Improving acceleration performance in football players

This work was commissioned by the English Football Association as part of their strategy to improve the physical capability and reduce the injury vulnerability through profiling English national team footballers. All references used appear at the end of the post.

I would like to thank Dr Ben Rosenblatt for his help in editing the draft of this article.

Introduction

Unlike the 100-m dash, football is not about who creates the highest linear running velocity during a single effort in a sprint lane, and maintains it towards the finish line. Most of the time, football is about changing velocity, i.e. accelerating one’s body mass. Starting from various velocities (rarely standing still) and positions, a player has to produce high amounts of “pure acceleration” to leave an opponent behind (offense), get the ball/position first, or catch-up with an opponent (defence). Other actions require the production of “mixed acceleration” (negative-positive) during changes of direction, cutting manoeuvres, or even vertically-oriented acceleration in the case of single- or double-leg jumps.
This article focuses on understanding the biomechanical determinants of sprint acceleration in football, how to evaluate them accurately in field conditions, and how to improve them on an individual basis. This may eventually help improve players’ physical performance, and better managing sprint related injuries. Note that our view of player’s performance and readiness/availability does not distinguish between physical preparation and injury prevention/rehabilitation. Seeking to improve players’ performance includes, de facto, injury management (primary, secondary prevention and rehabilitation post-injury)

1 – Understand: produce and transmit force to the ground…under constraints

A payer’s body accelerates by interacting with the ground. The “prime mover” is the lower limb neuromuscular machinery (pelvis, glutes, thigh, lower leg and foot muscles), but acceleration occurs as the result of the ground (playing surface) reacting to the player’s pushing actions. The driving and propulsive force is the ground reaction force (GRF). The more you push into the ground, the higher the GRF magnitude. Thus, accelerating and sprinting fast is about (i) producing force, but also (ii) transmitting it to the supporting ground in a mechanically efficient manner, so that (iii) the resulting GRF will propel the player’s body mass where and how quickly the player wants to. Sir Isaac Newton’s fundamental law of motion overall states that the acceleration is proportional to the sum of the external forces applied to a mass, and inversely proportional to that mass. In football, it means that maximizing acceleration means producing high amounts of GRF, per unit of body mass. However, the second (too often overlooked) part of this law states that the direction of the acceleration is the direction of the force. In football, it means that the direction of the acceleration depends on the orientation of the GRF.So it is not only how much force you can produce, it is also how you can orient it in the direction you want your body to accelerate.
Football players, like sprinters, also face two other major constraints during their sport-specific force production to accelerate: contact time and velocity of motion.
During sprint acceleration, the foot is in contact with the ground for 100 to 200 ms (1 to 2 tenths of a second). By definition, propulsive actions and GRF application can only occur during this very short time frame. Thus, no matter how much absolute force output a player is capable of. On the pitch, it is about how much GRF is produced within this short contact duration. This has major consequences on training, since explosive force output (the force that can be produced within the first ms of a maximal voluntary effort) rather than absolute maximal force (which may require much more time to be reached) should be the main objective.

The second sport-specific constraint is the velocity of motion at which the GRF is produced. One major physiological feature of skeletal muscle physiology is that the level of force output depends on the shortening velocity, as evidenced by English Nobel Laureate Archibald V. Hill. The slower the contraction velocity, the higher the possible force output, and vice versa. In practice, during sprint acceleration, there is an individual linear relationship (the “force-velocity relationship”) between a player’s GRF output in the direction of motion and their running velocity. For example, if two players are able to produce the same GRF while jogging at 2 m/s, they may have totally different capabilities when accelerating from a 5 m/s run, etc… And they also may totally differ when tested for maximal force output during very slow, gym-based exercises like a back-squat or a leg press. From the above described constraints of football-specific motion, it is clear that there is a major gap between isolated, single muscle or lower limb maximal force output (as measured in the gym through 1RM or isokinetic tests) and sprint/football-specific force output.

In summary, how muscle force output will “transfer” to football-specific acceleration depends on:

–      The velocity of motion
–      The forward orientation of the ground reaction force produced
–      The contact phase duration

The best athletes (sprinters, rugby, football players) in terms of sport-specific accelerations are not those with the greatest gym-based testing numbers, but those who are able to produce the highest amounts of GRF, in the horizontal direction, during 100-200 ms contact phases, at already moderate-to-high running velocity.

In other words, this sport-specific context implies that the players with the greatest isolated muscle group or lower limb strength capabilities at low velocity of motion and/or during longer, two-leg exercises will not necessarily be the best at producing and applying force effectively during a sprint acceleration.
The key is thus how to evaluate this sprint-specific force output and how it is oriented when applied onto the ground.

2 – Evaluate: from the laboratory to the field

Although innovations are still needed to evaluate sprint GRF during real soccer actions, recent research has made on-field assessment possible, that where hitherto only performed in laboratory setting. Research so far has presented the aforementioned concepts and data using instrumented sprint treadmills, and sprint tracks equipped with force plate systems. For each step of a linear acceleration, GRF is measured and the vertical and horizontal components computed, so that the ability of the athlete to (i) produce and (ii) orient the GRF vector forward is measured accurately, along with increasing running velocity.
The issue was that all in all, three or four labs in the world could produce this kind of comprehensive analysis, and to date no specific data have been published on high-level football players. The good news is that our research group group has worked in the recent years to propose a field method based on at least 3 split times (10, 20 and 30-m) or velocity measurements that allow accurate computations from much more accessible inputs. This method is based on the laws of dynamics and has been validated against force plate reference systems. Basically, the practical and cost-time-effective aspects of this macroscopic approach far outweighs the inevitable slight loss in accuracy. We also published a spreadsheet and tutorial to run the entire analysis from only three to five split times, and get the results directly. Note that an iPhone and iPad app named “MySprint” also runs these computations—all you need is athletes to run a 0-to-30m all-out sprint, and to record the sprint with the iOS device (slow motion at 120 to 240 frames per second). 

Since the assessment and subsequent computations only require players to give their best effort over a 30-m, it becomes very practical to accurately know the players’ individual maximal force, velocity, power output specific to sprint acceleration, but also their force output at any given velocity or distance (through the force-velocity profile), and how horizontally they orient their push throughout the sprint, from early steps to maximal velocity.

Finally, using the same slow motion video approach, another app based on our research (named “Runmatic”) allows accurate quantification of running contact time and GRF output during high-speed treadmill sessions. Thus, added to the standard, non-specific, gym-based assessment of “absolute” strength capabilities, a much deeper analysis of each player’s force-velocity-power profile (strengths and weaknesses) is now possible, at any point of the season: pre-season, post-congested schedule, before-after seasonal break or international games, post-injury (hence the importance of pre-injury data collection… i.e.regular monitoring), during and post-rehabilitation, prior to clearing the player for return-to-play, etc…
Of course, the main aim of this individual profiling is to tailor and perform more individual, thus efficient, training programs. Physical preparation in high-level team sport should aim for “collective individualization”.

3 – Improve: towards a really individual approach

Since football players on a same given team or group almost never show the same sprint mechanical profiles (many combinations are possible from the above-listed mechanical determinants of sprint acceleration performance) we think it is impossible to provide efficient strength and conditioning programs in football that do not include, at least partly, different training stimuli. Providing the same program (exercises, loads, velocity of motion, targeted mechanical features) for developing acceleration performance to a group of team sport players would mean that they allshare the samemechanical features. It is obviously not the case. We don not want the team to sprint faster on average, but each single player.

In this section we provide some evidence-based tracks of improvement of the sprint acceleration capability, within the specific mechanical context of football actions: acceleration capability is proportional to force divided by body mass, depends on the running velocity, on the amount of GRF produced within the very short contact duration, and on the overall orientation of the GRF vector (more horizontally-oriented is mechanically more effective). The following sections discuss key points that should be taken into account when designing individual training programs, allowing to progress from non-specific to sprint-specific force capability.

Body mass: avoid the “stronger-heavier-slower” cycle
Football definitely requires a certain amount of lower limb maximal absolute strength. There is no doubt it is useful when facing contacts, loss of balance, or overall for a better “toughness” and resistance to stress and strains induced by training and competition, including some injury mechanisms. That being said, in the context of acceleration performance, improving players’ strength must absolutely be considered within the context of the associated gain in muscle mass. Not to mention fat mass, of course. If the objective is to develop acceleration performance only, it is crucial to consider force development methods that are not associated with an increase in muscle mass, and to carefully monitor the training-induced changes in both force and body mass.  As explained before, acceleration depends on the ratio of force output to body mass. Consequently, if the relative improvements in strength are smaller than the associated change in body mass, stronger, yet heavier players, may also become slower. This phenomenon may be even worst if, as we will see now, the strength work led to non-sprint-specific force improvement.

Running velocity: what force at what velocity?
A “strong” player is usually seen as “a player with a high maximal force or 1RM capability”. This is confusing, since force capability depends on the velocity of motion. The question “is this player strong?” must be answered with “it depends on the velocity”. Some sub-10s sprinters are not strong in the sense of gym-based 1RM numbers, but they can produce more GRF than anybody else when their body is moving at 8-10 m/s. This is key since after only 2 or 3 steps, the running velocity of a football player who started to sprint from a standing still position is already 4 to 5 m/s…way faster than during any gym-based exercise. This is a conservative estimation since most accelerations in football start with an initial velocity that ranges from jogging to high-velocity pace. Always remember that stronger athletes in a low-velocity context are not necessarily stronger in a very-high-velocity context, and that, in trained athletes, the “transfer” of improvement in maximal strength (at low-velocity) to sprint-specific strength (high-velocity) is far from certain.
The sprint force-velocity profile allows to know exactly what force a player produces (be it the overall GRF or the specific horizontal component of this GRF) throughout their entire, individual running velocity spectrum (from zero to their individual maximal speed). Thus, inter-players comparison may easily show what player needs what kind of training stimuli, to improve what specific part of the profile. We have heaps of examples of players with similar 30-m sprint performance and very different (sometimes opposite) force-velocity profiles, or players with different 30-m performance resulting from very different “weak points” in their profile.

Training exercises, loads and programs should thus be designed with respect to this individual force-velocity profile. For example, some pilot results we obtained in football players show that progressive, well designed, very heavy sled work (80% body mass or higher) led to improvements in the mechanical features on the force end of the profile: maximal horizontal GRF, maximal forward orientation of the GRF vector. Alternatively, very heavy sled may not be an efficient training stimulus to develop the middle part of the profile (intermediate velocity, i.e.maximal power zone), and likely inefficient to develop force output at the very high velocity end of the profile. Here, high-speed and “overspeed” training may logically be efficient training stimuli, which is what our current research is testing.
Just as the assessment and interpretation of the force-velocity profile should be player specific, the training solutions should be specific and depend on the training objective. A lot of “it depends”… but that’s what elite sport performance is all about.

Contact time: as strong as possible…within 100 to 200 ms
No matter how much maximal, absolute, force a player is able to produce during a two-leg, gym-based test (not to mention isokinetic single joint torque tests), what counts on the pitch, and what will accelerate their body mass is the amount of GRF they are able to produce during a single-leg contact of roughly 100 to 200 ms. Again, no systematic direct correspondence exists in trained athletes between “explosive force” and “maximal force”. So specific training programs should absolutely take this time constraint into account.
Plyometrics, rebounds (preferably horizontally-oriented) and other explosive lower limb actions are interesting stimuli, provided that the general rule is “high force within short support time”. It is not either high force or short contact time, it is both high force and short contact time.
Very interestingly, no strength and conditioning exercise is better than top-speed running to put the players in this high-GRF-low-contact-time context. Added to the potentially preventive feature of regular, optimally dosed top-speed exposure, maximal velocity work is very likely a recommendable stimulus for this specific mechanical capability.

Ground force vector orientation: accelerate your body forward
Some football actions require to jump and accelerate in the vertical or oblique direction, but this is beyond the scope of this article. In the case of sprint acceleration and change of direction manoeuvers, the effective force is the component of the GRF that is directed horizontally. Note that the vertical component is not useless; it allows the body to remain balanced and standing. But as much as possible, an effective GRF orientation should be made in the direction of the targeted motion. All exercises (bounds, resisted sled push, pull, broad jumps) should follow this “as horizontally as possible” rule. Here again, very heavy sled push or pull are an efficient training stimulus to improve the players’ ability to orient their GRF forward, from the beginning of the sprint (maximal effectiveness) and as running velocity increases (limit the decrease in effectiveness with increasing velocity).
One possible explanatory mechanism may be a specific overload of the hip extensor muscles (glutes and hamstring), that have been related to an effective GRF orientation. On this point, experimental evidence confirmed the logical anatomical and mechanical analysis showing that a rapid and powerful hip extension before and during the sprint running stance is associated with a high horizontal (backwards) force production (thus forward propulsion of the body). This is especially the case when the player is already running, in an upright body posture, as it is almost systematically the case in football.
High loads will also allow a more inclined position of the body during the sprint acceleration push, and also keep this inclined position longer over the acceleration, so more time is spent applying very horizontally oriented ground force (which is not possible with lighter loads).
Practice and observations also show that the work done at the ankle and foot to transmit the power generated by the lower limb is huge in heavy sled conditions compared to lighter loads.

This also contributes to improved “technique” through less energy dissipation at the ankle and more energy transferred to the ground. This is key; whatever your lower limb power generation capability, if your ankle-foot system is not able to transmit that power output into the ground and it “deforms” under tension, this impairs your technique—and by extension, your acceleration performance. We currently have no consistent experimental results to support this view of things but our practice shows that heavy sled work “magnifies” this foot-ankle weakness, that we don’t evidently observe with lighter loads.
For overall weaker players, the above-mentioned points still apply, but in addition, the overload generated by heavy sleds may potentially add to lower limb strength (and horizontal ground reaction force in particular) in both absolute and sprint-specific terms.

 

One final note concerns the longer contact times induced when sprinting against high resistances. This seems counterproductive in light of the previous section, but an efficient training program must include all the key mechanical features and their associated stimuli, while ensuring that the long-term, overall “net” effects are positive. We are confident in the fact that the very heavy resistance work described in this section, if appropriately balanced with the short-contact/high-GRF work described before, may be part of an effective program. In addition, current research aims at testing the hypothesis that very-high resistance sprint training induces an increase in ankle stiffness, which is very likely a key feature of the ability to produce high amounts of GRF over short contact time periods.

Conclusion

The individual, comprehensive analysis of the players’ mechanical profile is quite young, we moved from split-time based testing of acceleration capability to the more in-depth profiling. Basically a split time is a very poor level of information, and only tells you a player is “slow”, “medium” or “fast” over a given distance. It does not give insights into the mechanical reasons explaining why, nor into the possible margins of improvement. Since a chain is only as strong as its weakest link, the individual profiling approach proposed here will help identify each player’s strength and weaknesses, and address performance “leaks”, while monitoring both progresses and performance “assets”. Future research should try to design and test more specific and individually tailored training programs and studies, instead of just giving X or Y common training regimen to a group of players and studying the group response. All the necessary methods are now easily available.

Last but not least, you may have the highest muscular power and the strongest hamstrings-glutei-quads in the world, you will not perform well and/or be at risk if you don’t have an efficient and safe sprint “technique” and overall movement pattern (pelvis-trunk control and tilt, knee-ankle position during swing and stance, etc…).  It is not only about how fast you can run, but also how you run…But that’s an other topic, and the aim of our current research. In Spiderman, Uncle Ben says to Peter “With great power comes great responsibility”…you should see sprint mechanics in football players this way: “with great power comes great movement pattern / stride mechanics responsibility”.

REFERENCES

This paper was initially written as a research practical synthesis, without systematic reference citations. All points made are based on published evidence, non-referenced opinions are stated as such. All references, articles and spreadsheets mentioned are freely available within this Website. See other blog posts and “Publications” sections

Some of these references are listed below:

Buchheit M et al. Mechanical determinants of acceleration and maximal sprinting speed in highly trained young soccer players. J Sports Sci. 2014 32(20):1906-1913.
Cross MR et al. Training at maximal power in resisted sprinting: Optimal load determination methodology and pilot results in team sport athletes. PLoS One. 2018 13(4):e0195477.
Cross MR et al. Optimal Loading for Maximizing Power During Sled-Resisted Sprinting. Int J Sports Physiol Perform. 2017 12(8):1069-1077.
Cross MR et al. Methods of Power-Force-Velocity Profiling During Sprint Running: A Narrative Review. Sports Med. 2017 47(7):1255-1269.
Mendiguchia J et al. Field monitoring of sprinting power-force-velocity profile before, during and after hamstring injury: two case reports. J Sports Sci. 2016 34(6):535-41.
Mendiguchia J et al. Effects of hamstring-emphasized neuromuscular training on strength and sprinting mechanics in football players. Scand J Med Sci Sports. 2015 25(6):e621-9.
Mendiguchia J et al. Progression of mechanical properties during on-field sprint running after returning to sports from a hamstring muscle injury in soccer players. Int J Sports Med. 2014 35(8):690-5.
Morin JB et al. Very-Heavy Sled Training for Improving Horizontal-Force Output in Soccer Players. Int J Sports Physiol Perform. 2017 12(6):840-844.
Morin JB and Samozino P.  Interpreting Power-Force-Velocity Profiles for Individualized and Specific Training. Int J Sports Physiol Perform. 2016 11(2):267-72.
Morin JB et al. Sprint Acceleration Mechanics: The Major Role of Hamstrings in Horizontal Force Production. Front Physiol. 2015 24;6:404.
Morin JB et al. Technical ability of force application as a determinant factor of sprint performance. Med Sci Sports Exerc. 2011 43(9):1680-8.
Jiménez-Reyes P et al. Effectiveness of an Individualized Training Based on Force-Velocity Profiling during Jumping. Front Physiol. 2017 9;7:677.
Rabita G et al. Sprint mechanics in world-class athletes: a new insight into the limits of human locomotion. Scand J Med Sci Sports. 2015 25(5):583-94.
Samozino P et al. A simple method for measuring power, force, velocity properties, and mechanical effectiveness in sprint running. Scand J Med Sci Sports. 2016 26(6):648-58.
Samozino P et al. Optimal force-velocity profile in ballistic movements-altius: citius or fortius?. Med Sci Sports Exerc. 2012 44(2):313-22.
Samozino P et al. A simple method for measuring force, velocity and power output during squat jump. J Biomech. 2008 41(14):2940-5.