REACTIVE STRENGTH INDEX Q&A:
Coach Conversations with Eamonn Flanagan
By Eamonn Flanagan
November 21, 2018
Since publishing the Reactive Strength Revisited article series, I’ve had some great conversations with coaches from around the world, across a range of sports. These conversations always provide alternative viewpoints, challenge current thinking and allow for the sharing of knowledge and ideas.
I’ve collected some of the best conversations I’ve had to help others learn more about this area. While the Reactive Strength Revisited article series outlined best practice of reactive strength measurement and training in line with the scientific literature, this Q&A is a more informal affair. The discussions here are conversational and at times “informed guesswork” rather than fully supported by the scientific literature. The Q&A below can be considered an “opinion piece” and outline my best interpretations of other coaches’ questions and their specific situations.
Every context is different and in many cases there is no gold-standard or “optimal” way of doing things – there’s simply the best “fit” for your context and your athlete group. Hopefully this Q&A demonstrates that and gives readers an opportunity to explore different ways of doing things that might be a good fit for their own individual context. In many cases, the questions below are more insightful than my answers. Thanks to all the coaches who have shared questions and ideas over the years.
We are looking at rolling out some reactive strength testing on our NBA guys, probably through repetitive jumps, possibly the 10-to-5 protocol. I know you are a fan of the test - what are the advantages of that test? Is there any reason you wouldn’t do it for a time period (let’s say 15 seconds) instead of 10 hops specifically? How would you assess the decrement in performance across the 10 hops of the test?
Sports Scientist at NBA Franchise
I’ve used repeat hopping as a way to test for reactive strength for some time. However, since Damien Harper developed and validated the 10-to-5 test, that’s helped tidy up my testing protocols. I’ve written about the specifics of the protocol here. The reason I like the test, compared to drop jumps is that with drop jumps in the busy training environment it can become a little “one size fits all” in terms of the drop height used across groups. If you have a particularly weak athlete, a fixed size dropping height may not be appropriate for them. If you have any athlete who is heavily fatigued, their usual “appropriate” drop height may not be appropriate on a day they are fatigued. In both these instances, you might not find out that the eccentric loading is too great for these athletes until it’s too late. I like repeat hopping because the eccentric loading is directly dictated by how strong you are or how high you can hop on that day. If you’re not hopping high because you are weak or fatigued, then your eccentric loading is not going to be very high. So it’s a nice way to get a safe snapshot of an athlete’s reactive ability.
That’s not to be dismissive of drop jumps. They remain an incredible tool to overload the eccentric strength capabilities of athletes and get a depth of information from tools such as force plates.
In the 10-to-5 hop test, there is not going to be a performance decrement across the 10 reps however. It’s not a fatigue test. In fact, athletes often get better across the initial hops. My gut feel is that athletes might need a certain number of reps to modulate their own lower limb stiffness in relation to the surface stiffness. Humans adjust their leg stiffness to depending on surface stiffness1. An athlete might need to use their initial jumps to sub-consciously adjust their “springiness” to get the most out of the surface they are on and get the highest scores possible from the test.
While I haven’t seen performance decrement in the 10-hop protocol, bear in mind that if you use a time period protocol, such as 15 seconds, the athletes will do a lot more total reps. It could be anything from 15-25 hops depending on contact times and jump height. A 15s protocol could double the number of repetitions. In this case you will likely start to see a decrement in reactive strength performance across time. Research on reactive strength endurance suggested that reactive strength will probably decay by around 15-20% in such a time period . So what do you want to get out of the test? If it is purely an assessment of maximal reactive strength, then run with 10-to-5 protocol. If you want to assess reactive strength endurance then use a longer time period of 15-30s depending on the sport and the athlete’s condition.
There are some other considerations with the 10-to-5 test. You may want to constrain contact time beyond the traditional 250ms fast stretch shortening threshold. I typically use that threshold but there are pros and cons to this. If athletes contact times are longer than 250ms and they are looking compliant at the knees and hips, then I would actively coach them to be stiffer through the leg spring and to get those contact times below the threshold. However, in most cases, I don’t give much more instruction than that. What I get out of that testing is a snapshot of what the default strategy is for each athlete. Is an athlete contact time dominant or jump height dominant? Two athletes could produce the same RSI but one might do it from a very short, elastic contact time of 150ms while another might utilize a longer 220ms contact time. That gives some indication as to which athletes might be more on the elastic side of the spectrum and who might be more strength dominant.
Another, reasonable method of testing would be to put a tighter contact time constraint on the test. If the reason you want to do the test is to get more insight into ankle function and you are dealing with athletes where top speed is key (which involves very short contact times) then you might want to apply a criteria for the test of 150ms or less. The shorter you go on the contact times the more you are emphasizing the ankle or the Achilles-calf complex specifically and the closer it is to the contact times observed in sprinting (80-100ms in sprinters, 100-120ms in field sport athletes). With a 150ms threshold, there just isn’t enough time to flex and extend the knees and the hips to the same extent. This shorter contact time also means you are probably observing a greater contribution from the elastic component. That type of testing could sit well alongside countermovement jump testing. You are getting a clear delineation between your ankle dominant, elastic focused testing (10-to-5 hop test) and your knee and hip dominant, contractile element focused testing (CMJ for maximal height). Ryu Nagahara showed that reactive strength as assessed via ankle dominant hops with short contact times (125-140ms) was closely related to late stage acceleration performance, where contact times are shorter and force application is more vertical in nature .
What are your thoughts on unilateral RSI measures? I am attempting to put together some objective exit markers for guys returning from calf or ankle injuries. We have been using a single leg 5 max hop and have been using asymmetry as our marker of readiness, however it would be good to have an external bench mark. Do you have any recommendations for RSI measures for unilateral exercises?
Strength & Conditioning Coach at an Australian Rugby Union Franchise
Although I did publish a table of “normative data” for reactive strength, I would approach these with caution. Many factors can influence the reactive strength score achieved such as type of test, RSI calculation method, warm-up protocol, instructional cueing and surface stiffness. These "thresholds" might look very different across different surfaces and different tests. Ideally, organisations will develop their own thresholds over time. In team sports, with large athlete numbers and high player turnover, it can be easily done.
I don't have any specific thresholds/recommendations for single leg hop tests. I have used a single leg version of the 10-to-5 test. I think it is important to look at the absolute reactive strength values as much as the athlete’s symmetry. For me, there are 4 possible scenarios for an athlete following this type of testing:
Weak and asymmetrical
Weak and symmetrical
Strong and asymmetrical
Strong and symmetrical
I use this protocol with some athletes who have large asymmetries (for example, Paralympic athletes with unilateral disability). Over time we sometimes see a change in strength (increase in RSI) but also an increase in asymmetry. In these cases, both limbs have increased their strength but the uninvolved limb has increased to a greater extent. This isn’t a bad thing in my opinion. We have affected a positive physical adaptation. In many cases, I would rather a strength increase in both legs, even if it means a small increase in asymmetry. For me, the most important thing is to build up robust profiles of the athletes when they are at their best or in a "good moment”. Regular monitoring over time means that you can cross references physical development alongside successful competition performance. This then becomes their individualized reference point for their return to competition post-injury. Many healthy, successful athletes will exhibit "tolerable" asymmetry from day-to-day.
Another note on single leg testing, is that single leg tests will often be “noisier” than bilateral testing. With a greater balance and control component, there can be more variability from rep to rep. This doesn’t mean the test is unreliable. It simply means that you need to account for this variability when assessing “meaningful change”. I have typically found that the coefficient of variance for RSI in single leg hopping can range from 4-12% (compared to 2-8% for bilateral hopping). So unless the measured change is beyond the athlete’s own coefficient of variance, we may not be looking at worthwhile change. Try to factor this into your assessment of progression.
If you are looking at ankle issues in particular, you may also want to consider "control" or stabilization as the ankle is a complex, multi-directional joint (or number of joints). For example, an athlete might exhibit symmetry in RSI between legs, but they may have a proprioceptive deficit on the involved side. They may hop high and fast, but they might be moving around from hop to hop, rather than "on the spot". Left vs right might produce similar jump heights, contact times and reactive strength indices but one limb might cover a greater "volume" of space across the 5 hops which might indicate poorer stabilization or control. Some testing systems can measure this “drift”. Accounting for this (even if just observing by eye) is important.
Can reactive strength be significantly improved? For example, is it be realistic for an athlete to improve their RSI by a magnitude of 0.5 – 1.0 units over the course of a year or so? Can reactive strength can be significantly improved without improving other strength qualities?
Strength & Conditioning Coach at a Private Training Facility
Yes, it can be trained and improved. There is plenty of research showing that everything from plyometric work, to high intensity isometrics and maximal strength work might increase reactive strength or highly related qualities. How well this transfers to sports specific skills is still unclear. In relatively untrained athletes (<3 years strength and power training), or those with low plyometric experience or significant reactive strength deficits, I have certainly seen improvements of 0.5 in 18-24 month periods. As coaches, we should constantly be asking “how strong is strong enough”. The principle of diminishing returns dictates that we should always be considering if the amount of work needed to develop a physical quality will actually contribute to a meaningful performance improvement.
All physical qualities are intertwined to some extent. I suspect reactive strength is optimised when complementary strength qualities are developed in tandem. Kris Beattie examined the relationship between maximal-strength and reactive-strength and showed they are likely related . Strong athletes exhibited greater reactive strength indices in drop jumping, particularly in drop jumps from large dropping heights (50-60cm) where eccentric loading is greatest. It is likely that a degree of maximal strength will be beneficial to producing higher reactive strength but this is not equivocal.
I have typically calculated RSI as jump height divided by contact time. However, it can be argued that calculating RSI as a quotient of flight time to contact time may be more appropriate as this produces a unit-less metric. In the practical setting, it’s my thought that the jump height method is a little more intuitive for athletes to comprehend.However, the flight time method may be more empirically valid. Ultimately practitioners should be aware that both methods will provide very different absolute values and aren’t comparable. The flight time method will provide greater RSI values. Both methods will be perfectly correlated as jump height on contact mats or timing gate systems is derived directly from flight time. For more depth on this topic, Robin Healy has done an excellent investigation highlighting the differences in the calculation techniques .
Our basketball program are looking at developing a reactive strength based readiness marker (via drop jumps) and also want to adapt training in terms of exercise selection in the off-season and pre-season. What is worth tracking across the season and how can RSI testing inform exercise selection?
Director of Strength & Conditioning at NCAA Division One College
If you are using RSI as a readiness marker then I would take the time to ensure each athlete has their own well-established baseline (maybe 6-8 weeks of baseline data) and then I would track their weekly RSI vs a moving average of their last 4-6 weeks. An idea is to develop a traffic light system using effect sizes of 0.6 and 1.2 of the athlete’s own standard deviation (across the baseline period) to denote moderate or large decrease from the baseline.
It’s always very difficult to comment without really knowing the overall program and what the goals of the program are. I would say is that if you have individuals with low RSI scores (let’s say <2.0) from 40cm box then I would be cautious in using high impact plyos with these athletes. I would focus on more remedial, moderate intensity plyometrics done in more extensive fashion. This would be plyometric work done with lower intensities of effort, a greater coaching focus on posture and foot position and for higher than typical plyometric training volumes. Regarding acute adjustment of program based off weekly testing, I would be looking at dialling back high intensity, fast velocity activities (high CNS load) when we think RSI is suppressed to a large extent.
A few other points on RSI monitoring:
Surface matters. Change in surface will affect scores. Within our own facility we have three different surfaces that result in measurably different reactive strength scores.
Warm-up matters. It needs to kept as standardised as possible. This is not always easy or practical, but it can help tidy up the data.
"Intent" matters. Athletes have to want to hit good scores. Are they dialled into the test? Are they going to give 100% effort? A slight dialling off in intent can have a big effect on scores.
Look at contact times, not just RSI. Some athletes might maintain RSI but may be shifting to slower CTs and high jump heights. For example, you might see this in training phases with a large maximal strength emphasis alongside very little plyometric and speed activity.
I’ve been collecting reactive strength data from our triathlon program. An important physical quality for running efficiency is stiffness around ankles/knees/hips and reactive strength. I’ve recently started measuring RSI weekly (via drop jumps) to monitor improvements in reactive strength and to inform strength and plyometric programming. My thought was to increase the intensity of plyometrics for those that can consistently produce an RSI of over 2.0 and have strength levels of approximately 1.3 – 1.5 times body mass in back squat.
Strength and Conditioning Coach at a National Institute of Sport
I think your idea to have RSI and maximal strength thresholds before progressing plyometric work is valid. The threshold of >2.0 might be a little high for an endurance population but the principle is a good one – don’t progress them until they are demonstrating a reasonable strength level.
One thing you should consider is whether or not Maximal RSI is most important for this group. Which is more important... an RSI of 2.5 or an ability to maintain an RSI of 1.8 for rep after rep with good posture and position? Perhaps Maximal RSI and Reactive Strength Endurance might be quite different things. If you have a large group of athletes, it might be worth assessing the correlation between RSI and one of their key running markers such as economy. Is there a relationship? If so, what does it look like? Is there an inflection point at which the relationship becomes stronger or weaker? If so, this might offer clues to “how strong is strong enough” with respect to reactive strength for this population.
Think about your population critically and try to assess if there is a minimum threshold you want them to meet, but also a max threshold beyond which you suspect their might be diminishing performance returns. Personally, I am not sure if putting in a lot of effort to get a 2.1 RSI up to a 2.4 RSI might be worth it for an endurance athlete... but spending some time with more extensive methods to help them maintain a 2.1 RSI for more repetitions might be more performance related.
Declan Browne’s research suggests that maximal reactive strength doesn’t appear to have a relationship to relative reactive strength endurance (the relative drop off in reactive strength in response to fatigue) . However, sport is a game of absolutes and a very large positive relationship was observed between maximal reactive strength and absolute reactive strength endurance. Higher maximal RSI, was associated with higher reactive strength throughout a fatigue procotol. So perhaps high levels of reactive strength offer a protective effect against decreases in performance when under fatigue. Logically, this “buffering” effect might be of particular importance in cyclical sports requiring repeated, sub-maximal reactive strength expression of the lower limbs, such as distance running. Some research suggests that training interventions which improve maximal reactive strength may contribute to enhanced endurance performance .
So here are our typical RSI testing trends for our volleyball squad. The athletes are worst at the lowest box “drop” heights (15cm), most of them are best at moderate heights (30cm) and from high box drop heights (45cm) their RSI score lies somewhere in between. The majority of athletes are producing their best RSIs and jumping highest from the 30cm drop height. So what do we do next?
Do we preferentially train in this optimal range? Or do we work on their weak points? Theoretically, I would prefer the athletes to be their best, to be incredible at 15cm – because we think this is closest to the sport. When they “cheat” at 15cm, they tend to hop up off the box - presumably to generate greater eccentric loading on landing, so they can summate more force. But that’s not volleyball. In volleyball we need them to summate force off very little “drop” in order to be better. So do we train in the sweet spot where they produce the greatest RSI but its less sport specific? Or do I make them work the weakness in the more specific range?
Lead Strength & Conditioning Coach at a National Volleyball Program
Let’s start by talking about the “optimal RSI” method and the trend that you are seeing which is some box heights are too low, you find a sweet spot in the middle, and then as you go higher the RSI drops again because maybe the athlete doesn’t have the eccentric strength to tolerate the loading. This is a common trend. Even though your athletes do most of their work (in the sport) from the equivalent of low dropping heights we still wouldn’t expect their RSIs to be higher from those heights. Because what typically happens from those very low dropping heights is two things: you don’t have enough eccentric loading to really stimulate some of those stretch shortening cycle reflexes and potentially you don’t have enough time to appropriately pre-activate the musculature to modulate stiffness most effectively. At the higher heights something else is happening. The strength levels aren’t high enough to tolerate the very high eccentric loadings and potentially there is inhibition to protect the musculature from such high forces. So you find an optimal performance range somewhere in the middle.
There are some studies which suggest that training at the “optimal range” can provide a positive training effect compared with generic dropping heights – its likely a safe and effective way to generically increase maximal RSI. But this doesn’t have to be the only approach. Let’s remember what plyometric training is…. Its “shock” training. It is supposed to overload the athlete to a level that they can’t handle which will drive real changes to the architecture and the neural pathways. So shock plyometrics are not optimal. They shouldn’t be! I think a better formulated research study would be if you took the optimal drop jump height and compared it to 30cm higher than optimal across the board. And I think if you set up studies like that, I don’t think the “optimal” method will always win out. I think the shock method might often produce more acute positive training adaptations. Obviously a problem with the shock method is that it is risky – so when I use the term “optimal” drop height, I mean that more from a safety point of view.
So what do you do next? You’ve identified that most of the sporting actions are at lower amplitude like hopping or falling from a 15-20cm drop height, so that is where you want to enhance performance. It also sounds like that contact time off the ground isn’t the most important thing. Absolute jump height might be more important. So what would I do based on the data you have? I probably wouldn’t spend a huge amount of time in the optimal range… because they are already good at that and importantly it’s not “typical” of the sporting action. I would probably work at the outer limits of the spectrum. I would work in the sport specific range (lower end) because that’s where you think is more relevant to the sport and alongside that I would potentially do small volume exposure to the upper range of the RSI spectrum to ensure we are keeping that “shock” stimulus and continuing to drive some of the architectural adaptation that might be inhibiting performance in some athletes. That is a risky piece of work – the higher the eccentric loading the more risk there is. So I would do it in small doses and only with athletes that have no “red flags”. For high level plyometrics I would consider red flags as recent foot/ankle/calf injury, patella or Achilles tendinopathy, high bodyweight with poor body composition or very low maximal strength levels.
I would do the majority of the work in the sport specific zone but I would try to find a way to overload that zone in some way. What I mean by that is… the work they are going to do in that zone is similar to what they do in their sport training a lot. So what are you going to do differently in their physical preparation training that they are not already doing on the court repeatedly and regularly? For me, that would be something like really exaggerating the jump height outcome. Rather than giving them an RSI score, we maybe set up the Vertec or an overhead target and encourage them focus on an external target for maximal jump height focus. Obviously if we are seeing really long CTs and you are seeing big displacement at the knee and the hip that is completely different to the sport, then you’ll need to dial back the jump height target. If you were seeing a different issue and if you had a contact time problem – for example if the coaches were seeing the athlete hit good jump heights but get there too slowly – then you might need an alternative approach. You could set-up the overhead target sub-maximally but you would record and feedback on contact time and you’re encouraging athletes to hit the same height from shorter and shorter contact times.
Ferris, D.P., Louie, M. and Farley, C.T., 1998. Running in the real world: adjusting leg stiffness for different surfaces. Proceedings of the Royal Society of London B: Biological Sciences, 265(1400), pp.989-994.
Browne, D. and Flanagan, E., Reactive Strength Endurance: Part 2 Is maximal reactive strength associated with reactive strength endurance?
Nagahara, R., Naito, H., Miyashiro, K., Morin, J.B. and Zushi, K., 2014. Traditional and ankle-specific vertical jumps as strength-power indicators for maximal sprint acceleration. J Sports Med Phys Fitness, 54, pp.691-699.
Beattie, K., Carson, B.P., Lyons, M. and Kenny, I.C., 2017. The Relationship Between Maximal Strength and Reactive Strength. International journal of sports physiology and performance, 12(4), pp.548-553.
Healy, R., Kenny, I.C. and Harrison, A.J., 2016. Assessing Reactive Strength Measures in Jumping and Hopping Using the Optojump™ System. Journal of human kinetics, 54(1), pp.23-32.
Beattie, K., Kenny, I.C., Lyons, M. and Carson, B.P., 2014. The effect of strength training on performance in endurance athletes. Sports Medicine, 44(6), pp.845-865.
BIO: EAMONN FLANAGAN
Eamonn Flanagan, PhD., is a strength and conditioning coach and sports scientist based in Dublin, Ireland. He has previously worked in professional rugby with the Scottish Rugby Union, Edinburgh Rugby, and The Irish Rugby Football Union. He is now Lead S&C Consultant with the Irish Institute of Sport and can be found on Twitter @eamonnflanagan