Skip navigation

Monthly Archives: September 2011

There are a variety of ways to train straight-ahead sprinting speed. Some tools involve assisted sprinting and resisted sprinting. Assisted sprinting helps the athlete run faster in training than he/she is normally capable of. The idea being that this trains the nervous system and muscles to move faster and this will eventually carry over to unassisted sprinting. Resisted sprinting makes the running process more difficult by providing resistance that the athlete must sprint against. The idea here being that the resistance will require the athlete to recruit more muscle fibers/motor units to sprint, which will eventually carry over to unresisted sprinting – making the athlete faster.

Both training tools are effective, though each has drawbacks. The biggest one being that too much disrupts running mechanics which may cause the athlete to develop bad sprinting habits. While both are effective, both are not for every athlete, every sport, or every situation. Like a lot of training tools, these are best for athletes that have reached a certain level of development.

Upton studied the effects of assisted and resisted sprint training on Division I female soccer players. The subjects were divided into three groups; assisted, resisted, and traditional sprint training. Each group performed 10×20 yard sprints, 3x/week, for four weeks. The assisted group used a commercial device where a research assistant pulled the athlete along, such as the one below from Power Systems.


The assisted group used a device that was attached to a pole, that allowed for the resistance to be selected at 12.6% of athlete’s body mass, such as the one below from Perform Better.

At the conclusion of twelve weeks of training:
• The assisted group improved their ability to accelerate during the first 15 yards of the 40 yard sprint, which resulted in a .
• The resisted group improved their ability to accelerate during the last 25 yards of the sprint.
• The traditional group made no statistically significant improvements in their performance.

This seems to suggest that in this population assisted and resisted training modes impact sprinting performance differently, with the assisted training being more beneficial for acceleration and the resisted training being more beneficial for maximal velocity sprinting.

Upton, D. (2011). The effect of assisted and resisted sprint training on acceleration and velocity in division IA female soccer athletes. Journal of Strength and Conditioning Research, 25(10): 2645-2652.


Plyometrics are a popular mode of exercise in the strength and conditioning of athletes. As I have posted elsewhere (see ), different plyometric exercises have different power outputs, ground reaction forces, and rate of force developments. This means that especially with advanced athletes, plyometric exercises have to be selected with this in mind so that they match up with the physical qualities that an athlete is attempting to train.

In the October issue of the Journal of Strength and Conditioning Research, Cappa and Behm compared jumping over hurdles (one-leg and two-leg) with counter-movement jumps (CMJ). The idea behind the study was to investigate and quantify the differences between the various jumps.

The authors studied thirteen male rugby and soccer players over two testing sessions. In the first session, the subjects performed maximal CMJs on a force platform. In the second session, the subjects jumped forward over a series of four hurdles set 50cm apart. A force platform was set up after the second hurdle. Subjects did two-legged jumps and single-legged jumps. For the two-legged jumps, the hurdle height was set at 100%, 120%, 140%, and 160% of maximum CMJ height in a random order. For the single-leg jumps, it was set to 70%, 80%, and 90% of CMJ height in a random order.

The results show differences between two-legged, single-legged, and CMJs:
• Two-legged jumps at between 100% and 140% of CMJ height had a ground contact time of around 185ms. The two-legged jump at 160% had a noticeably longer ground contact time at 210 ms.
• One-legged jumps at all heights had a ground contact time longer than the two-legged jumps, at around 260ms for all heights.
• In terms of force production, the CMJ resulted in around 2100 newtons. Two two-legged hurdle jumps resulted in between 4000 and 4300 newtons, independent of the height (i.e. there was no pattern of increasing or decreasing force production with increasing heights). The single-legged jumps had a force production of around 2400 newtons.
• Rate of force production for the CMJ was around 4600 newtons/second. It decreased as height increased on the two-legged jumps, ranging from 41,000 N/s to 32,000 N/s. For the single-leg jumps the rate of force production was around 14,000 N/s.

It needs to be kept in mind that a specific population was studied and this could effect the results. However, the trends are important when selecting exercises for a strength and conditioning program:
• The shorter two-legged hurdle jumps had a short ground contact time, less than 200 m/s. The one-legged jumps had a longer ground contact time (around 260ms). So, when training for activities with a short ground-contact time (like sprinting), two-legged jumps may be the way to go.
• The two-legged hurdle jumps had a significantly greater force production and rate of force production that the one-legged or the CMJs. Again, this needs to be factored in when selecting exercises.
• The rate for force production was almost 300% greater when comparing two-legged jumps to one-legged and almost 900% greater than the CMJ. This brings into question the value of the CMJ as a conditioning exercise when rate of force development is important.

Cappa, D.F. and Behm, D.G. (2011). Training specificity of hurdle vs. countermovement jump training. Journal of Strength and Conditioning Research, 25(10): 2715-2720.

Luge is a sport contested in the winter Olympics where the athlete, after an explosive start, navigates a sled through an icy track. The start is critical to performance because it is the only time than an athlete can accelerate the sled. Typically luge athletes are trained with variations of the Olympic lifts (to develop power) and many upper body exercises to help with the start.

Crossland et al in the October issue of the Journal of Strength and Conditioning Research sought to look at the relationship between upper body performance measures and anthropometrics and the luge start. This study is extremely interesting for two reasons. First, there isn’t a lot of published research on the luge. Second, the subjects were members of the senior and junior U.S. national teams including several Olympians.

The normal height, weight, age information was recorded for the subjects. Their upper body anthropometric measurements (finger-tip span, biacromial width, hand length, acromio-radial distance, acromio-olecranon distance, sitting cervical height), their start time was measured, and the following upper-body strength measures were taken: bench press 1-RM, prone row 1-RM, weighted pull-up 1-RM, and number of pull-ups in 15 seconds.

In the senior group, all the upper-body strength testing measures were significantly correlated with start time (the correlation ranged from 0.69 for 15-second pull up test to .82 for prone row 1-RM). In terms of anthropometric measures; finger-tip span, biacromial width, acromio-olecranon distance, and height were all significantly correlated with start time (correlations ranged from 0.58 to 0.62).

In the junior group, only bench press and prone row 1-RM had statistically significant correlations with start time (0.76 each). In terms of anthropometrics, only the acromio-olecranon distance was significantly correlated with start time (0.74).

According to the authors, the differences between the senior and junior group reflect that fact that the luge start technique is very complicated and takes years to master. Once athletes master the technique, then outside factors play a larger role in performance. This is why more variables are impacting performance for the seniors than the juniors.

This study is a really good start in terms of factors that influence performance. It also reveals that certain anthropometric variables might predispose someone to be more successful at the luge start, which could help to determine athlete selection and recruitment though a lot more research is needed on this.

Crossland, B.W., Hartman, J.E., Kilgore, J.L., Hartman, M.J., and Kaus, J.M. (2011). Upper-body anthropometric and strength measures and their relationship to start time in elite luge athletes. Journal of Strength and Conditioning Research, 25(10): 2639-2644.

Ho et al, in the October issue of the Journal of Strength and Conditioning Research, published a study looking at the biomechanics of the snatch exercise. The snatch involves taking a wider than shoulder-width grip on the barbell and lifting if from the floor to overhead in one, continuous movement which typically takes 1-2 seconds. The barbell must be moved very fast during a successful snatch, usually around 1.6-2 meters/second and this exercise involves one of the highest power outputs seen in strength training exercises. For this reason it is widely used in strength and conditioning.

While it has a lot of benefits, it is also a highly technical lift and is not very forgiving of errors. The authors point out that most analysis of this lift focus on the path and velocity of the barbell. They hypothesized that a lifter’s joint kinematics may have an impact on the success or failure of the lift.

With this in mind, the authors studied a single, novice lifter (age 25, snatch 1-RM 80kg, bodyweight 74.8kg). The subject performed lifts with 75%, 81%, 88%, 94%, and 100% of 1-RM. These were done in order (light to heavy), though the number of repetitions at each lift was randomized. Each lift was filmed so that the subject’s joint angles could be analyzed. It was also recorded which lifts were successful and which were not.

The results are interesting, though not always statistically significant:
• Peak barbell velocity for successful lifts ranged from 2.35 m/sec (lightest weight) to 2.25 m/sec (heaviest weight). The velocity for successful lifts decreased until 94% of 1-RM, at which point it leveled off suggesting that there is a minimum effective barbell velocity.
• In the unsuccessful lifts, the barbell velocity ranged from approximately 2.2 m/sec (lightest weight) to 2.2 m/sec (heaviest weight). There are differences between the successful and unsuccessful barbell velocities, but these were not statistically significant.
• Looking at the right side of the body, there were small but statistically insignificant differences between the joint angles of successful and unsuccessful lifts. The only exception was at the ankle joint where about 2 degrees of extra dorsiflexion was present during the start of the unsuccessful lifts.
• Statistical modeling revealed that the hip and pelvis joint angles (at the start) were very important for ultimate success of the lift.

The authors conclude that the hip and pelvis angles are very important for ultimate success of the lift and discount the impact of barbell velocity. It’s true that, in this study, the differences between successful lift barbell velocities and unsuccessful lift barbell velocities was between .15 m/sec and .5 m/sec, but it needs to be kept in mind that this study only looked at one subject – meaning that definitive conclusions cannot be drawn based upon this.

Ho, K.W.L., Williams, M.D., Wilson, C.J., and Meehan, D.L. (2011). Using three-dimensional kinematics to identify feedback for the snatch: A case study. Journal of Strength and Conditioning Research, 25(10), 2773-2780.

There are a series of articles in the latest issue of New Studies in Athletics which take a biomechanical analysis of the throws during the 2009 IAAF World Championships. I’m not a throw’s coach, but the data that is presented is interesting for a strength and conditioning professional. The table below summarizes the top five performances in each event (except women’s shot put, which is not analyzed to this extent in the article).

A few trends to notice:
• The men’s javelin has the highest release velocity, the men’s shot put the lowest. Excepting the shot put, the release velocities are between 23 and almost 30 meters per second.
• The hammer has the highest angle of release, with the rest of the events being very close to each other at 35-36 degrees.
• In the discus and hammer, the men and women have similar release velocities.
• In all events, the angle of release is similar between men and women.

  Distance (m) Release Velocity (m/s) Angle of Release (deg)
Discus, Men




Discus, Women




Hammer, Men




Hammer, Women




Javelin, Men




Javelin, Women




Shot Put, Men




In terms of application of this information. An argument is beginning to be found in the literature against the existence of velocity specificity. One of the tenants of this argument is that it is not possible to perform strength training exercises that have a similar velocity of movement as the sporting event that they seek to train. While the Olympic lifts potentially generate 31 to 37 watts of power (Garhammer 1981), the bar velocity rarely exceed two to four meters/second – noticeable less than the 14-30 in the throws. This goes to underscore the fact that strength training is a “general” training tool for many athletic events and not a “specific” one (to borrow Matveyev’s language).

The angle of release for all the events is also interesting for the strength and conditioning practitioner. There is a debate about whether throwers should bench, incline, or simply push press/jerk. Given that strength training isn’t going to resemble the throws in terms of velocity, it seems that the incline press and push press or jerk would be more relevant to the throws.

Badura, M. (2010). Biomechanical analysis of the discus at the 2009 IAAF World Championships in athletics. New Studies in Athletics, 25(3/4): 23-35.

Garhammer, J. (1981) Force-velocity constraints and elastic energy utilization during multi-segment lifting/jumping activities, Medicine and Science in Sports and Exercise,
13(2): 96.

Isele, R. and Nicdorf, E. (2010). Biomechanical analysis of the hammer throw at the 2009 IAAF World Championships in athletics. New Studies in Athletics, 25(3/4): 37-60.

Lehmann, F. (2010). Biomechanical analysis of the javelin throw at the 2009 IAAF World Championships in athletics. New Studies in Athletics, 25(3/4): 61-77.

Schaa, W. (2010). Biomechanical analysis of the shot put at the 2009 IAAF World Championships in athletics. New Studies in Athletics, 25(3/4): 9-21.

Poulos et al, in the latest issue of New Studies in Athletics, look at postactivation potentiation and adolescent sprinting performance. As I’ve blogged about elsewhere ( , , ) the results of studies on this topic are very inconsistent and may have a lot to do with strength levels and level of athletic ability. This study reinforces some of that.

The authors studies eight adolescent track and field athletes (seven sprinters, one long jumper). Athletes had a personal best on the 100m ranging from 10.61 to 12.2 seconds. The study was organized as follows:
• Testing session: Back squat 3-RM (using IPF rules on squat form), 1-RM was estimated from the 3-RM.
• Treatment sessions: 50 meter sprint followed by 10 minutes rest, then 5 reps on the back squat followed by four minutes rest, followed by a 50 meter sprint. The squat + sprint was performed for four sets.
• There were three treatment sessions, each separated by a week. One involved the back squat being performed at 65% of 1-RM, one at 75%, one at 85% of 1-RM.

The results are interesting:
• There was no improvement in performance (and no reported decrease in performance) as a result of performing squats prior to sprinting.
• There was a strong relationship between the athlete’s relative 1-RM squat (i.e. their squat in terms of their bodyweight) and their 50 meter time. Basically the stronger athletes (relative to bodyweight) were faster.
• There was a strong relationship between the athlete’s relative 1-RM squat and their sprinting speed.

While the study did not find an effect from the squats, the information about the relationship between relative strength and sprinting performance in adolescents is very important. This reinforces the need for sprinters (and long jumpers) to be strong, at least to a point.

Some limitations need to be kept in mind. First, these athletes may be too young to benefit from this type of training. Second, the rest interval between the squats and the sprint may have been too short. Third, the volume or the intensity of the squats may have not been optimal to create a performance improvement. Finally, since this type of training doesn’t result in a decrease in performance it’s still a valid tool especially during the season when there is a need to be even more efficient with an athlete’s training time.

Poulos, N., Kuitunen, S., and Buchheit, M. (2010). Effect of preload squatting on sprint performance in adolescent athletes. New Studies in Athletics, 25(3/4): 95-103.

Kilduff et al in the September issue of the Journal of Strength and Conditioning Research looked at the impact of postactivation potentiation on swim sprint starts. To me this is a really interesting study because it is looking at something other than a vertical jump or a dryland sprint. The authors studied nine elite sprint swimmers (member of the British Sprint Development squad), all of whom had swim times that were within 5% of the national record in their respective events.

The authors studied several things. First, the athletes performed a set of three reps at 87% of 1-RM with their back squat. They then performed a counter-movement jump (CMJ) 15 seconds, 4, 8, 12, and 16 minutes after the squat. The idea being to determine optimum rest interval between the squat and the jump. Second, the subjects performed the 1x3x87% squat then performed a swimming block start to 15 meters. This was compared to the subjects performing the same start after their normal warm-up.

The results are interesting:
• First, in the jumping trails the amount of time between the squat and the jump had a significant effect on the power output from the jump and the height jumped. Essentially the 15 second separation resulted in diminished performance, four minutes resulted in a return to baseline performance, and eight minutes resulted in the highest power outputs and jump heights (all greater than baseline).
• Second, in the swimming trials the squat stimulus resulted in greater horizontal force and greater vertical force coming off the starting blocks, but this did not translate into a faster time swimming to 15 meters.

There are a number of interesting implications from this study, good and bad. First, under these experimental conditions using this group of subjects, eight minutes appears to be the optimum time between the heavy squat and the power movement that it is attempting to improve. Second, performing a heavy strength exercise prior to an explosive swimming start does not improve swimming performance to 15 meters even though force measures increased. Third, performing a heavy strength exercise prior to an explosive swimming start results in the same swimming performance as performing a traditional swimming warm-up. In other words, this approach may substitute for a traditional swimming warm-up.

This study has several limitations. First, it may be population-specific. Second, it may be situation-specific (i.e. swim to 15 meters), we don’t know if performance would have increased/decreased for an entire event as a result of this warm-up approach. Third, we are not given the time interval between the squat and the swimming start. This could have implications for performance and it is always possible that just because eight minutes was optimal for a vertical jump, a different time might have been better for an explosive swimming start.

Kilduff, L., Cunningham, D.J., Owen, N.J., West, D.J., Bracken, R.M., and Cook, C.J. (2011). Effect of postactivation potentiation on swimming starts in international sprint swimmers. Journal of Strength and Conditioning Research, 25(9), 2418-2423.

Andrews et al, in the September issue of the Journal of Strength and Conditioning Research, investigate complex training and whether the speed of the strength training exercise impacts the effectiveness of complex training. The idea being that by pairing a heavy strength training exercise with a plyometric exercise, the strength training will create something called “post activation potentiation (PAP)” which will enhance performance on the plyometric exercise. I have blogged about post activation potentiation and complex training ( elsewhere because the results are pretty mixed on this. The reality is that the effectiveness of complex training may depend upon the type of plyometric being employed and on the level of the athlete.

The authors studied 19 Division I and II collegiate athletes with a mean back squat of almost 150% of bodyweight and a mean hang clean of more than bodyweight. The study was organized around having the subjects perform several sets of counter-movement jumps (CMJs), having them perform 3x3x75% back squats paired with CMJs, and having them perform 3x3x60% hang power cleans paired with CMJs (each in a pairing in a different session). The idea was to see which had the greatest impact on the height of the CMJs.

The results show that fatigue impacts vertical jumping height. After the first set of exercise, the athletes were only able to CMJ at 96-99% of their best CMJ. From the first set to the third, the CMJ-only group and the back squat complex group both lost about 4% of their CMJ height. In contrast, the hang clean complex group only lost .5% of their CMJ height from the first set to the third.

The results suggest that if the desire is to minimize the effects of fatigue on CMJ height during complex training, the hang clean (i.e. a fast strength training exercise) may be more effective than performing the CMJ alone or pairing it with the back squat.

There are a number of limitations with this study. First, the results may be population specific. Second, it’s possible that the resistance employed was insufficient to generate a PAP effect. Third, there is not a real smoking gun in this study demonstrating that any protocol is more effective at increasing CMJ height. It needs to be kept in mind that this study is looking at the acute response to narrowly defined training stimuli and does not demonstrate long-term adaptations from this type of training.

Andrews, T.R., Mackey, T., Inkrott, T.A., Murray, S.R., Clark, I.E., and Pettitt, R.W. (2011). Effect of hang cleans or squats paired with countermovement vertical jumps on vertical displacement. Journal of Strength and Conditioning Research, 25(9), 2448-2452.