• Introduction
• Part 1: Maximum Strength/Strength Training For Strength/Power Sports
• Part 2: Maximum Strength/Strength Training For "Endurance" Sports
• Summary

Introduction

The purpose of this brief review is to consider the association of measures of maximum strength in relation to sport performance endurance and endurance related factors. Evidence from different types of cross-sectional research as well as observational data was considered. Collectively the data indicate that the association between maximum strength and sport performance endurance factors is stronger then might be expected. While, explaining performance/endurance in sports is a multi-factorial problem there is little doubt that maximum strength is a key component.

Strength can be defined as the ability to produce force (Siff 2000, Stone 1993). Because force is a vector quantity, the display of strength would have characteristics including a magnitude (0-100 %), a rate and a direction. Furthermore the generation of force can be isometric or dynamic. The magnitude of force production and its characteristics are determined by a number of factors including the type of contraction, the rate of muscle activation and the degree of muscle activation. The direction of force production is related to the motor unit and muscle activation patterns.

The importance of force production can be ascertained from Newton's second law: F = ma

Thus acceleration (a) of a mass (m) such as body mass or an external object depends directly upon the ability of the musculature to generate force (F). Furthermore, power production is the product of force and velocity and is likely the most important factor in determining success in most sports. Thus, the "ability to generate force (strength)" is an integral part of power production and therefore may be a key component in determining athletic success. Exactly how these factors impact upon endurance is still not totally clear.

While these two observations/mechanisms can be used to explain much of the association between maximum strength and endurance capabilities - they do not necessarily explain the increases in endurance associated with strength gains for all activities. For example: Although in sports and daily living it can be argued that absolute loads are regularly encountered there may be few (if any) instances where true relative loads (relative to maximum strength) are encountered. Fundamentally, these two observations/mechanisms do not consider additional "underlying" mechanistic possibilities.

As can be ascertained there are several possible reasons as to why strength training may enhance endurance. This discussion will be broken into two parts. Part 1 deals with strength training and strength/power sports, this type of endurance may be termed high intensity exercise endurance (HIEE). HIEE can be defined as the ability to sustain or repeat high intensity exercise. Part 2 deals with strength training and long-term endurance activities, this type of endurance may be termed low intensity exercise endurance (LIEE). LIEE would be the ability to sustain or repeat low intensity exercise.

It should be noted that as with any type of training goal (i.e. maximum strength, power, endurance) the impact of the training program depends upon training factors including mechanical specificity (see Explosive Exercise) the training volume and intensity factors, rest period length and the trained state.

Part 1: Maximum Strength/Strength Training For Strength/Power Sports

Effects of enhanced maximum strength correlational studies: A correlation is the strength of the relationship among variables - the correlation coefficient (symbolized as r ) ranges from -1.0 to 1.0; the closer the coefficient is to 1.0 the stronger the relationship. A positive correlation between two variables would mean they increase together, a negative correlation would mean an inverse relationship. Hopkins (1997) has ranked correlations as r =

Glaister et al (2000 - unpublished data) studied the relationship between the 1 RM parallel squat and tests of agility, jumping capabilities, and endurance using elite Scottish badminton players (n =13). This study was part of the ongoing sports testing/sports science programme initiated by the Scottish Institute of Sport. The results (Glaister et al 2000) indicated that the 1 RM squat was strongly related to weighted and unweighted counter movement and static vertical jumps as well as tests of agility (r = 0.65 - 0.87) and was strongly related to the ability to short-rest repeated agility runs. Glaister et al (2000) created a badminton specific agility test (X-test) that could be repeated producing a measure of endurance. The X-test was repeated 15 times with a 14 s interval for males and a 16 s interval for females (simulating the exercise: rest intervals during badminton). The correlation between the 1 RM squat and the repeated X-test was 0.69. Glaister et al's (2000) results indicate that 1 RM squat strength (and 1 RM squat per kg body mass) have significant relationships with power, speed and speed-endurance related variables.

Using previously trained (n = 33) subjects Robinson et al. (1995) showed that high volume strength training for 5 weeks could increase power output and HIEE. Power and HIEE was measured by 15- 5sec maximum effort cycle rides with 1 min rest intervals (0.1 kg x body mass). Robinson et al (1995) showed that maximum strength as measured by the 1 RM squat had strong and increasing correlations with cycle PP, average PP (APP15) over 15 rides and average work accomplished (ATW15) over 15 rides:

These studies Glaister et al (2000) and Robinson et al. (1995) indicate that maximum strength is related to HIEE. Additional studies indicate that greater maximum strength can be related to increased power and endurance in various activities including sprint swimming (Costill et al. 1980, Davies 1959, Sharp et al. 1982) and sprint cycling (Stone et al. 2003). However, cross-sectional/correlational data do not necessarily imply cause and effect.

Effects of enhanced maximum strength longitudinal studies: It is well established that stronger athletes have a greater absolute endurance (Anderson and Kearney 1982); however, it is not uncommon for these athletes to undergo periods of high volume strength training (strength-endurance training) or power training (power-endurance training) during specific portions of a training cycle (i.e. general and specific preparation phases). Part of the reason for using a "strength-endurance or power-endurance phase" is the belief that HIEE will be enhanced beyond that of typical strength training. To address this issue McGee et al. (1992) compared 3 different training groups consisting of:

GpL (n = 8) = low volume - 1 set of 8-12 RM to failure*
GpV (n = 9) = multiple set variation group**
2 wks - 3 x 10 RM
3 wks - 3 x 5 RM
2 wks - 3 x 3 RM
GpH (n = 10) = 3 x 10 RM**

*One light warm-up set
** three sets light >> moderate

The subjects trained using large muscle mass exercises and emphasized leg and hip strength-endurance. Training was 3d/week for 7 weeks. The subjects were all trained in the same manner for two weeks prior to the study. Endurance was measured by two methods; cycle ergometry to failure (< 5 min) at a constant load (4.5 KP) and parallel squats to failure with increasing loads. Pre-post testing found that while all groups improved, the greatest percent improvement for both tests was GpH > GpV > GpL. Additionally it was noted that although the greatest improvements were specific (i.e. squats), considerable improvement in cycle endurance also occurred. The authors concluded that the degree of strength training induced adaptations in HIEE was volume dependent agreeing with the general observations and conclusions of Stone and Coulter (1994).

It is commonly believed that shortening the rest interval between sets enhances the endurance training effect. Robinson et al. (1995) used moderately trained subjects and investigated rest interval effects on HIEE. Three different inter-set rest periods were studied:

Gp 1: (n = 11) 3.0 min rest intervals
Gp 2: (n = 11) 1.5 min rest intervals
Gp 3: (n = 11) 0.5 min rest intervals

The subjects trained 4d/week, for 5 weeks using exercises that emphasised the legs and hips. All subjects performed 5 x 10 repetitions for all major exercises emphasising HIEE only the rest intervals were different. Pre-post tests included the vertical jump (VJ), 1 RM squat and 15- 5sec maximum effort cycle rides with 1 min rest intervals (0.1 kg x body mass). Gps 1 and 2 showed non-significant improvements in the VJ, while Gp 3 showed a non-significant decrease, GP 1 significantly increased in the squat compared to Gp3. All 3 groups improved significantly on the cycle tests (see correlational studies) with no differences between groups. The authors (Robinson et al. 1995) concluded that shortening the rest intervals did not produce an advantage for developing HIEE agreeing with the observations of Nimmons (1995). In similar investigations Kulling et al (1999) found that longer inter-set rest periods facilitated HIEE adaptations. Kulling et al. (1999) found that 90 s rest periods, compared to 30 s, resulted in more repetitions to failure in bench presses at a percentage of body mass (60% for men and 40% for women) after 12 weeks of training. The longer rest periods allowed a higher training intensity, which facilitated adaptations in strength and endurance. These data indicate that if inter-set rest periods are too short (< 90 s) then training intensity (i.e. average load) and subsequent adaptations are compromised. These observations bring into question the practice of using circuit resistance training (CRT) to enhance strength-endurance. CRT uses short rest intervals attempting to increase the average metabolic expenditure. However, the short rest periods can compromise exercise loading and subsequent adaptations.

1. Although specificity is evident, strength training can produce adaptations in endurance, which is "transferable", i.e. adaptations can take place in exercises not used in the strength-training program.
2. Higher volume training can affect measures of endurance to a greater extent than low volume training.
3. Within the context of strength-training short rest periods (< 90 s) do not enhance endurance beyond using typical rest periods and can compromise strength and power gains. If rest periods are too short (< 30 s) loading may be compromised sufficiently to result in smaller gains in strength power and possibly HIEE.

Part 2: Maximum Strength/Strength Training For "Endurance" Sports

Among coaches and athletes strength training for LIEE has been quite controversial (Suslov 1997, Reuter 2000). Recently data from several longitudinal studies have indicated that strength - power training can enhance long-term endurance (LIEE). This brief review will deal with those studies.

Correlational/Descriptive Studies: Several studies have shown that strength or power measures are associated with endurance performance, For example: Among road cyclists anaerobic power was a major factor separating higher and lower rank athletes (Tanka et al. 1993). Anaerobic power has been shown to be a critical factor determining success among cross-country runners with similar a VO2max (Bulbubian et al. 1986). Additionally evidence indicates that distance runners with more powerful muscles are more likely to succeed (Nokes 1988). Several studies have shown strong correlations between swimming performance up to 400 m and maximum strength/power of the upper body (Costill, et al. 1980, Davis 1959, Hawley and Williams 1991, Sharp et al. 1982, Toussaint and Vervoorn. 1990). These data indicate the potential for strength training and increased maximum strength to enhance endurance.

Longitudinal Studies: Several longitudinal studies have noted a relationship between increased strength and increased anaerobic power and measures of endurance as a result of strength training in untrained or minimally endurance trained subjects (Hickson 1980, Inbar et al. 1981, Marcinik et al. 1991, O'Bryant et al. 1988, Petersen et al. 1984, Rutherford et al. 1986, Smith, 1987,). Strength training has also been shown to produce increases in endurance among trained subjects and well-trained athletes. For example:

Hickson (1988) studied the effects of adding strength training to the training programs of already endurance trained subjects (8 men, 2 women, n = 10). The subjects were moderately endurance trained (> 50 ml x kg-1x min-1). Ten weeks of strength training (3d/wk) emphasizing leg and hip strength resulted in significant gains in maximum strength (20-38%). Although there was little change in aerobic power incremental treadmill and cycle times were markedly increased as was time to exhaustion on a cycle ergometer at a constant work-rate (80-85% of VO2max). From a practical side 10k running time decreased from 42:27 ± 1:59 to 41:43± 1:45 (n =9).

Paavolainen et al. (1999) investigated the effects of "explosive strength training" on the performance capabilities of 18 male well-trained orienteers (VO2max = 65 ml x kg-1 x min-1). In an attempt to partially control for training load differences, endurance training time was replace with strength training (32% of total time) so that the total approximate training time was equal between experimental (GpE, n =10) and control (GpC, n = 8) groups. Interestingly (compared to GpC), GpE showed a decrease in VO2max over the 9-week experimental training period. However, GpE showed superior gains compared to the control in maximum strength (isometric leg press) a 20M sprint, jumping ability, anaerobic capacity (VMART), running economy and most importantly 5K time.

Strength training has also been shown to have beneficial effects on endurance factors associated with road cyclists. Bastiaans et al. (2001), using 14 male competitive road cyclists, investigated the effects of explosive strength training on endurance related factors. As with Paavolainen et al. (1999) endurance training time was replaced with strength training (37% of total time) so that the total approximate training time was equal between experimental (GpE, n =6) and control (GpC, n = 8) groups. While the addition of strength training resulted in small increases in power output and riding efficiency the major effect dealt with "short-term performance". Short-term performance was measured by calculating mean power output at a fixed pedal rate (60 RPM) during a 30 s ergometer test. It was shown that GpC lost mean power and GpE showed small increases over the 9 week period. The authors (Bastiaans et al 2001) suggested that the data indicated strength training attenuated the commonly observed loss in power and sprint ability associated with long-term endurance training. GpE showed slightly greater improvement in work accomplished during a 1 h ergometer time trial. These data suggest that replacing endurance training with explosive strength training can preserve or enhance the ability to maintain high power outputs for short periods without compromising endurance.

1. Maximum strength can be associated with LIEE.
2. Strength training can improve LIEE or factors associated with LIEE.
3. As with strength/power sports there is a degree of specificity in the endurance adaptations.

1. One possibility is that strength training has little effect on endurance factors - in the authors' opinion this factor is unlikely because a) there are ample studies indicating an effect and b) athletes and coaches are very pragmatic, most coaches and athletes do advocate some form of strength training for their endurance athletes in the belief that it will enhance performance, it is quite unlikely that athletes/coaches would waste time and effort on training that does not produce reasonable results.
2. It is possible that the type of resistance training programme used was not specific enough for the task (sport event). For example: Bastiaans et al. (2001) argue that one possible explanation for Bishop et al's (2001) finding of no improvement in endurance with strength training deals with the type of contraction used. Bishop et al (2001) used typical heavy slow velocity strength training, which may not match the characteristics of the task ("high speed endurance cycling"). In this context it is interesting that both Paavolainen et al (1999) and Bastiaans et al. (2001) used dynamic explosive moments for the training intervention which may have matched the characteristics of the task better than slower movements. However, Millet et al. (2002) used typical heavy strength training procedures and found improvements in movement economy among very well trained cross-country skiers. Differences in the trained state may also have affected the out come.
3. Another factor which may affect outcome is the total training volume, Paavolainen et al (1999) and Bastiaans et al. (2001) substituted strength training for endurance activities thus (to a point) maintaining total training volume - studies adding strength training to existing training regimens may have increased the total volume such that chronic fatigue interfered with adaptations.
4. No longitudinal study has demonstrated that maximum strength, power or specific performance variables including endurance adapt at exactly the same rate. Often gains in sports perfromance variables lag behind the measured gains in strength and power (Stone et al. 2003). It is possible that the lack of direct correspondence between maximum strength gains and other performance variables is associated with a lag-time (Abernethy and Jurimae 1996, Stone et al. 2003). Lag-time deals with a period of time in which the athlete learns how to use the increased strength; the lag time may extend many months in some cases. It is possible that lag-time may be reduced by careful coaching strategies in which the potential link between strength and technique/endurance is pointed out to the athlete. This may partly be accomplished by pointing out similarities between training exercises (i.e. mechanical specificity) and performance exercises.
5. Furthermore, the increases in strength may continue after the changes in sport performance become asymptotic. This observation may indicate that a change in the type of strength training being used is necessary.
   

Summary

Figure 1: Adaptive Mechanisms (Modified from Paavolainen et al 1999)