Sprint acceleration force-velocity mechanical profile versus split times: what added value?

This first summer post summarizes recent discussions we’ve had with Pierre Samozino(PhD, University of Savoy) and Martin Buchheit (PhD, Paris SG Football Club) around the practical interest of computing players’ sprint acceleration mechanical profile versus collecting short (0-5m) and long (flying 30-40m) split times. Martin and I thought interesting to write our respective arguments down as a “point-counterpoint”. So I will try to explain our points and “defend” our approach, although our discussion about these issues has always been friendly.
My aim here is not to “preach” or convince scientists or S&C pros to implement this approach. It is to bring some arguments and practical details to fully inform researchers and sport practitioners so they can make their own, enlightened decision. They can then decide among the possible, interesting, and important variables to monitor, from sand to big rocks…

Quick reminder about the sprint force-velocity-power (FVP) mechanical profile

What we refer to as “sprint acceleration mechanical profile” or FVP profile, is the integrative approach described in our recent IJSPP commentary, in which we quantify the horizontal net force output, velocity and mechanical power of athletes, and their effectiveness of ground force application. In addition to the IJSPP paper, the main references for this approach are: Morin etal. MSSE 2011 and Samozinoet al. SJMSS 2016. It has also been used in >10 recently published papers about sprint, soccer or rugby performance. Briefly, the muscular output of the athlete is described by an inverse linear relationship between the horizontal net force output and running velocity. This spectrum ranges from the theoretical maximal force (F0) to the theoretical maximal velocity (V0) an athlete is able to produce. These variables are produced at the very beginning and the very end of the acceleration phase, respectively, and determine the maximal power of the athlete in sprinting (Pmax): Pmax = F0.V0/4.
The FVP profile also includes the mechanical effectiveness of ground force application, which is described with two main variables: the ratio of force (RF) i.e. the ratio of the effective horizontal component of the GRF to the resultant GRF, and how quickly this ratio drops as the running velocity increases (decrease in the ratio of force, Drf). For instance, if an athlete as a RF of 30% during the first few meters of an acceleration, it means that 30% of his total ground force output is directed in the direction of motion (horizontally) and propels the body mass forward. An athlete with a 25% RF will be less efficient and vice versa. Then, a Drf of -5% (or -0.05) means that for each new 1 m/s of speed increase, RF decreases by 5%, until the end of the acceleration. At that point, top speed is reached, and the step-average angle of the ground force vector is vertical. In previous studies (Morin et al.EJAP 2012 and Rabitaet al. SJMSS 2015) we observed that the mechanical effectiveness of ground force application was a key factor in sprint acceleration and overall performance, from non-specialists to sub-10 sprinters. But what really counts for most team sports is what happens before top speed…

Point: “short and long splits bring sufficient information about force and velocity muscular capabilities, respectively”

As the head of performance and sport science in one of the world’s greatest football clubs (and renown sport physiologist), Martin’s points were as follows: 
“I didn’t find (personally) the need for such data when profiling players, so I still use 5 m as a proxy of horizontal force production (since it is the main determinant as per our 2014 JSS article) and max speed (flying 30-40m) as a measure of maximal horizontal velocity property. 
1. It is faster and I don’t have the resources to calculate the profile (spreadsheet or software)
2. Once you know the determinants of 5m and flying speed, what do you have in addition with the full profile?
Since you already know the mechanical determinants of the splits the profiling doesn’t add much in terms of training guidance and prescription”

Counterpoint: mechanical FVP profiling brings a significant added value

Here is a list of arguments to support the added value of the FVP profiling.
 
1. Not exactly the same information: sprint performance versus underpinning mechanical and muscular variables
Overall, we agree with Martin that 5-m splits and 30-40 flying splits are well correlated with F0 and V0, respectively. We consistently find such good correlations in our own research and consultancy practice. However, our simulations performed using the published equations show that two players may have the same 5-m splits and different FV profiles leading to different F0 (up to 10-15%) so same 5-m split, very different maximal horizontal force capability. We also found such players in the real life during our testing with elite rugby teams for instance. Interestingly, we observed that a long/short split ratio (20-m/5-m times) was well correlated with the slope of the linear F-V profile. Computing the FVP profile is a bit like doing a performance analysis on all the possible distances between 0 and 30-m and synthesizing it through 2 or 3 key variables only. In our opinion, considering the sprint performance output as an interchangeable information for the underlying neuromuscular properties is a bit like considering jump height is a good proxy for lower limbs power. It is overall roughly the case, but with substantial individual variability. In summary, we agree that timing splits is a more convenient analysis on the field, and gives overall a valuable yet limited (see point 3.) information. As we wrote in our 2014 JSS paper the F-V profile is interesting since it brings more information as to what causes performance than just splits, other details in our IJSPP commentary. But since we deal with Ferraris in the elite sports context, our tools should be as accurate as possible. Who uses the Cooper 12-min test as a proxy for VO2max in top-level athletes?
2. Not that complicated-expensive to measure and compute
Based on the equations published in our SJMSS paper, it is possible for sport science MScs or PhDs to design their own spreadsheets to compute the entire FVP dataset from 6 split times or velocity data (radar gun or laser devices). Furthermore, two recent solutions have been proposed by our colleagues to facilitate the implementation of our approach.
First, Pedro Jiménez-Reyes (PhD, UCAM, Murcia, Spain) has designed an iPhone/iPad application named MySprint. As seen in the promotional video below, this app allows easy and practical measurement of split times, and then runs the entire set of computations. We have no commercial relationship whatsoever with Pedro on this app, but our own practice and a paper under review show very high agreement between the split times measured with the app compared to both timing gates and radar devices. Our own comparisons clearly confirm those made by our Spanish colleagues in their submitted comparison study (see correlation below). Provided of course that the app instructions are thoroughly followed by the users…This app + the 240 fps device necessary (iPhone 6 or >, or iPad pro 9”7) is not a huge investment.
 
 
Another solution has been developed by Matt Brughelli (PhD, AUT, Auckland, NZ) with a software that instantaneously runs the entire set of computations directly from the radar raw files. We compared the software results to our own computations, perfect match. For all details, and a user license, contact Matt directly, his email appears at the end of this promotional video.
Of course we agree that all these solutions are more complex than reading split times, but we think the additional information is worth the additional processing energy.
Finally, we are available for professional education and workshops consultancy, to teach S&C staffs how to be autonomous and effectively implement this method, and add the FVP approach to their jumping/sprinting performance toolbox.
 
3. Split times do not differentiate between force production capability and the effectiveness of force application
This is in our opinion the most important argument (until point 4. Is finalized). Even if one considers 5-m split time as a proxy for maximal horizontal force production (F0), it does not tell anything as to the mechanical effectiveness of force application (maximal value of RF in particular for the first steps). As also detailed in our IJSPP commentary, two players may have the same F0 but very different effectiveness values. We frequently screen players and observe similar F0 yet with huge differences in initial acceleration RF, meaning that some players have good effectiveness with overall low strength (like the French sprinter Christophe Lemaitre in our 2012 EJAP paper), whereas some others on the contrary have good strength capabilities, but low RF. If you want to specifically and effectively train each of these players, then you should target different features of their FVP profile. And give each players the training input he/she needs. You know these needs in details when using the FVP profile, you have no clue from just analyzing the 5-m splits. Note that training studies showing how to specifically target the various components of the FVP profile (F0, RF, V0, Pmax…) have been submitted. We hope that these clear practical applications will be published shortly. Some have already been presented at the ECSS congress, such as MattCross’ recommendations on optimal loading for maximal power in resisted sprint training (see his slides here). We also recently submitted a controlled pilot study in which we show that amateur soccer players improve their F0 via a better RF using very heavy sled training (80% body mass) compared to controls training with no resistance.
 
4. Important update in progress: individual optimal profile for sprint acceleration performance
Speaking of individualized training, our jumping studies clearly show the superiority of a F-V imbalance individualized training compared to a one-size-fits-all training program (see Pedro Jimenez ECSS slides, paper submitted soon). The optimal FV profile is computed using the typical inputs needed to draw the actual FV profile, and the imbalance between the two gives the magnitude and direction of the training program (rather force- or velocity-oriented, see our IJSPP commentary of Pedro’s slides for more details). This very promising individualized approach for jumping performance has not been transferred to sprinting yet. The only possibility for now is to compare players to themselves over time or to other players within or between teams/groups/sports/levels. This is already interesting…but soon we’ll show the existence and usefulness of a sprint optimal profile (exactly like in jumping) and yes then the training objective will be individualized. The approach will give some clues about who needs what in terms of speed and/or force and/or mechanical effectiveness to run faster on a given (target) distance of acceleration. Equations are finalized, and Pierre Samozino will start writing a draft soon. This process could be faster should French university professors not have to teach >200h per year!! 
– whining mode off –
This computation of the sprint individual optimal profile (as per jumping performance) will be done through the F-V profile equations, impossible with 2 splits only.
 
 
Finally, although we do not consider it as a key argument (some high-level professionals sometimes use crappy devices/techniques/approaches on a daily basis), this approach makes sense to an increasing number of athletics, rugby or soccer coaches and staffs around the world. So we are happy to see that the approach is used (that’s what we work for), but we are also ok with the fact that many other elite professionals do not feel like it is a useful resource for them. 
 
To conclude this post, our opinion is that despite some necessary data collection/processing we fully acknowledge, the sprint FVP approach brings a more informative picture than split times only. We are working now on providing some facilitated processing techniques, personalized professional education to implement them correctly, and an improved analysis with the individual sprint optimal profile. Stay tuned…
 
 
All references are available on my ResearchGate pages.
 

Next post: the impressive physical capabilities of Master athletes…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close