*2.1. Participants*

Twelve regional- and national-level competitive male swimmers specializing in various competitive distances volunteered to participate in the study (see Table 1). Participants had competitive swimming training background of mean (standard deviation: SD) 8.5 (1.7) years, and they participated in a daily swimming training (6 days per week) with duration of approximately 2 h per session. Each participant provided written informed consent after a thorough explanation of the study. The local institutional review board approved the experimental procedures (Approval No. 1007/26-4-2017), which were in accordance with the Declaration of Helsinki for Human Subjects.

#### *2.2. Study Design*

Physiological and biomechanical parameters calculated by a 5 × 200-m progressively increasing speed swimming test were compared with those measured during a constant speed 5 × 400-m intermittent swimming training set in this study. The speed during the training set was prescribed based on the speed vs. lactate concentration curve drawn after a progressively increasing speed swimming test.


**Table 1.** Participant characteristics in the current study. The data are presented as mean values with SD in parentheses.

#### *2.3. Preliminary Testing—The 400-m Test*

The study was conducted during the specific preparation mesocycle of training and swimmers were tested in three testing sessions 48 h apart (Figure 1). All swimming tests were conducted in a 25-m indoor swimming pool with a constant temperature of 25–26 ◦C. During the first visit and following a standardized warm-up (400-m slow swimming at 60% intensity, 4 × 50-m front crawl leg kicking, 4 × 50-m front crawl drills and 4 × 50-m front crawl swim with progressively increasing speed), swimmers participated in a 400-m front crawl test with maximum intensity. Immediately after the completion of the 400-m test, a face mask was applied to the swimmer for expired gas collection during recovery and VO2peak determination (VO2OOO; MedGraphics, Saint Paul, MN, USA; [15]).

**Figure 1.** Study design of the current study; 400-m: 400-m front crawl; 5 × 200-m: five repetitions of 200-m front crawl, 5 × 400-m: five repetitions of 400-m front crawl, h: hours.

#### *2.4. Applying the 5* × *200-m Test*

On the following day, all swimmers participated in a standardized swimming warm-up (following the same procedure as before the 400-m test) and 10 minutes later performed a 5 × 200-m front crawl test at intensities corresponding to 60%, 70%, 80% and 90% of the 200-m maximum speed progressively during the first four repetitions, exerting maximum e ffort in the last 200-m repetition. During the 5 × 200-m test, each repetition started every 5.5 min with a push-o ff start from within the water. Fingertip blood samples were collected after each repetition and were analyzed for blood lactate concentration (BL) using the reflectance photometry enzymatic reaction method (Accutrend Plus; Roche, Germany). Rating of perceived exertion (RPE) was indicated using a 10-point scale after each 200-m repetition [16]. Heart rate (HR) was recorded continuously using telemetry (s610i; Polar Electro, Oy, Kempele, Finland). V4 was determined for each swimmer by interpolation from a second-order polynomial function of swimming speed vs. lactate concentration data (mean (SD) R<sup>2</sup> =

0.97 (0.03), mean r = 0.98 (0.01)). Heart rate corresponding to V4 (HR-V4) was determined for each swimmer individually by the linear regression between swimming speed and HR obtained during the 5 × 200-m test (mean R<sup>2</sup> = 0.96 (0.05), mean r = 0.98 (0.02)). Stroke rate (SR) was calculated by the time (T) to complete three arm-stroke cycles (180·T−1), and stroke length (SL) was calculated by dividing swimming speed every 50 m (V) by SR. SR and SL corresponding to V4 (SR-V4 and SL-V4) were calculated by the interpolation of the best-fit regression line of SR and SL vs. swimming speed during the 5 × 200-m test (SR: mean R<sup>2</sup> = 0.97 (0.03), mean r = 0.99 (0.01); SL: mean R<sup>2</sup> = 0.98 (0.02), mean r = 0.99 (0.01)). Similarly, the rating of perceived exertion (RPE) corresponding to V4 was calculated by interpolation (mean R<sup>2</sup> = 0.97 (0.02), mean r = 0.98 (0.01)).

#### *2.5. Intermittent Swimming Training Set of 5* × *400-m*

Swimmers completed an intermittent swimming training set 48 h after the completion of the 5 × 200-m test. A standardized warm-up as described for the previous testing sessions was applied before the training set. Ten minutes after warm-up, swimmers completed a 5 × 400-m training set at a constant speed corresponding to V4 with a resting interval of 30 to 45 s between repetitions to allow blood sampling. Swimming speed was kept constant by using a sound transmitter attached next to the swimmer's ear (FINIS tempo pro, Finis Inc., Livermore, CA, USA) and according to the individual V4 that was determined by the 5 × 200-m test. Swimmers were advised to touch the swimming pool wall with their legs in each 25-m lap when hearing the transmitted sound. Additionally, one of the researchers recorded the time for each 50-m split in all 5 × 400-m repetitions (HS-80; CASIO, Guangzhou, China), and the mean speed of the test was calculated (V-5×400). BL concentration was collected after the first, third and fifth 400-m repetitions, while HR was recorded continuously and RPE was indicated after each repetition. SR and SL were calculated during each 400-m repetition of the 5 × 400-m swimming training set. The mean values of, BL-5×400, HR-5×400, RPE-5×400, SR-5×400 and SL-5×400 were used for the statistical analysis.

## *2.6. Statistical Analysis*

Student's *t*-test for paired samples was used to compare physiological and biomechanical parameters corresponding to V4 and calculated after the 5 × 200-m test with those measured during the 5 × 400-m intermittent swimming training set with constant speed. Specifically, V4, BL-V4, HR-V4, RPE-V4, SR-V4 and SL-V4 were compared to V-5×400, BL-5×400, HR-5×400, RPE-5×400, SR-5×400 and SL-5×400. Pearson *r* correlations were used to examine the relationship between relevant parameters. Additionally, the effect size for paired comparisons was calculated with Cohen's *d* [17], using the pooled standard deviation as the denominator. The effect size was considered small if the absolute value of Cohen's *d* was less than 0.20, medium if it was between 0.20 and 0.50 and large if it was greater than 0.50. The 95% confidence limits (95% CL) were also calculated for the mean differences between parameters obtained by the two tests. Agreement of measured parameters was tested using Bland and Altman plots [18]. SPSS, software (v.23, SPSS Inc., Chicago, IL, USA) was used for data analysis. Data are presented as mean and SD (in parentheses). Statistical significance was set at *p* < 0.05.
