A new tool but same old issues

A new tool to measure training load in soccer training and match play is a paper authored by Rebelo et al (2012) and published in the International Journal of Sports Medicine in April of this year. The aim as set out by the authors was to “assess the relationship between a new method of training load and established heart rate (HR) methods of training load”.  The new method being a Visual Analogical Scale (VAS) and the established methods being Banisters TRIMP and Edwards TRIMP.  Now if you have read the training load section of this website you probably know the word “established” is going to be a gripe for me with regards to these methods. But anyway what does this study show?

Well the authors correlated two versions of the new VAS method to Banister and Edwards TRIMP from soccer training sessions and matches. You’ll have to have a look at the scales in the paper to fully appreciate them (unfortunately I can’t post a picture due to copyright). They also assessed the relationships between VAS and performance measures in a match. The relationships between VAS and the HR methods in training and matches showed significant relationships (r=0.33–0.72; p<0.05). The relationships between VAS, Edwards TRIMP, Banisters TRIMP and the distance based measurements in matches showed significant relationships (r=0.43–0.78; p < 0.05) too. It must be noted that only VAS measurement related to high intensity activities in matches.

So what can we read into the results? Could VAS be a good surrogate measure of training load? Given what I think about the so-called established methods I think the answer to that question is pretty clear. But you may ask what is wrong with session RPE given that it also shows relationships with Banister and Edwards TRIMP (Impellizerri et al, 2004). If you read the introduction of this paper it is riddled with contradictions.

The authors use the rationale of the HR–VO₂ relationship as a reason for using HR based measures. As I’ve highlighted elsewhere “intensity” is not the same as “load”. Therefore this rationale needs looking at with a bit more scrutiny. However they go on to criticise HR as it’s not “accurate in the differentiation of high intensity exercise above max VO₂….or modes such as resistance training… etc”. If HR is not accurate then why are HR based methods being used to justify the use of VAS. Although this could be true, in sprint activities research has shown the contribution of the aerobic system to supra-maximal exercise (Gaitanos et al, 1993). It is my belief that the elevated HR and VO₂ consumption between these bouts add to the training load after the event. Furthermore as training load calculations consider time, this represents 2% of the time in games (Stolen et al, 2005), possibly even less in training. So any underestimation would be minimal. Secondly, I agree with the latter half of the statement but resistance training activity wasn’t part of the exercise the players did in this study. Unfortunately we are some time away from combining load measurements from the two types of training and may never do.

What did they have to say about RPE? They acknowledge the relationships between RPE, blood lactate and HR. However they also considered that RPE only explains 50% of variability in heart rate (Impellizerri et al, 2004). The justification for the assessment of VAS comes from the reported relationship of VAS itself with RPE (r=0.99). So they are basically the same as a measure of intensity and given they are both scored on scales of 0-10 what could be different? Also if you are creating a new subjective tool to replace an old subjective tool why not compare them both? Surely any differences would highlight potential issues and show the superiority of one over another.

In summary, in my view this paper has failed to consider the validity of the HR methods of training that the VAS has been compared to. Although I don’t consider session RPE a valid measurement, my question would be what does this method give you that RPE doesn’t? The arguments made by the authors are that subjective methods take into consideration psychological factors whereas HR based methods don’t. This may well be true, but as long as certain HR based methods (iTRIMP) show stronger dose-response relationships than subjective methods and relate to changes in outcome (Akubat et al, 2012; Manzi et al, 2009; Manzi et al, 2012) this is the way I’d want to monitor the training process.

What are your thoughts? Agree or disagree I look forward to your comments.