Are you interested in improving your gaps and sidehops? Performance coach and sports therapist David McFall has recently conducted some research, where he looks at the correlation between jumping with and without the bike. If you plan to spend a lot of time in the gym over the winter, this is a must-read.
Only scratches the surface
As part of a recent study on lower limb asymmetry in trials riders, I recorded a number of jumps both on a bike and in the gym, which has allowed me to look at correlations between the two. These relationships may reveal clues as to whether there’s any physical training we can do which will directly influence how high or far a rider can jump. Strength and power training is very popular amongst elite riders, but the research looking at the topic is pretty thin. This first look at the data only scratches the surface of the possibilities, but there’s certainly some interesting findings.
What’s this thing called correlation?
Let me start by explaining the concept of a correlation coefficient, because it’s going to come up a lot in this article. Basically the correlation is the relationship between two sets of data, ranging from 0 (no link between them at all) to -1 or 1 (there is a perfect negative or positive link between the measurements). So to give examples from my jump tests, if for every centimetre higher you could jump in a countermovement jump you could automatically side hop 1 cm higher this would be a perfect linear relationship (a correlation of 1). If for every 0.1kg of body fat you were carrying you could clear 0.1m less of gap this would be a perfect negative correlation (-1). Now we know that trials riding isn’t purely a physical sport and there’s a huge amount of skill involved, so the correlations aren’t likely to be perfect, but using these correlations can help us to see which of the tests and physical factors examined are likely to be the most important and which don’t really transfer between the gym and the bike.
These relationships may reveal clues as to whether there’s any physical training we can do which will directly influence how high or far a rider can jump.
The bike tests
Six 20″ riders took part in the testing and performed two jumps on a bike – one horizontal and one vertical. The horizontal jump was basically a gap to the rear wheel. The vertical jump mimicked the side hop technique, but instead of jumping onto a box the rider just pulled up the rear wheel as high as possible. In both cases the jumps were videoed and the distance/height jumped calculated using the rear axle as the reference point. The riders didn’t try to clear a defined gap or jump onto pallets because we just wanted to see the maximum displacement. We could have organised it in a contest style with ever increasing targets but that takes a lot more time and repetitions, which could cause fatigue effects.
The gym tests
In the gym, we again performed two jump tests – a horizontal standing long jump and a vertical countermovement jump (CMJ). For the CMJ I was able to borrow a couple of force platforms from Vald Performance in order to dive deeper into the underlying physical characteristics. The good news for those without access to a such expensive equipment though, is that the test with the very best correlation to bike performance was the humble standing long jump. The correlation between this simple test and performance on the bike was excellent – to the side hop 0.83 and to the gap 0.94. Both of these results achieved statistical significance, which means there is only a very small chance that the results were due to chance and not real. In fact the chances that the results were just a coincidence were only 4% and 0.6%.
If you want to monitor performance in the gym over the winter, then the standing long jump is an excellent choice.
The CMJ couldn’t match the same level of correlation to bike performance. The correlation to the side hop was pretty good at 0.73 but to the gap only 0.55. Why is there such a difference between these results? Kotsifaki et. al. have shown that there are different contributions from the ankle, knee and hip when hopping upwards compared to forwards and my speculation is that the techniques and associated physical demands involved in jumping on a bike are best simulated by performing a horizontal jump. Think of how important the hinge position is on a bike, for example look at the movement through positions a to c in the pictures above. Intuitively this movement should have a higher dynamic correspondence to a standing long jump than a CMJ.
Daan Boverhof demonstrating a deep hip hinge on a trials bike. 
Nina Reichenbach hinging in the gym.
Hinging is key
The importance of the hinge isn’t exclusive to trials either, Chris Kilmurray wrote an excellent article talking about the body mechanics of going fast on a downhill bike, where the hip hinge position also features prominently. If this theory is proven to be true, then it would mean that gym exercises using the hinge position should be prioritised when trying to peak for the trials season. So deadlifts, kettlebell swings and maybe the snatch would be better at transferring to the bike than squats or jerks. Of course this theory is a massive leap from the results of this study, but I believe it to be worthy of future investigation. What I can say though, is that if you want to monitor performance in the gym over the winter, then the standing long jump is an excellent choice as it’s easy (and cheap) to perform, and has the best correlation to jumping on a bike.
Riders can’t really go wrong with getting strong and generating very high peak forces in the gym.
The sidehop is a difficult puzzle to crack
The data from the force platforms used in the countermovement jumps was a bit more complicated to interpret. The headline figure of peak take-off force had a good correlation to bike performance at 0.72 to the side hop and 0.70 to the gap. This was also true for the individual phases of the CMJ (concentric 0.73 to the side hop and 0.70 to the gap, eccentric 0.69 to the side hop, 0.71 to the gap). Otherwise though, there were differences between what’s important for a side hop and what’s important for a gap. If you divide the force generated by the rider during the different phases of the CMJ by their bodyweight we find excellent correlations to gap performance (eccentric phase 0.84, concentric 0.88, overall 0.86). What is surprising though is that this doesn’t carry over to side hop performance (0.30, 0.38 and 0.30). Conversely the average force during the CMJ phases correlate better to the side hop than to the gap test (concentric 0.69 vs. 0.43, eccentric 0.74 vs. 0.51). Why is body weight so much more important in a gap than a side hop? Because of the greater displacement of the body? Why is mean force more important in a side hop than a gap where it’s only really the peak that counts? It’s hard to answer these question with the small amount of data available because, as stated during the introduction, these observations were just a bonus from the left-over data from a different study. As is the way with science I’ve ended up with more questions than I started with.
What I have taken away from these results is that my riders can’t really go wrong with getting strong and generating very high peak forces in the gym. That’s a nice thing for a strength coach to hear, but peak force isn’t everything and the side hop is still a difficult puzzle to crack. Exercises using the hip hinge position remain king for cycling performance and I’ll definitely be monitoring my riders’ standing long jump performance this winter in the gym.
Thanks to St Mary’s University, London for help with the initial asymmetry study, hopefully the results will be published soon.
