Garmin Fénix 7^® Underestimates Performance at the Lactate Threshold in Comparison to Standardized Blood Lactate Field Test

Figures & data

Table 1 Corresponding Literature on the Accuracy of Physiological and Biomechanical Estimates of Different Garmin^® Smartwatch Models

Authors	Smart Watch	Variables Tested	Reference Method	Results	Conclusion
Anderson et al 2019^Citation14	Garmin^® Fenix 5x	VO2max	Bruce protocol with spiroergometry	t = 2.21, p = 0.037	Sign. divergent estimate, maybe still an easy and affordable alternative
Wagner et al 2020^Citation10	Garmin^® Fenix 5	VO2max, energy expenditure (EE), stress level	Spiroergometry, calorimetry, electrocardiography	VO2max: CCCLin = 0.50; p = 0.77 EE: CCCLin = 0.43 p = 0.52 Stress level: m-d-Plot; p = 0.89	Garmin fenix 5 estimates not sufficiently concordant
Carrier et al 2020^Citation12	Garmin^® Fenix 3 HR	VO2max, cadence, ground contact time (GCT), vertical oscillation (VO).	Graded exercise testing with spiroergometry, visual 3D	8.05% overestimation of VO2max (R = 0.917). MAPE: VO: 26.38%; cadence, GCT and VO2max < 10%.	Valid VO2max estimate. Cadence and GTC accurate. VO underestimated.
Carrier et al 2021^Citation11	Garmin^® Fenix 6	Speed at LT, HR at LT, speed at OBLA	Graded exercise testing with blood lactate analysis	MAPE: 8.38%, 6.20%, and 10.41%; CCCLin: 0.85, −0.03, and 0.79	Underestimation of LT speed, poor agreement of OBLA and HR at LT

Figure 1 Procedure on Garmin watch face: getting to the guided lactate threshold test.

Figure 2 Study Design.

Table 2 Descriptive of Study Sample

	Age	BMI (kg/m^2)
Total	25.97 (± 6.26)	24.58 (± 2.8)
Female	27.00 (± 8.58)	23.93 (± 3.06)
Male	25.65 (± 5.57)	24.78 (± 2.75)

Table 3 Descriptive Values of the Dependent Variables as Mean (Std)

	Lactate Pace Watch (km/h)	Lactate Pace Dmax (km/h)	Lactate HR Watch (bpm)	Lactate HR Dmax (bpm)
Total	11.87 (± 1.26)	13.28 (± 1.72)	174.42 (± 4.62)	177.42 (± 9.99)
Female	10.38 (± 1.06)	11.75 (± 1.74)	174.17 (± 5.85)	178.50 (± 11.31)
Male	12.32 (± 0.94)	13.73 (± 1.46)	174.50 (± 4.37)	177.10,178.50 (± 9.861)

Table 4 Differences Between the Measurements for All Participants

Sample	Variable (Watch-Dmax)	Median	Mean (STD)
Total	Difference in Pace (km/h)	−1.56	−1.40 (±1.17)
	Difference Pace (%)	−12.06	−11.96 (± 10.60)
	Difference HR (bpm)	−2.00	−3.00 (±8.33)
	Difference HR (%)	−1.14	−1.71 (±4.78)
Female	Difference in Pace (km/h)	−1.29	−1.37 (±1.64)
	Difference Pace (%)	−11.85	−13.58 (±16.62)
	Difference HR (bpm)	−2.50	−4.33 (±10.21)
	Difference HR (%)	−1.39	−2.50 (±5.90
Male	Difference in Pace (km/h)	−1.56	−1.41 (±1.05)
	Difference Pace (%)	−12.06	−11.48 (±8.62)
	Difference HR (bpm)	−2.00	−2.60 (±7.94)
	Difference HR (%)	−1.14	−1.47 (±4.54)

Figure 3 Differences between the LT pace output from the smartwatch and the modified Dmax method.

Notes: Differences between the LT pace output from the smartwatch and the modified Dmax method. Error bars show the SE. * = p < 0.001. Graphs were created using Microsoft Excel (Microsoft, 2019).

Figure 4 Differences between smartwatch LT HR output (bpm) and the modified Dmax method.

Notes: Differences between smartwatch LT HR output (bpm) and the modified Dmax method. Error bars show the SE. Graphs were created using Microsoft Excel (Microsoft, 2019).

Figure 5 The distribution of all data points (pace).

Notes: The distribution of all data points (26 per comparison group and graph) was illustrated using rain-cloud plots.²² Here, we compared individual km/h measured with the Garmin Fénix 7^® (green) and the Dmax Method (orange), not stratified by sex. The graph was created using JASP (JASP Team, 2022).

Figure 6 The distribution of all data points (HR).

Notes: All the 26 data points were included in the analysis. Illustrated by rain cloud plots.²² Here, the comparison of individual HR measured with the Garmin Fénix 7^® (green) and the Dmax Method (orange), not stratified by sex. The graph was created using JASP (JASP Team, 2022).

Figure 7 Bland-Altman Plot for differences in pace (A) and differences in HR (B) plotted with mean values.

Notes: Bland-Altman Plot for differences in pace (A) and differences in HR (B) plotted with mean values. It shows that the deviations between the measurement methods (here: smartwatch and field test with Dmax method) for pace and HR at the lactate threshold are within the level of agreement.

Anderson JC, Chisenall T, Tolbert B, Ruffner J, Whitehead PN, Conners RT. Validating the commercially available Garmin fenix 5x wrist-worn optical sensor for aerobic capacity. Int J Innov Educ Res. 2019;7(1):147–158. doi:10.31686/ijier.vol7.iss1.1293

 Google Scholar

Wagner M, Engel F, Klier K, Klughardt S, Wallner F, Wieczorek A. Zur Reliabilität von Wearable Devices am Beispiel einer Premium Multisport-Smartwatch. German J Exerc Sport Res. 2021;51(1):49–62. doi:10.1007/s12662-020-00682-7

 Web of Science ®Google Scholar

Carrier B, Creer A, Williams LR, et al. Validation of Garmin fenix 3 HR fitness tracker biomechanics and metabolics (VO2max). J Meas Phys Behav. 2020;3(4):331–337. doi:10.1123/jmpb.2019-0066

 Google Scholar

Carrier B, Cruz K, Farmer H, Navalta J. Validation of the lactate threshold estimate from the Garmin fenix 6 fitness tracker. Med Sci Sports Exerc. 2021;53(8S):48. doi:10.1249/01.mss.0000759640.87686.98

 Google Scholar

Allen M, Poggiali D, Whitaker K, Marshall TR, van Langen J, Kievit RA. Raincloud plots: a multi-platform tool for robust data visualization. Wellcome Open Res. 2021;4:63. doi:10.12688/wellcomeopenres.15191.2

 PubMedGoogle Scholar

Data Sharing Statement

All anonymized data are available upon request.