food values and barcode matches need your review; missing macros are not zero and provider values are not verified recommendations. Protein tracking can be used without calories. Targets are optional personal tracking choices, not medical advice. LiftNext does not reward eating less, weight loss or burning more calories.

Where Energy estimates settings are available (0.7.2 and later), Mifflin–St Jeor estimates resting energy from explicitly confirmed local inputs; entering body fat selects the lean-mass equation often called Katch–McArdle. These are equation estimates, not measured metabolism or personal accuracy guarantees. They are not valid during pregnancy or breastfeeding and are not medical advice. The method sheet shows the equation and inputs. A resting estimate plus a partial recorded-activity window is an incomplete subtotal: it is neither expenditure so far nor full-day TDEE. Missing activity, unrecorded movement and the thermic effect of food remain unknown. No activity multiplier is applied in this model, and no eating allowance, adaptive expenditure or suggested intake target is derived from this partial subtotal. You can pause energy estimates or hide calories while retaining protein tracking.

Burned calories are estimates, not an allowance to eat back. A device active-energy reading is shown alone, without adding workout or step estimates. Local estimates use confirmed active durations, explicit body weight and conservative activity factors above resting energy; walking distance can use an assumed step length. These factors and scenario ranges are product heuristics, not individual measurements or guaranteed limits. Local numeric estimates are limited to adults19–59; device estimates have their own limitations. Unmeasured workout/step overlap is disclosed and represented by a range, not a precise sum. Planned session time is never substituted for actual active minutes. Older versions can show a whole-day profile activity scenario; it is separate and must not be added to these values. The newer formula model replaces that scenario.

Exercise guides

Written and visual guides are learning aids. If a guide conflicts with your equipment, qualified coaching or safe joint motion, do not follow it—report it from the app.