Friday, May 1, 2009

Core strength doesn't matter?

Here is one for the "functional" training debate that we have been having recently. Here is the abstract:

Nesser TW, Lee WL. The relationship between core strength and performance in Division I female soccer players. JEPonline 2009; 12(2):21-28. To identify relationships between core stability and various strength and power variables in division I female soccer players. 16 NCAA division I female soccer players (height 163.6 cm ± 5.2 cm, weight 60.7 kg ± 7.5 kg) completed strength and performance testing prior to off-season conditioning. Subjects were tested on two strength variables (1RM bench press, and 1RM squat), three performance variables (countermovement vertical jump, 40 yard sprint, and a 10 yard shuttle run), and core strength (back extension, trunk flexion, and left and right bridge). No significant correlations were identified between core strength and strength and power. The results of this study suggest core strength is not related to strength and power. Core strength does not contribute significantly to strength and power and should not be the focus of any strength and conditioning program with the intent to improve sport performance.

Key Words: Training, Athlete, Core Stability

The whole study is available for you to read.

Here is some more:

Despite the fact significant correlations were not identified between core strength and athletic performance, it does not warrant neglect of the core. At the same time, it appears the core is no more important than any other body part.


Based on the results of the current and previous research, it is believed core training is necessary for optimal sport performance and should not be dismissed. However, it should not be the emphasis of any resistance training program. The core is one part of the body thus it is should not be the focus of any training program taking time away from other body parts which may lead to a muscle imbalance and possible injury.

interesting, challenging?

Hat tip to Luke


Anonymous said...

Ha! I knew Outside Magazine was full of shit.


Anonymous said...

Defining the core becomes problematic and makes these discussions take on a philosophical flavor.

Anonymous said...

I'd like to see more studies... I'm not sure I believe it. I do know that releasing core muscles... psoas, quad. lumborum, I"ll include hip flexors in that grouping... definitely seems to improve MY performance, whether running, turning, squatting down or lifting up. Something as simple as a cobra or Upward Dog pose, and or a slow and deliberate triangle pose, does the trick. And I'm not even a yoga guy :)

Physical Therapy Blog said...

Disappointing research, thanks for sharing, I enjoyed your conclusion. Agree with Anony above, defining "core" and it's true role in sports & function is often the challenge.

Mike Reinold

Jeff said...

Might be a stupid observation, but this might make sense with regard to beef. The most tender part of the cow is the tenderloin, located on the middle of the cow. It is tender since it does little work. The tougher parts are farther out to the legs which are carrying the load. Similar to the athletes, making the core of the cow stronger would be unlikely to make it faster either.

I would rather see a study of athletes who strength train their legs with squats/leg press, etc and see if that translates to speed and power performance. Do you know of any studies like that?

John Sifferman - Real World Strength Training said...

I agree with Anonymous and PTB above, if you don't define what you're talking about (especially when you're talking about the core), then nobody knows what's going on. This is one of the problems with research often being misunderstood or misinterpreted. Everyone has a different definition of "core muscles."

Chris and Jeff,

Here is a collection of studies on this topic... I think the take-home message that Chris has correctly stated earlier in the discussion is that sports performance is not DIRECTLY improved from strength training/functional training. However, performance can be improved INDIRECTLY through things like lifting/swinging weights, plyometrics, etc.

So, no your bench press training will not transfer over to a more effective/efficient boxing jab, but the strength/power you gain will help you to apply your skills with more strength and power once you can connect those attributes with your skill level. It just doesn't happen automatically.

Your conditioning is what gives you better access to your skills and your ability to apply them effectively, but it doesn't make you more skilled. Does that make sense?

Here is some of that research...

Luecke, T., Wendeln, H., Campos, G. R., Hagerman, F. C., Hikida, R. S., & Staron, R. S. (1998). The effects of three different resistance training programs on cardiorespiratory function. Medicine and Science in Sports and Exercise, 30(5), Supplement abstract 1125.
Four groups of young men performed different strength training activities:
LR (N = 9) performed 4 x 3-5 repetitions;
IR (N = 11) performed 3 x 9-11 repetitions;
HR (N = 7) performed 2 x 22-26 repetitions, and
C (N = 6) did not train.
Training lasted eight weeks and comprised three lower extremity exercises.
All groups improved in 1 RM on each of the training exercises. No changes were exhibited at 60% of 1 RM. VO2max and VE were unchanged due to training. HR was the only group to show a significant increase in time to exhaustion and maximal power output.
A high number of repetitions was associated with specific local muscular endurance changes.
Implication. Strength training improves strength in the exercises used for training. High repetitions are required for improvements in muscular endurance but then, the improvements are exercise specific.

Harris, G. R., Stone, M. H., O’Bryant, H. S., Proulx, C. M., & Johnson, R. L. (2000). Short-term performance effects of high power, high force, or combined weight-training methods. Journal of Strength and Conditioning Research, 14, 14-20.
The purpose of this study was to examine the effects of three different resistance-training methods on a variety of performance variables representing different parts of the force-velocity curve. After four weeks of high-volume training and pretests, male college football players were assigned to high-force (N = 13), high-power (N = 16), or a combination (N = 13) training group. Training was conducted four days per week for nine weeks.
The high-force group trained at intensities of 80-85% of 1 RM. The high-power group used intensities equivalent to 30% of peak isometric force. The combined group used both forms of training. Measurement variables included 1RM parallel squat, 1 RM quarter squat, 1 RM mid-thigh pull, vertical jump, vertical jump power, Margaria-Kalamen power test, 30-m sprint, 10-y shuttle run, and standing long jump.
The high-force group improved in the squat, quarter squat, mid-thigh pull, and M-K test.
The high-power group improved in the quarter squat, mid-thigh pull, M-K test, vertical jump, and standing long jump.
The combined training group improved on the squat, quarter squat, mid-thigh pull, vertical jump, vertical jump power, and shuttle run.
Body mass and composition did not change over the study period.
This investigation showed improvements occurred in activities that were like those involved in training. Slow strength movements improved slow-strength activities. Fast-power movements improved fast, powerful movements. When both forms of training were combined, the greatest number of performance factors was recorded. No form of training improved the 30-m sprint, the activity that is most likely to represent a sporting demand.
The findings of this study are limited to the actions trained and the actions tested, most of which were strength-training-type activities. It would be wrong to generalize and contend that combined training will improve sport performances that involve a variety of modalities. That assertion is too big a jump for what was found in this investigation. This study is limited to demonstrating its specific training effects.
Implication. Strength and power training activities improve strength and power activities.

Fagan, C. D., & Doyle-Baker, P. K. (2000). The effects of maximum strength and power training combined with plyometrics on athletic performance. Medicine and Science in Sports and Exercise, 32(5), Supplement abstract 659.
Ss (M = 19; F = 14) were randomly assigned to two training groups; maximum strength (85-90% 1 RM) and plyometrics, or maximum power (30% 1 RM jump squats) and plyometrics. Female competitive soccer players (N = 6) served as a control group. Training was performed twice a week for 10 weeks.
Both groups improved in lower body power and strength. Both forms of training were equally effective in increasing squat strength to perform plyometrics. However, sprint speed over distances of 5-40 meters did not change, therefore, this form of training was very specific and did not carry-over to a useful athletic pursuit.
Implication. Strength training only has specific effects on the trained exercises.

Laviano, T., Kierfer, S., Otto, R. M., Wygand, J., & Carpinelli, R. (2000). The relationship of benhc press and push-up performance to muscular strength and muscular endurance. Medicine and Science in Sports and Exercise, 32(5), Supplement abstract 1780.
An evaluation of the ability of different speeds of push-up and bench-press to predict muscular strength and endurance was conducted. Ss (N = 31) engaged in six crossover trials (three push-ups, and three bench-press). Muscular strength was evaluated with a free weight 1 RM bench-press. Muscular endurance was measured using 70% 1 RM bench-press at each of three durations: 2-sec concentric, 2-sec eccentric; 4-sec concentric, 4-sec eccentric; and self-selected pace. Push-ups were also performed at the three durations with females doing a modified push-up.
There was a significant difference in the number of repetitions performed between the three conditions. Only three low statistically significant correlations between the strength and endurance measures and the performances were revealed.
Faster movement speed exercises facilitated higher numbers of push-up and bench-press repetitions. Thus, those exercises relate to performance depending upon speed of execution and therefore, cannot be used for prediction unless movement speed is standardized. Although push-ups and %1 RM bench-press are considered to be indices of muscular endurance, they were not shown to be in this study.
Implication. Although common indicators, push-ups and bench-presses are not good predictors of muscular strength and endurance.

Weiss, L., Fry, A., Wood, L., & Melton, C. (1998). Comparative effects of deep versus shallow periodized squat training by novice lifters. Medicine and Science in Sports and Exercise, 30(5), Supplement abstract 942.
Ss (M = 10; F = completed periodized machine-based heavy-resistance training to determine if manipulating range of motion would have an effect on strength and power adaptations. Three groups were formed: a) deep squats that required the tops of the thighs to be parallel to the floor, b) shallow squats that were half the depth of the deep squats, and c) controls that did not participate in strength or power training. Training occurred three times per week for eight weeks.
Two forms of vertical jump were not improved by either form of training. The deep squat group was the only group to improve 1 RM shallow-squat strength.
It was concluded that training protocols were specific in their effects. Deep squats appear to elicit the best improvements for both shallow and deep squatting performance. Mahcine-based, periodized squat training does not enhance velocity-controlled squatting force and power or vertical jumping performance.
Implication. Machine trained squats do not transfer effects to other forms of performance. Strength gains are particularly specific.

Harney, R. G., Purcell, M., Martinez-Arizala, G., Reed, E., & Serfass, R. (2001). Relationship between anthropometric measurements, traditional modes of testing and training, and blocking performance in collegiate football linemen. Medicine and Science in Sports and Exercise, 33(5), Supplement abstract 1387.
The performances of blocking and charging football skills were correlated with their traditional weight-training activities (bench press, squat, and power clean). College football linemen (N = 10) performed the skills against an instrumented blocking sled as well as maximal assessments of the weight-training activities.
Low, non-significant correlations between the training activities and skill effectiveness showed that training on those weight activities does not transfer to skill performance. Training would have to be more specific to be of value.
Implication. Traditional weight training activities do not transfer to the American football skills of linemen.

Mayhew, J. L., Ware, J. S., Johns, R. A., & Bemben, M. G. (1997). Changes in upper body power following heavy-resistance strength training in college men. International Journal of Sports Medicine, 18, 516-520.
The effects of heavy-resistance training on measures of bench press power using absolute loads and seated shot put performance were measured. College men (N = 24) trained twice weekly for 12 weeks. Bench press power was measured by timing free weight actions at 30%, 40%, 50%, 60%, 70%, and 80% of 1 RM.
1 RM performance increased significantly (9.1%) after training. There was no change in shot put performance. Peak power was produced between 40-50% of 1 RM before and after training. There was no relationship between changes in shot put performance and changes in resistance-training strength.
Implication. Heavy-resistance training improves the activities involved in the training, such as 1 RM. Such gains are not transferred to more explosive activities. This adds support to the specific nature of resistance training and its inability to transfer to other activities.

Hetzler, R. K., DeRenne, C., Buxton, B. P., Ho, K. W., Chai, D. X., & Seichi, G. (1997). Effects of 12 weeks of strength training on anaerobic power in prepubescent male athletes. Journal of Strength and Conditioning Research, 11, 174-181.
Two groups of 10 prepubescent and pubescent male baseball players trained three times per week for 12 weeks using a variety of general free-weight and machine exercises designed for both strength and power acquisition. One group was experienced in strength training while the other comprised novices. A comparable control group (N = 10) did not perform the training program but did participate in all other non-experimental activities.
For the experienced, novice, and control groups respectively, the following gains were recorded: leg press — 41%, 40%, and 14%; and bench press — 23%, 18%, and 0%. Both training groups were significantly better than the control group. Similarly, the two training groups improved in vertical jump. However, the control group improved to a significantly greater degree in peak and mean anaerobic power and the 40-yard dash.
The training regime improved the training activities but did not transfer to functional performance measures. One could argue that the training actually caused anaerobic power and 40-yd dash measures to decrease, particularly in the experienced strength-training group.
The metabolic changes in training groups did not transfer changes in energy potential to dynamic cycling, supporting the principle of specificity. In particular, the high force/low velocity aspects of the training did not transfer to high velocity activities.
Implication. Strength and power exercises in pubescent males improved training exercises but produced worse performances in functional strength and power activities than in a non-training comparable group. Performance benefits from such training for this class of athlete are unlikely.

Miller, T. A., White, E. D., Kinley, K. A., Clark, M. J., & Congleton, J. J. (1999). Changes in performance following long-term resistance training in division 1A collegiate football players. Medicine and Science in Sports and Exercise, 31(5), Supplement abstract 1467.
This study analyzed performance changes in the squat, 20-yard shuttle run, 40-yard dash, bench press, vertical jump, and power clean in collegiate football players who had experienced a long-term, periodized strength and conditioning program at Texas A&M University. Players from 1993-1998 (N = 261) were tested twice per year. Ss were assigned to groups based on playing position: 1) defensive backs, running backs, and wide receivers; 2) kickers linebackers, tight ends, quarterbacks, and specialists; and 3) linemen. Relationships between performance changes and training time, body fat, and bodyweight were determined.
Body fat had a significant negative association with performance in all six activities for all groups. Neither training time nor bodyweight was related to 20-yard shuttle running or the 40-yard dash but both were related positively to the bench press and power clean. Bodyweight was significantly related to squatting performance. Results in the bench press, power clean, squat, 20-yard shuttle run, and 40-yard dash were consistent across all groups. For vertical jump, time showed a slight positive association only for group 1. Body weight had a positive effect on all three groups, being strongest in group 1 and weaker for the other two groups.
A strength and conditioning program was related to performance changes in strength and conditioning activities. However, there was little to no association between program training and the dynamic performance activities of vertical jumping, 20-yard shuttle run, and 40-yard dash, they being activities that could be transferred to game situations.
Implication. Strength and conditioning programs for football players make them better strength and conditioning trainers. There is little evidence of transfer of training effects to dynamic performances that are likely to be more associated with football playing performance. Increased body fat appears to hinder performances.

Bell, G. J., Petersen, S. R., Quinney, A. H., & Wenger, H. A. (1989). The effect of velocity-specific strength training on peak torque and anaerobic rowing power. Journal of Sports Sciences, 7, 205-214.
Eighteen varsity oarsmen from the University of Victoria were divided into three training groups: (a) high-velocity repetition (HVR) training, (b) low-velocity repetition (LVR) training, and (c) a no-training control. Rowing-specific exercises were performed on Hydra-Fitness machines in a repeated circuit format with the HVR group performing 18 to 22 repetitions and the LVR group performing six to eight repetitions of each exercise.
Training effects were measured on a rowing ergometer. A 90-seconds maximum performance was measured every 15 seconds with the 15 to 30 seconds interval being used as the measure of peak power output. The high lactic acid levels recorded in the subjects validated the test as being a measure of anaerobic capacity and power output.
It has been estimated that the contribution of anaerobic energy to rowing ranges from 14 to 23 percent. Usually, those contributions are greatest in the starting and finishing efforts of a race. The point behind this study’s resistance training program was that it should increase power and rowing ergometer performance should improve since the exercises used the muscles that are involved in the sport. The investigation assessed how much of the specific-resistance training effects transferred to ergometer work and thus, reflected the benefit of such training for rowing performance.
The results showed that there were specific changes in the performance of the specific resistance exercises, that is, the athletes became better resistance exercisers. Those changes were specific to the velocities of training. The HVR group performed better in the high velocity range of movements while the LVR group was better at low velocity actions. Contrary to what has been reported by Moffroid and Whipple (1970), each of the training groups changed specifically, that is, the high-velocity group did not show any improvement in low-velocity movements.
The control group worsened in performance. There was no change in either training group in peak power output or lactic acid levels. This finding was surprising because the strength program was specifically designed to enhance the strength of the muscle groups involved in rowing. Since power is dependent on both force and velocity, the observed improvements in torque with resistance training should, theoretically, have contributed to an increase in rowing power. That theoretical position was not supported by the results of this study in these high-caliber athletes. The lack of improvement contradicts the recommendations of many coaches and the content emphases of many rowing training programs. This negative finding might be explained by the fact that the movement patterns involved in rowing are very complex and require a high degree of skill. The training effects that were observed in this study were specific to the resistance-training mode and did not transfer to the more complex action involved in the sport. This restriction supports the training principle that training effects achieved on simple activities (such as specific resistance exercises) do not transfer to complex activities.
This study failed to show performance benefits that are supposed to result from resistance training programs. It supports the absolute specificity of training principle and suggests that an emphasis on resistance training in high-level athletes is not useful for improving performance. Such programs may even restrict the volume of beneficial specific training that can be achieved because of the level of fatigue that results from their execution. Neither modern training theory nor the mounting evidence of the ineffectiveness of specific resistance training programs supports the continued emphasis on this type of training as a means of generating performance improvements in high-caliber athletes.
Implication. Traditional use of resistance training programs that are “meant” to improve performance should be questioned. The only time that resistance training may be of value would seem to be in the transition (off-season) for basic preparatory training phases. There is the possibility that fatigue generated by strenuous resistance activities will: (a) diminish the physical resources that can be applied to specific beneficial training;
(b) detract from the amount of available training time so that the volume of specific beneficial training is reduced; and
(c) the training effects from resistance training will be incompatible and interfere with beneficial specific training effects (principally those of aerobic adaptation).

Hetzler, R. K., DeRenne, C., Buxton, B. P., Nelson, K. R., Seichi, G. M., Chai, D. X., & Ho, K. W. (1994). Effect of 12 weeks of strength training on anaerobic power in pubescent male athletes. Medicine and Science in Sports and Exercise, 26(5), Supplement abstract 469.
Control, experienced, and inexperienced males (13.6 +- 3.9 yrs), after strength training (3 days/wk) were evaluated on the Wingate, Margaria, and Sargent Jump tests, 40 yd dash, leg press (1RM), and bench press (1RM). It was found that improvements in strength occurred in the training exercises but there was no increase in anaerobic power tests. Implication. The benefits of strength training are limited in adolescents. They do not transfer to anaerobic performance in activities other than the training exercises themselves.

Bloomfield, J., Blanksby, B. A., Ackland, T. R., & Allison, G. T. (1990). The influence of strength training on overhead throwing velocity of elite water polo players. Australian Journal of Sience and Medicine in Sport, 22(3), 63-67.
The relationship between muscular strength and morphology with overhead throwing velocity was examined in elite water polo players (N = 21). A strength training and no-training control group were formed.
An 8-week program using “Nautilus” equipment and emphasizing upper body strength development was employed. Regular swimming and game practice continued.
Significant relationships were found between throwing velocity and standing height, body mass, lean body mass, stem length, bicromial width, arm girth, and forearm extension strength.
Following strength training, no change in throwing velocity was observed in either group. In the strength training group there were significant increases in arm girth, mesomorphy, and arm medial rotation strength.
The authors explained the results this way:
“It is more likely that this homogeneous group of elite water polo players already possessed optimum levels of upper body strength . . . and that diminished strength returns were gained from the extra training. More substantial strength gains would have been expected from players of lower calibre with poorer overall physiques.” (p. 67) Implication. The study really shows strength training on unrelated activities does not improve speed actions. Strength training had no carry over to the skill tested because it was neither neuromuscularly nor modality specific.

Butchar, J., & Becque, M. D. (1966). Effects of high and low intensity weight training on iEMG and force. Medicine and Science in Sports and Exercise, 28(5), Supplement abstract 1139.
High and low repetition training programs were evaluated for effects on the EMG to force relationship. Low repetitions (LR) were 2-6 and high repetitions (HR) were 10-15, both for 3-4 sets per session twice per week. EMG recordings were integrated (iEMG).
The iEMG to force ratio increased for the LR group and decreased for the HR group. Low repetitions increase activation and strength. High repetitions increased strength but did not increase activation. It was concluded that the mechanism for strength increase in experienced lifters is dependent upon the loads used.
Implication. Weight training effects are specific to the load used. Consequently, any changes demonstrated are unlikely to transfer to other activities because the training effects are so specific.

SB said...

The study discussed in the post seems like a classic example of restriction of range . . . i.e., a handful of people who are already accomplished athletes. Moreover, it's not causal: All it shows is that some people are perhaps stronger in their squat than in their core, or vice versa, but that doesn't tell us what would happen if they did more planks.

Bryce said...

Defining the core . . . just the abs and spinal erectors? Do you add in the glutes and hip flexors? It's extremely vague. Even the lats connect to the trunk musculature. Are the part of it?

I don't think this study is very conclusive, but at the same time, I don't think it's conclusions are very controversial. I don't really think you should train the core (as the study would describe it) any more or less than anything else.

Chris said...


Thanks for those references


eshlow said...

Bryce... the core is the lower back, abs and everything else... transverse abdominus, QL, obliques, etc.


This study blows because there's no need for the core in those movements tested.

Core in athletics especially hitting sports is the ability of the body to resist twisting:

1. In baseball, the power of the swing comes from the legs, and the core acts as a "brace" to prevent twisting to transmit the leg power to the upper body to hit the ball.

2. Same thing occurs in the golf swing... all leg power and the "core" is used to brace the club when hitting the ball.

When evaluating core strength you need to look at movements in sports where you have transmission of power/energy from an action of the lower body to upper body (hitting sports) or upper body to lower body (some gymnastics comes to mind).

I don't think anyone is going to argue that the best baseball players or gymnasts don't have strong cores for a reason.

If your premises are junk then your study is going to be junk as exemplified by these studies.

eshlow said...

Also, throwing sports and weightlifting (obviously) use core as well.

In fact, the ability of the core to resist stress during weightlifting is one of the key ways to develop it.

No one is going to put up a 500 lbs DL or a 500 lbs squat without a strong core.

So next time you see a study like this you can shake your head because they don't know what they're doing.

Anonymous said...

Bad study.

The trunk testing was a measure of strength-endurance. The performance tests were tests of maximal strength and power.

From this we can conclude that apples do not taste like oranges.

Bill Hartman

Sascha Fast said...

I think that transfer should be definied more precise. Thus there is no immediatly improvement of skill, there could be an increase to improve the skill further. A increasement in muscle tissue in glutes and hamstrings could lead to performance increase, if someone lacks in this issue.

Through my personal experience (of course unsafe induction) my boxing skills improved very rapidly after including some general (Deads, Press, Chins etc.) strength and sprint training, after I avoided it trough the first two thirds of my boxing career.

The very small sized concept of specifity is challenge through Schöllhorns differential lerning.
Following (wfs) link includes some English papers.

Following the nontransfer argument my boxing skills would be confused and therefor decreased through shadow boxing, because there I don't wear gloves which should be comparable to playing soccer with a signifikant lighter ball.

Greets from Germany
Sascha Fast