Performance Condition
Performance Condition is a real-time metric that shows how the body is performing relative to its established baseline during a run or ride. The score appears after six to twenty minutes of activity and updates continuously for the remainder of the session. Each point on the scale represents approximately a 1% deviation from the athlete’s baseline VO2 max estimate.
Developed by Firstbeat Analytics and introduced on the Forerunner 630 in 2015, the feature analyses the relationship between running speed and heart rate, using the athlete’s VO2 max as a reference point. The scale runs from -20 to +20. Garmin’s own documentation uses a +5 score to indicate a rested, fresh state capable of a good run or ride.
The metric depends on the accuracy of the underlying VO2 max estimate. Where that estimate is poorly calibrated — as is common in the first weeks on a new device — the deviation scores carry proportionally greater uncertainty.
What the Number Actually Means
The score is anchored to the individual’s own baseline rather than to any population norm. A +3 for one athlete describes a different absolute performance level than a +3 for another.
Early in a run, the score most often reflects recovery status and external conditions. As the run progresses, a declining score typically indicates the onset of fatigue — the body working harder to maintain its pace than in its fresher state.
Garmin publishes the scale range (-20 to +20) and the +5 example. No official named bands or colour zones are published for Performance Condition specifically.
How Garmin Calculates It
Performance Condition is a Firstbeat Analytics algorithm. It examines the relationship between running speed and heart rate, using the established VO2 max estimate as the baseline for comparison.
The calculation requires six to twenty minutes of recorded activity before the initial score appears. This window allows sufficient heart rate variability data to accumulate alongside pace and heart rate for a statistically stable result.
After the initial score appears, the value updates continuously throughout the activity. The device requires either wrist-based heart rate or a paired chest strap. A chest strap provides more accurate HRV input and generally produces a more stable reading. Several qualifying runs or rides are required before the VO2 max estimate stabilises sufficiently for the Performance Condition to become reliable.
What Affects the Reading
Uphill running or a strong headwind slows pace relative to effort. The algorithm reads this as increased cardiovascular demand at the given speed, and the score temporarily drops. It recovers on descents and in tailwind conditions. Garmin’s technology page explicitly acknowledges this behaviour.
On a bike, external load is measured directly from the power meter rather than inferred from pace and heart rate. A tailwind or gradual descent that would produce an anomalously high score on foot has less influence in cycling, where power output provides an objective load reference.
Heat elevates heart rate, independent of changes in fitness, as cardiac output is partially diverted to thermoregulation. This produces a negative score even in a fit, uninjured athlete. The score in isolation makes no distinction between heat suppression and genuine fatigue.
Sleep deprivation, early-stage illness, high accumulated training load, and significant dehydration all elevate exercise heart rate or suppress HRV. These produce negative scores that frequently precede the athlete’s subjective awareness of limitation — the feature’s primary practical value.
GPS quality affects pace accuracy, particularly in urban canyons or under heavy tree cover. When pace data is unreliable, the speed-to-heart-rate relationship the algorithm relies on is compromised. Multi-band GPS devices are less susceptible to this source of error.
Profile accuracy matters. The user profile fields — age, weight, and sex — contribute to the VO2 max estimate that forms the baseline. Outdated or inaccurate profile data introduces a systematic error that propagates into Performance Condition scores.
Wrist-based optical sensors introduce more noise into the HRV input than chest straps. Accuracy further degrades in cold weather, when peripheral blood flow is reduced.
How Accurate Is It
No published peer-reviewed study has directly validated Performance Condition scores against laboratory-measured physiological performance deviation. The algorithm is proprietary and has not been published in a form that permits independent replication.
Accuracy must be assessed indirectly through research on the baseline VO2 max, since the feature depends on it. Pearson et al. (2018) found mean VO2 max estimation errors averaging 7.6 ml/kg/min in high-fitness participants using GPS watches—a margin that translates into deviation scores for trained athletes. Anderson et al. (2019) examined the Garmin Fenix 5X and reported a mean absolute error of 2.16 ml/kg/min under controlled conditions, suggesting more recent hardware achieves smaller baseline errors.
The feature is more reliable as a directional indicator within a workout than as a precise measurement.
A pattern of negative scores across multiple sessions, as accumulated fatigue builds, carries actionable information even when the precise number is uncertain. Garmin acknowledges that results stabilise after the first few runs as the device accumulates training history.
Competitor Equivalents
There are no direct competitor comparisons. These share similarities.
- Polar: Running Index provides a post-run efficiency score calculated from speed and heart rate. It is produced after the activity ends and benchmarked against population norms rather than the individual’s own baseline. There is no real-time in-workout equivalent.
- Apple Watch: Cardio Fitness provides an estimated VO2 max and workout heart rate data. Apple Watch produces no real-time deviation score during a run.
- Coros: EvoLab provides post-session analysis of training load and fatigue. Coros produces no real-time within-activity performance deviation score of this type.
- Suunto: Recovery and readiness indicators are available, but Suunto produces no real-time within-activity performance deviation metric equivalent to Performance Condition.
- Wahoo: The ELEMNT platform produces no equivalent feature. Training load and recovery metrics are available via third-party platform integration, but are calculated externally.
Which Garmin Devices Support It
The Performance Condition feature was introduced in 2015 on the Forerunner 630. It has since been standard on all mid-range and flagship Garmin running and multisport devices.
Current support covers the Forerunner 265 series and later models in the same tier; the Forerunner 570 series and later; the Forerunner 965 series and later; the Forerunner 970; the Fenix 7 series and later; the Fenix 8 series (AMOLED and Solar, 47mm and 51mm, including Fenix E); the Epix Pro (Gen 2) series; the Enduro 3; the MARQ (Gen 2) series; and the Tactix 8 series. On Garmin Edge cycling computers — including the Edge 540, Edge 840, and Edge 1050 series — the feature incorporates power meter data where a meter is paired.
The feature is absent on entry-level devices, including the Forerunner 55, Forerunner 165, Venu series, and Vívoactive series. These devices support VO2 max estimation but do not surface the real-time deviation calculation.
Where to Find It
Performance Condition appears as an automatic alert on the watch approximately six to twenty minutes into a recorded run or ride. The device displays the score at that point without requiring any action from the athlete.
After the initial alert, the metric is available as a data field added to any training screen via the device’s activity data screen customisation settings.
Performance Condition is an in-activity value only. It is absent as a standalone widget, widget glance, watch face complication, and Morning Report component. The score does not persist after the activity ends.
In Garmin Connect mobile, the session’s score appears within the activity detail view after syncing, in the performance metrics section of the activity summary. Aggregated trend data across multiple activities is absent.
On Garmin Connect, Performance Condition appears in the activity detail view. A dedicated aggregated dashboard is absent. No Garmin Connect Plus subscription is required to view the data.
The mid-activity notification can be disabled on the device by going to My Stats > Performance Notifications > Performance Condition.
Common Problems and Misreadings
The most common source of confusion is a score that contradicts perceived effort. The metric measures cardiovascular cost, and subjective sensation and physiological output routinely diverge — particularly in early-stage illness, mild dehydration, or accumulated fatigue. Early in a run, the score most often reflects recovery status rather than fitness level. A negative score at the ten-minute mark is more often a readiness signal than a performance problem.
Sustained high scores — persistently positive readings across multiple sessions — are frequently a sign that the VO2 max estimate is set too low. When the baseline is underestimated, the algorithm reads normal efforts as above-expectation performance. Sustained high scores typically precede an upward revision of the VO2 max estimate.
A score that rises continuously throughout an activity with no obvious cause may indicate bad sensor input data. Where the score behaves in a way that terrain, conditions, or fatigue cannot explain, check that the heart rate sensor is making clean contact and that GPS reception was adequate throughout the session.
Scores that fluctuate substantially during a run are most often explained by terrain and wind. Uphill sections cause the score to drop; descents or tailwind sections cause it to recover. Runners who continuously monitor the data field often overinterpret this normal variation. The stable reading produced after the initial calculation window carries more interpretive weight than the subsequent fluctuations.
New device users frequently encounter unreliable scores during the first two to four weeks. The VO2 max estimate must stabilise before the deviation calculation becomes meaningful. Garmin’s own documentation explicitly acknowledges this settling period.
How to Improve It
Performance Condition is a relative indicator. An athlete who improves fitness substantially will see scores that continue to centre around zero on normal training days, because the baseline rises in step with fitness. The goal is a pattern in which positive scores reliably appear before targeted hard efforts and races — not a permanently elevated score.
The most direct lever for consistently positive scores is managing cumulative fatigue. Training hard through accumulated fatigue produces chronically suppressed scores, regardless of fitness level, because the cardiovascular cost remains elevated relative to the adjusted baseline. Structuring training with adequate recovery between hard sessions, and treating negative scores as early indicators rather than anomalies to override, allows the pattern to become a useful planning signal.
Sleep quality has a pronounced and measurable effect on next-day scores. A single night of significantly shortened sleep is sufficient to elevate resting heart rate and suppress HRV, producing a negative reading in a well-trained athlete on a low-load day. Consistent sleep duration and quality are among the most direct practical influences on the pattern of scores observed across a training block.
Frequently Asked Questions
- Why does Performance Condition show negative when training feels fine? The metric measures cardiovascular cost against the athlete’s established baseline, not subjective sensation. Early-stage illness, mild dehydration, residual fatigue from a recent hard session, or alcohol intake in the preceding twenty-four hours can each elevate heart rate or suppress HRV sufficiently to produce a negative score before any conscious limitation is registered. Where the pattern persists across several days without an obvious explanation, check sleep quality and accumulated training load over the preceding week.
- How long does it take for the score to appear? The device requires six to twenty minutes of activity data before displaying the initial score. The window varies depending on how quickly the algorithm reaches a statistically stable reading. A chest strap rather than a wrist-based heart rate monitor typically shortens the calibration window and yields more stable results.
- Does Performance Condition appear as a trend in Garmin Connect? The score is recorded within each activity file and is visible in the activity detail view in both Garmin Connect mobile and Garmin Connect web. A dedicated trend chart plotting scores across multiple sessions is absent. Runners who want to track the pattern over time should review individual activity records or export data for analysis in a third-party platform.
- Why does the score drop on hilly routes? Uphill running increases the cardiovascular demand at a given pace, which the algorithm interprets as elevated effort relative to the input pace. The score recovers on descents. Runners whose baseline was established primarily on flat terrain will see more pronounced fluctuation on hilly routes than those whose baseline reflects mixed terrain.
- Can Performance Condition replace Training Readiness as a pre-run assessment? Training Readiness is a pre-activity assessment calculated before the run begins, using overnight HRV, sleep score, Body Battery, recovery time, and stress history. Performance Condition appears only after six to twenty minutes of activity. The two metrics are complementary: Training Readiness informs the decision about what to attempt; Performance Condition informs whether to adjust targets once the session is underway.
- Does Performance Condition work on the treadmill? Support varies by device and activity profile. Treadmill runs using wrist-based heart rate and internal accelerometer-based pace estimation can produce a score. Still, accuracy is lower than for outdoor GPS activities where pace is measured directly. Pairing a foot pod for more precise pace input improves the quality of treadmill readings.
Scientific Basis
Firstbeat Technologies. Sports Technology White Paper: VO2max and Performance Condition. Firstbeat Analytics, 2014 (updated 2017). Describes the Firstbeat model for real-time aerobic capacity estimation from heart rate and HRV data during exercise, and the derivation of the Performance Condition deviation score from a VO2 max baseline. This is the primary technical precursor document for the algorithm underlying the feature.
Engel FA, Masur L, Sperlich B, Düking P. Validity of V̇O2max estimates from the forerunner 245 smartwatch in highly vs moderately trained endurance athletes. Eur J Appl Physiol. 2026 Jan;126(1):591-603. doi: 10.1007/s00421-025-05923-x. Epub 2025 Aug 6. PMID: 40770433; PMCID: PMC12881131.
Anderson, James & Chisenall, Trent & Tolbert, Blake & Ruffner, Justin & Whitehead, Paul & Conners, Ryan. (2019). Validating the Commercially Available Garmin Fenix 5x Wrist-Worn Optical Sensor for Aerobic Capacity. International Journal for Innovation Education and Research. 7. 147-158. 10.31686/ijier.Vol7.Iss1.1293.
Buchheit, M. “Monitoring training status with HR measures: Do all roads lead to Rome?” Frontiers in Physiology, 2014. Establishes the physiological basis for using altered heart rate and HRV responses to a given workload as indicators of current performance capacity — the conceptual foundation on which the Performance Condition metric rests.
How It Connects to Other Features
Performance Condition is downstream of VO2 Max. The VO2 max estimate provides the baseline against which all deviation scores are calculated. Any factor that causes an abrupt shift in the VO2 max estimate — illness, a sudden jump in training volume, or a device change — temporarily destabilises Performance Condition readings until a new baseline is established. The Forerunner 620 (2013) was the first Garmin device to estimate VO2 max; Performance Condition followed on the Forerunner 630 (2015), requiring an established VO2 max before it could function.
The metric complements Training Readiness. Training Readiness is a pre-activity prediction; Performance Condition is a real-time observation during the activity. The two signals align directionally on most days and diverge when an unexpected stressor is absent from the pre-activity estimate.
A persistent pattern of negative scores across a training block is a signal that recovery between sessions is insufficient and, eventually, manifests in Training Status as Unproductive or Overreaching. Reviewing the two features together gives a more complete picture than either provides alone.
On cycling-capable devices and Garmin Edge computers, Performance Condition incorporates power output alongside heart rate where a power meter is paired, using the power-to-heart-rate relationship rather than pace as the primary effort reference.
