THE SWIMMING POWER. NEW METHOD TO TRANSFER THE POWER FROM DRYLAND TO THE WATER

129 Резюме: Вивчено вплив метода силового тренування на потужність плавання 20 спортсменів-ветеранів, яких було умовно розподілено на дві групи – силової (n = 10, ST) і плавальної (n = 10, SW) підготовки. Тренувальні заняття проводилися упродовж 6 тиж. і включали в групі SW плавальну підготовку та силову з подальшим плаванням з максимальною швидкістю. Результати в обох групах оцінювали на основі максимальної–механічної–зовнішньої потужності (ММЕР), застосуванням ергометра для вимірювання сили, швидкості і потужності в воді. В групі ST спостерігали значне підвищення ММЕР (5,79 %; р < 0,05) разом із збільшенням сили (11,70 %; p < 0,05) і зниженням швидкості (4,99 %; р < 0,05). В групі SW виявлено зниження ММЕР, сили і швидкості (7,31, 4,16 % і 4,45 %; р < 0,05). Дослідження показало, що метод, заснований на поєднанні силового тренування (на суші)з подальшим швидким плаванням, істотно збільшує потужність плавання у спортсменів-ветеранів. Ключові слова: екологічна валідність, тестування в польових умовах, тест потужності в воді, результат, силове тренування.

Introduction. The metabolic demands of the swimming competitions is very different, indeed aer obic and anaerobic systems [1] are related to the race time (from 20 sec for 25m to 900 sec for 1500m) Nevertheless, the performance of swimmers was con tinuously improved due the enhancement of tech nique [2], the evolution of the facilities [3] and the improvements of the physical skill of the ath letes [2]. Swimming action recruits many muscles for propulsion, mechanical power, and for drag con trast [4]. Therefore, the muscle strength plays a crucial role to increase the swim velocity [5]. Al though some authors [6,7] have shown that the ad justments related to technical movements performed in «dry conditions» using overloads, may be use ful to improve the technique of the swimmers in the water, this was not confirmed by field swimmer's coaches. Currently, two methods are mainly used for strength training purposes in swimmers: «drymet hods», namely with session out of water composed by exercises with loads of general type [8][9][10], or by «simulating» the swimming movements [11]. The simulation approach was carried out with «aquatic methods» training session, when the swim is over loaded with tethered [12] or tools that increase the dragging force [4].
However, it is not yet entirely clear on the actual effectiveness of these methods [ 1] , as it appears dif ficult to increase the strength as the power of swim mers through «aquaticmethods» into load session [13,14]. Similarly, an increasing strength method obtained with «drymethods» showed some limits on the «transferability» on specific technical swim movements [9,15]. Recently, several inwater meth ods [5,16] were used to assess the strength and the power of the swimmers through the assessment of the drag, providing conflicting results [5,16,17]. The strength and power estimates from swimming velocity doesn't seem adequate [ 8,18,19] because the swimming velocity was related to muscle power, and both propulsion efficiency and drag coefficient of swimmer [5]. In rare cases the use of tethered test has been reported with some limitations (the swim mer cannot effectively advance in water, and thus the technical gestures are altered).
Among the different methods of training alternat ing dry weights and swimming we chose the method proposed by Prof. Cometti [20]. Despite never hav ing been studied with scientific rigor its principles are clear. This innovative method aimed to improve the swimmer performance, using an approach poten tially valid even in other disciplines (such as Team Sport). Through the Cometti method [21] stimulate the muscle fibers using a weight of about 80 % of a maximum repetition immediately followed by the execution of technical movements specific of each discipline (in our case swimming [20,21]). The goal would be to stimulate the muscle fiber with an over load in water, that is impossible to reproduce be cause of the lack of «stable points of resistance». Therefore the aim of this study was to verify a Com etti training method based on mixed «dryland phase with overloads with a series of fast swimming» on the swimming power with a specific semitethered swimming test.

Participants
Twenty senior male master swimmers belonging to the same team were recruited for the study and randomly assigned to either the strength training (ST, n = 10) or swimming training (SW, n = 10) groups. Their main anthropometric data, as well as their best performances on 100 meter crawl. In or der to be included in the study, participants had to: 1) participate in at least 90 % of the training sessions (see following chapter about training pro gram), 2) have regularly competed during the pre vious competitive season, and 3) possess a medical clearance. There were no dropouts from the exper iments and no injuries occurred during the experi mental training or testing sessions. Indoor field tests were completed in a certified swimming pool. Base line tests started at 5:00 p.m. (26,5 ± 0,12°C, water temperature), while postassessments were carried out at 5:00 p.m. (26 ± 0,16°C, water temperature). The participants were healthy and clear of any drug consumption. The groups were homogeneous with regard to their training status (more 10 years back ground competitions). Each subject was fully in formed and trained about the test's procedures and everyone gave the written informed approval to par ticipate in the study in accordance with the guide line of the Muscle, Ligament and Tendons Journal [22]. All experimental procedures were approved by the University Human Research Ethics Committee, which followed the ethical principles laid out in the 2008 revision of the Declaration of Helsinki.
Testing A parallel, twogroup, randomized, longitudinal (pretest/posttest), singleblind experimental de sign was used. After baseline measurements, partici pants were randomly allocated to either the strength training (ST) or swimming training (SW) groups with an allocation ratio of oneto one [23]. The in dependent variable was «training type», so no con trol group was used. The study lasted 6 weeks (from September to November in preseason) and consisted of one session of test (preand posttraining) before and after one week training sessions. No addition al strength, power and/or plyometric training was completed by the subjects out of the training inter vention of the present study.
Training outcomes Before and after (testretest) the training period, participants performed one testing session of semi tetheredswimming to assessment Maximal Mechan ical External Power (MMEP). Before each testing session, participants were instructed not to eat for at least three hours before testing and not to drink cof fee or beverages containing caffeine for at least eight hours before physical testing. Tests were completed at the same time of the day, with the operators una ware of the participant's allocation.

Maximal Mechanical External Power Test
The test consisted in 15 m allout frontcrawl swims across the pool while pulling a different load during each trial, besides the reliability of the test has been shown in previous studies to be very high (Intraclass Correlation Coefficient > 0,80) as shown by DominguezCastells et al. [ 24] After a standardized 800 m warmup, the test started with a load of 45 N. The load increased by 25 N for each tri al. Swimmers rested for 5 min between 2 consecutive 5 repetitions. The protocol ended when the swimmer was not able to complete a trial. Data related to the first and last 2,5 m was discarded to consider only constant speed conditions [ 24] . The MMEP param eters of interest were acquired by means of a dedi cated custom ergometer designed and built by Tec nologicamente S.r.l. (Italy) with the collaboration of the workshop of the Department of Mechanical, Chemical and Materials Engineering of the Univer sity of Cagliari (Italy). The ergometer used for the experimental sessions was linked to the swimmer us ing a belt as described in the following.
This device is basically composed by a 28» wheel (acting as a drum with a winding circumference of 2092 mm), a cable, two sensors (force and speed) and an electronic apparatus necessary to proper ly conditions and transmits the data to a Personal Computer. The wheel is equipped with a disc brake (Shimano disc 160 mm diameter and Hayes Nine brake caliper) and a reflective encoder wheel with 72 pulses per turn read by an optical speed sensor (Optek OPB704). A 500 N miniature tensioncom pression load cell (F2220, Tecsis GmbH, Germany) was hosted inside an aluminum cylindrical (160 mm long, 47 mm diameter with a nose cone to mini mize the hydrodynamic resistance effects) that act as waterproof case and was connected to the swimmer through a belt equipped with a system composed by a light aluminum bar and four twines. The load cell signal is conditioned and powered by a Mecos train 2038 module embedded in the cylindrical alu minum case.
Prior to the tests, a calibration curve hydraulic pressure vs. resistant force was obtained using cali brated weights (corresponding to a 10 < 150 N force range). Both force transducer and speed sensor sig nals were properly acquired by a National Instru ments DAQ Module USB 6009 (8 channels, 14Bit, 48 kS/s). A custom routine was developed in the National Instruments LabView ® environment to col lect and store data in form of ASCII files during the trial. The resulting files were then postprocessed with a Matlab TM 10 software routine that transforms the raw data into a fourvectors text file contain ing time, traveled distance, instantaneous force, and speed values.
Training program The training program was performed during six weeks, divided in three sessions for both groups in according to Cometti method [ 21] . All participants (ST-SW) after 15 minutes of standardized warm up carried out the same set of exercises in water, that comprises several sprints in front crawl at maxi mum velocity with sets and recovery balanced. Each swimming session had a duration of approximately 2 hours and was repeated 5 times per week. During the swimming training the same distance was per formed for both groups (ST-SW).
Particularly, ST group were performed as suggest ed by Cometti [20,21], the strength training program during swimming training (mixed: weight trainingswim maximum velocity and vice versa). The one repetition maximum (1RM) test on bench press was conducted to determine maximal upper body strength as recommended by Padulo et al [25]one week before the training. Particularly, du ring the exercise with weigh (85 % 1RM [ 6] ) or body load, subjects were asked to perform 6 fast repetitions [6] according to Cometti method [20,21]. To minimize the effect of the passive recovery [20,21] inbetween weight train ing and swimming exercises (~5s), each participant was encouraged by the coach.
Statistical analyses Normality of the data was verified using the Sha piroWilk test. The null hypothesis was tested to re veal no difference between groups using multiple un paired ttests. A twoway mixed analysis of variance (ANOVA) was used on each continuous dependent variable. The independent variables included one bet weenparticipants factor, training intervention with two levels (ST and SW), and one withinpar ticipant factor time, with two levels (pretest and posttest). ANOVAs was used to test the null hy pothesis of no difference in the change over time be tween ST and SW (training intervention Ч time in teraction), and the null hypothesis of no difference in the change over time in response to the train ing intervention (main effect for time). With this statistical design, the following variables were ana lyzed: MMEP (Watt), Force (N) and speeds (m·s -1 ). The effect sizes were also calculated (eta squared, η 2 ) for better interpretation of the results and pva lue < 0,05 was considered significant. Testretest re liability [26] was satisfied in previous study [24] using the Intraclass Correlation Coefficient (ICC). Statistical analysis was performed using Sigma Plot TM software 11,0 (Systat Software, Tulsa, OK).

Discussion
The results shown that the transfer is effective an improving MMEP in masters male swimmers and might represent a technique useful to achieve better performance. In the last years, several authors [27,28] investigated new methods to improve the swim ming performances. Particularly, in the swimming history, several reasons have limited scientific knowl edge in water sports. Many technical approach were due to the environment that requires special equip ment; in fact, it is still difficult to validate the differ ent training methods so far tested in swimming [14]. DominguezCastells et al. [24] showed for the first time a new interesting method to assessing mechan ical power output as a reliable predictor of per formance of the swimmers [24]; for this aim the DominguezCastells methodology has been used in the present investigation.
Our findings are partly in agreement with the results of Morouco 14, who showed the existence of a relationship between dry land strength and power measured during swimming performance 14, 29. In dry conditions Morouco studied upperlower limb muscle strength and revealed high associations between swimming performance vs. muscular strength method 29. From our point of view, our work tries to change from that performed by Moruco particularly in two key points: entering the fast movements with weights (85 % 1 RM), and mixed training (weight training immediately followed by swimming sprints).
Force Considering the effects of the this method [20,21], the results indicate that mixed training in creased the strength in ST group by 9.03N (11,70 % increase). This effect could be emphasized espe cially for short time trials or several track compe titions (e.g. 50-100 m) where the results are of ten highly contested with close finishes. The present study results are in line with Schmidtbleicher et al. [30] and Padulo et al. [ 6] that have shown that few repetitions and maximal loads (> 80% 1 RM) in duced recruitment of fasttwitch motor units [6,30] and increased muscle strength (10,20 % p < 0,05) for ST, compared to repetitions with low loads and free speed. This interesting improvement of the strength in ST obtained with a mixed model training can be analyzed as a further deepening of the understand ing of the strength development in swimmers [13]. It seems that the adaptations of the swim intensity stimulate more than the mechanisms that trigger of aerobic capacity and limiting the development of the contractile muscle structure.
Conversely a decreased 4,16 % in SW does not stimulate fairly motor units. Indeed, swimmers train over many miles daily, and in the case of master swim mers this is more evident, at low intensity. Moreover the force applied in water requires particular sensitiv ity and gradation of effort [31]. The training for an enhanced MMEP, emphasizing neural adaptations, led to significant changes in rate of force develop ment using weight training. These results showed an increased in rate of force development and thus pow er production, rather than the increased swimming workout. For this reason the MMEP was no changed in SW group. These results in the SW group can be related to lower stimulation of muscle strength with only swimming as a main training activity. Velocity The swimming speed was measured during the semitethered test at maximal load, because crucial component of the power value obtained. Concern ing the maximum speed the MMEP showed a drop ping of ~5% (0,04 m·sec -1 ) in both groups (ST-SW) that represented a decrease with respect to the pre test value, resulting from the training interventions. Moreover, the velocity reported small differences (~5%) with no significant effects in both groups, in ST (min/max: 0,74-1,04 m·sec -1 ) this effects showed a shift on low speed of maximal power out put (5,73% with p < 0,05). In according with Mo rouco et al. [29] the velocity must not be assessed as a negative effect on swim performances because this velocity represent the ratio between power output and force in MMEP.

Maximal-mechanical-external-power (MMEP)
The results showed ( Figure 2C) an increase of 4.04w (MMEP), that representing 5,73% of pretest values in ST. Improvements in ST of MMEP could be relat ed to force production [ 32] more than in SW. The in creased MMEP in ST on is in agreement with explo sive movements on the neuromuscular systems [33]. In this regard, MMEP showed more accuracy in re lationship to the ecological validity because in our semitethered test the swimmers performed 15 meters of swimming with external loads. As confirmed by DominguezCastells et al. [ 24] and Morouco et al. [ 29] , the power test in swimming were altered when each subject were constrained to swim without wear on.
Combined effects of the variables studied The innovative method suggested by Comet ti highlights that for water sports, mixed train ing (land and water) is favorable to stimulate mus cle strength, in relationship at the combination of movement in dry conditions (weight training) with out other resis tance as Drag. In addition, the vari ous phases of eccent ric/concentric [34] contractions during exercise in land are not altered by the hy drostatic pressure. On the same topic, di Prampero showed that the greatest fraction of the energy ex penditure is utilized to overcome water resistance or Drag [17]. The 6 weeks explosivetype strength training resulted in considerable improvements in selected neuromuscular characteristics, although a large volume of endurance training was performed at the same time. An hypothesis is that trainingin duced alterations in neural control during stretch shortening cycle exerci ses (such as running and jumping [35]) might take place in both voluntary activation, inhibitory and facilitatory reflexes [36].
From our point of view it is not clear if MMEP and strength increased in the ST, obtained through an intense workout of 6 weeks training with combi nations «weight and swim training», has to be con sidered an important value to satisfy the research of mayor higher power. But again emphasizes how dif ficult it is to deve lop strength in the sport of swim ming, as reported in considerations of other authors [13,37].