Ankle sprains represent from 38 to 50% of the total sport injuries (Jones et al., 2000; Leaf et al., 2003; Thacker et al., 1999; Verbrugge, 1996). Garrick and Requa, 1988 estimated that one-sixth of the total time lost by sport injuries was attributed to ankle sprains. Functional taping and ankle braces are passive preventive measures frequently utilised in sports (Osborne and Rizzo, 2003; Robbins and Walked, 1998). Studies on the influence of functional taping on sports tasks during actual competition are scarce (McCaw and Cerullo, 1999; Riemann et al., 2002), and most of them only analyse the passive ROM restriction (Hume and Gerrard, 1998). The studies that analyse jump tests and static balance are the most common among those that assess the influence of ankle taping on performance tasks in sports (Hume and Gerrard, 1998; Cordova et al., 2002). Research that studied jump performance focused on the changes in jump height with taped subjects (Burks et al., 1991; Mackean et al., 1995; Verbrugge, 1996). Some of them reported decreases in jump performance (Burks et al., 1991; Mackean et al., 1995; Verbrugge, 1996), but this remains a controversial issue.

In addition, a few studies have analysed drop landings and functional taping (McCaw and Cerullo, 1999; Riemann et al., 2002), showing decreases in the time to dissipate landing forces and adverse effects on the landing kinematics. The risk of 'overuse' injuries will increase if the ability to reduce landing forces is impaired by limiting the mobility of the lower extremities (Dufek and Bates, 1991; Hewett et al., 2005).

The studies on balance have focused on the centre of pressure (COP) trajectories to evaluate performance (Bennell and Goldie, 1994; Cordova et al., 2002; Feuerbach and Grabiner, 1993; Hertel et al., 1996; Kinzey et al., 1997; Paris, 1992). Better performance is shown by shorter trajectories or narrower areas of the COP. Nonetheless, some authors have utilised less accurate methods to evaluate balance, such as counting the number of times the subject needed to keep his balance (Bennell and Goldie, 1994), or the time spent by the subject on a fixed bar (Paris, 1992). There are contradictory results on the influence of preventive ankle taping on balance tests. Hertel et al., 1996 found no differences between subjects with and without taping in three balance tests. One of them was performed with static monopodal stance and the other two were dynamic tests. However, other authors such as Bennell and Goldie, 1994 concluded that ankle taping led to a decreased postural control in similar balance tests.

Therefore, studies on the effects of ankle taping during specific movements, such as jumps or balance tasks, are scarce, and its influence on sports performance is controversial. The present study analysed the changes in ground reaction forces and the path of the COP during balance tests. The performance of taped subjects during static and dynamic balance tasks could be improved by the increase in exteroceptive input provided by the taping (Feuerbach and Grabiner, 1993). From previous studies, we hypothesised that prophylactic ankle taping on uninjured subjects would decrease their jump performance and increase the peak vertical forces during the landing phase. On the other hand, we expected an increase in the subjects' performance in the balance tasks, especially in the static balance tests. More biomechanical research on the effects of functional taping on sports performance is necessary to clarify its effects during actual sports tasks. Therefore, the research question of this study was: does prophylactic ankle taping influence on performance of two balance tests (static and dynamic balance) and the push off and landing phase of one jump test?

Participants
Fifteen physically active subjects, seven men and eight women, volunteered for the study. Their physical characteristics are given in Table 1. The participants are regularly involved in recreational sports, at least twice a week, but none of them had competed professionally. None of the subjects have used ankle taping or bracing (Bennell and Goldie, 1994) or have had lower limb injuries in the last 6 months (Greene and Hillman, 1990; Gross et al., 1991). An experienced physiotherapist confirmed this information with a medical history and a physical examination, including ligamentous and range-of-motion tests one week before testing. The subjects performed all the tests with indoor court shoes.

Intervention
The anthropometric characteristics were determined using a calibrated scale with height rod (Seca Ltd, Hanover, Germany), an anthropometer (GPM, SiberHegner Ltd., Zurich, Switzerland), a 1.5-m flexible tape (Holtain, Croswell, Crymmych, UK), a bicondylar caliper (GPM, SiberHegner Ltd., Zurich, Switzerland), and skinfold calipers (Holtain, Croswell, Crymmych, UK). Fat mass was calculated from six skinfold measurements (triceps, subscapular, umbilicus, suprailium, thigh, and lower leg) according to the equations of Carter, 1982. Fat free mass (FFM) was calculated by subtracting fat mass from total mass and muscular mass (expressed as a percentage of total mass) was calculated by subtracting bone and residual mass from FFM.

A prophylactic taping, modified Gibney closed-basket-weave (Wilkerson, 1991) (designed for subjects without previous ankle injuries to restrict ankle inversion) was done in both ankles by a physiotherapist, with a prewrap, to protect the Achilles tendon and restrict ankle inversion. Two adhesive anchors were applied to the skin according to the subjects' body dimensions (Figure 1). The inferior adhesive anchor was applied over the metatarsal head with six active strips that limited ankle inversion, and 13?17 strip locks were utilised, depending on the size of the lower limb.

Each participant performed the three tests in two different situations: with taping (T) and without taping (NT). The tests were as follows: countermovement jump (Figure 2), static balance (Figure 3), and a dynamic posturography test (Figure 4). The tests and conditions (T-NT) were randomly performed. Static balance tests were performed on a force platform (Piezoresistive force platform Dinascan 600M; IBV, Valencia, Spain). The force data were digitally converted and stored in a computer for subsequent analysis using the software Estabilometría (IBV, Valencia, Spain). The force-time data from the countermovement jump were assessed on a Quattro Jump Portable Force Plate System (Kistler, Winthertur, Switzerland) at 500 Hz. This sample rate has been previously utilised for assessing landings in the studies of Hopper et al., 1999 and Ozguven and Berme (1988). The forces were normalised and expressed as times body weight (BW). A standardised 10-min warm-up was carried out by the participants before each session. The warm-up consisted of 5 min at 175 W on a cycle ergometer Ergomedic 894 Ea (Monark, Varberg, Sweden), stretching of the lower limb muscles directed by the researcher, and six jumps (three submaximal and three maximal).

Outcome measures
Countermovement jump: The subjects performed the test on the force platform with the hands placed on the hips during the whole jump. The knee angle during the counter movement was not controlled. The participants performed three valid trials and the one with the greatest jump height was recorded for further analysis (Figure 2). The variables analysed during the push-off phase of the jump test were jump height (h), from the flight time, peak vertical forces (PF) and peak power (PP), obtained from the integration of the force-time record. In addition, in the landing phase, we analysed the first and second peak vertical force values (F1 and F2), the time that elapsed from the feet contact to F1 and F2 (T1 and T2, respectively), and the time from feet contact until the vertical ground reaction forces reached the subject's weight for the first time after the landing movement (TBW).

Static balance on monopodal stance: The subjects had to remain as still as possible standing on the right leg, with the left lower limb at 90º of hip and knee flexion, during 15 s. Their hands had to be placed on the hips throughout the test, and the feet were placed in the same location on the plate in all the trials. The aim of the test was to keep to the minimum the area in which the movement of the subject was taking place, defined by the trajectory of the COP (Figure 3).

The variables analysed were the area covered by the COP and the average position in the antero- posterior axis (average of X values) and medial-lateral axis (average of Y values). Three trials were completed and the best performance, that is, the one with the lower area, was recorded for subsequent statistical analysis.

Postural sway test: Dynamic balance was measured using computerised dynamic posturography: the subjects were in standing position on a force platform with hands on hips, and balance was assessed by modifying visual feedbacks and asking the participants to score a circle as fast and as accurately as possible in response to the changes in the visual feedback by moving their bodies. Eight red circles, projected in a wide screen in front of the subject, were randomly lit for periods of 4-6 s. The test lasted 40 s. The analysis of the transitional period from one lit centre to another included the calculation of the time to reach the lit centre and the percentage of the time during which the subject remained inside the centre as a percentage of the overall time of the lighting of the centre (hits). The best of three trials, that is, the one with the longest time into the target, was recorded for subsequent analysis (Figure 4).

All the variables analysed were recorded from the best trials because we aimed to compare maximal performance and not patterns obtained by averaging the data from several trials (Bosco et al., 1999; Macpherson et al., 1995).

The reliability of the main variables was assessed with the intraclass correlation coefficient (ICC) and the typical error, from three measurements of each variable (Hopkins, 2000). In a pilot study, carried out with six subjects, the ICCs were very high for all the variables (0. 94-0.99). Typical errors in the jump height, F2 value, area covered by COP in the test of the static balance on monopodal and hits from the postural sway test were 0.16 cm, 0.11 BW, 7.37 cm² and 2. 47%, respectively.

Data analysis
Based on the data obtained in a pilot study, the minimal number of subjects required with a power of 0.8 and a level of significance ? of 0.05 was calculated to be 14, considering differences in F2 between T and NT. Descriptive statistics included mean and standard deviations; relationships between variables were examined using Spearman´s correlation test. Differences between T and NT conditions were assessed with the Wilcoxon matched-pair test. Significance was accepted at the level of P < 0.05.

Tables 2 and 3 show the means, standard deviations, percentage differences, and the levels of significance of the variables studied in the balance and jump tests, respectively. There were only significant differences in the average of X values in the static balance test (3.23 cm, 95% CI -1. 28 to 7.74) and in the F2 value of the landing (0.66 BW, 95% CI -0.64 to 1.96), with greater values noted in the T condition in both cases.

The most important correlations between the T and NT conditions are shown in Table 4. There were significant correlations among variables in all the tests, with the exception of the postural sway test. There was a significant negative correlation between F2 and T2 in both conditions (T: r = -0.66 (95% CI -0.88 to -0.23), p < 0.01; NT: r = -0.58, (95% CI -0.85 to -0.10), p < 0.05).

• Ankle taping has no influence on balance performance.
• Ankle taping does not impair performance during the push-off phase of the jump.
• Ankle taping could increase the risk of injury during landings by increasing peak forces.