Would love to read.

Yup, agree. These kinds of coaches haven’t invested much effort to think about their practice. And if they aren’t interested in diving deeper, it’s very difficult to support their learning.

Instead, can we support coaches to recognise the value and consequences of ALL tools they have, and with this make better decisions around what tools to utilise against their coaching aim.

Potentially worth a conversation to ensure we are not missing each other’s points.

But I absolutely agree that there is a prevalence of repetition based practice, across all sports.

I’m not sure that the answer is in devaluing some coaching tools (like instruction, feedback and drills) and placing more value on others.

Even if coaches were following an IP view on skill, the overuse of drills is a poor coaching choice. Marteniuk (1976) said once a learner has a basic idea of skill he/she should practice in gamelike conditions.

I have experienced the same as you with grassroots coaches, but I would challenge that it’s because they have a particular view of skill. Often coaches (particularly at that level) don’t put much thought into their tasks (and just copy), and it’s hard to go against the grain of what’s commonly done.

If talking about coaches, what is the practical trade-off between fidelity and representativeness? Genuine q.

I would suggest a sophisticated understanding of skill recognises skill is more than technical competence, but technical competence facilitates skill development (Barrett et al, 2025)

Finally, their practical applications are spot on. Coaches tailor their sessions to what athletes need. Sometimes a more exploratory (NLP) approach is needed, sometimes a more exploitative (LP) approach is needed.

Rather than a dichotomy (LP v NLP), is a continuum (LP <-> NLP) better?

Another perspective is that NLP led to significant improvements over LP, so it is “better”. But I think we both would agree that there is far more nuance than that.

Also, comparing group means is somewhat misleading and probably more insight would be gained from individual responses.

Results interesting. For LP, there is improvement from pre-post, and (very slight improvement) pre-ret. Depending on the paradigm, you could draw different conclusions.

One way to look at it, both interventions led to improvements, so the question becomes when and where would you use LP vs NLP?

Cog science promotes practice fidelity. So creating “artificial environments” is again a poor coaching choice.

Looking at the intervention characteristics, they have artificially exaggerated the differences (IMO), when an authentic coach would likely use a combination of both approaches.

I’m not a fan of how they have represented LP, saying it creates “over-standardization of movement patterns reduces tactical flexibility”

It’s fair to say it creates a standardised movement pattern, but an “over-standardization” which limits tactical flexibility is a poor coaching choice.

Thanks for signposting. I think they lay out their methods well and it’s easy to follow what they did. Hats off to them.

A few things I think are worth considering (and would be interested in your thoughts too):