5 Comments

Hi Griff - I've been away with work for a bit recently so haven't had much time to work on my riding skills but I have been playing around with Chat GPT in my spare time. Have to admit that my earlier views haven't changed much - you need to know what you want from it before it gives you that. So whilst it is useful for code snippets or rewriting a draft it's not so good for finding out information. The suggestion it provided in your post to practice each skill in the circuit for 1.5 mins is an example of that - in most cases that would mean two attempts at most once you have rolled into the jump or skinny.

I read an article from a teacher about how it's making education a lot harder as before there was generally only a couple of common misconceptions about each topic, and these could be easily countered. Now each student has their own set of misconceptions based on how they entered the query and which model they used.

Anyway - I'm back now and have almost recovered from the inevitable post-travel bug so time to get back on the bike. I recently picked up an almost-mint Trek Sawyer rigid singlespeed 29er which is a beautiful bike to ride but is tough on the wrists when you case a jump!

Expand full comment

I'll get back to you, Dogtank. I've been camping the last several days with poor internet connectivity!

Expand full comment

Dogtank, I'm glad to hear that you're playing around with ChatGPT, even tho you're less than impressed!

You wrote:

"The suggestion it provided in your post to practice each skill in the circuit for 1.5 mins is an example of that - in most cases that would mean two attempts at most once you have rolled into the jump or skinny."

In many cases, you're right. The information/coaching providing is not quite right, especially after the first inquiry. So the skill of interacting with a chatbot often involves one or more follow-ups.

For example, I asked ChatGPT:

"You suggested an interleaving routine for a 45-minute practice session focusing on skinnies, small tabletop jumps, and logovers (mountain biking skills).

"You included a suggestion to practice each skill in the circuit for 1.5 minutes. In most cases that would mean two attempts at most once you have rolled into the jump or skinny.

"Only two attempts? That seems extreme to me. I'd prefer a dozen at least! Can you explain in more detail why you think such a minimal # of attempts might be better?"

ChatGPT replied:

============

The idea behind having fewer attempts per skill during a practice session is rooted in the principles of interleaving and varied practice. Here’s a detailed explanation of why this approach might be beneficial:

### Cognitive Load and Focus

1. **Attention and Focus**: By limiting the number of attempts, each run becomes more significant. You are more likely to focus intently on each attempt, leading to higher quality practice.

2. **Reduced Cognitive Fatigue**: High-quality attention can wane over many repeated attempts. Short bursts help maintain mental sharpness and engagement.

### Learning and Memory

3. **Better Retention**: Interleaving different skills helps in better retention and transfer of skills. It forces your brain to repeatedly engage in retrieval practice, which strengthens memory and improves performance under varied conditions.

4. **Contextual Interference**: Practicing different skills in short bursts creates a desirable difficulty. The brain has to work harder to switch contexts, which leads to better long-term learning even if it feels more challenging initially.

### Physical Conditioning

5. **Muscle Fatigue**: Repeated attempts without sufficient rest can lead to muscle fatigue, which might degrade the quality of practice. Short, focused attempts allow muscles to recover slightly before being used again, maintaining the quality of physical execution.

6. **Injury Prevention**: Especially with skills that have a risk factor (like jumps and logovers), fewer attempts reduce the likelihood of overuse injuries or mistakes due to fatigue.

### Skill Mastery

7. **Deliberate Practice**: Fewer attempts can encourage more deliberate practice. Knowing you have limited tries, you might take more time to visualize the attempt, review what went wrong, and plan for improvement before each run.

8. **Precision and Technique**: Focusing on fewer, high-quality attempts can help hone precise movements and technique, as opposed to ingraining potentially sloppy habits through repetition without reflection.

### Adaptability and Versatility

9. **Versatility in Skills**: Interleaving helps in becoming more adaptable. It mimics real-world scenarios where you rarely repeat the exact same movement multiple times in succession. This adaptability is crucial in mountain biking where conditions and trails can vary greatly.

10. **Building a Robust Skill Set**: This approach helps in building a more robust skill set. You become proficient at quickly switching between different skills, which is invaluable in dynamic sports like mountain biking.

### Practical Implementation

To ensure you still get enough repetitions, consider multiple short circuits within your session:

1. **Multiple Rounds**: Instead of one long session of 45 minutes, break it down into 3 circuits of 15 minutes each. In each circuit, you can focus on 1.5 minutes per skill, repeating this across the circuits.

2. **Focused Sessions**: On alternate days, have focused sessions where you do more repetitions of a single skill. This complements your interleaved practice, giving you the best of both worlds.

=========

I also posted that same question to Claude.ai and got a completely different answer.

Claude replied:

============

I appreciate you bringing this up, as it allows me to clarify and refine the practice routine suggestion. You're absolutely right that two attempts per skill would be quite limited, especially for developing complex mountain biking techniques. Let me explain my thinking and offer a more suitable approach:

1. Misinterpretation of time allocation: I may have misjudged the time required for each attempt, leading to an unrealistically low number of repetitions. This was an oversight on my part.

2. Quality vs. quantity: While I was likely aiming for a balance of focused, high-quality attempts with adequate rest, I went too far in reducing the number of repetitions.

3. Skill-specific needs: Different skills may require varying numbers of attempts and recovery times. Skinnies, small tabletop jumps, and logovers each have unique physical and mental demands.

A revised approach:

For a 45-minute session, we could structure it like this:

1. Warm-up: 5 minutes

2. Three 12-minute blocks (one for each skill)

3. Cool-down: 4 minutes

Within each 12-minute block:

- 8-10 minutes of active practice

- 2-4 minutes of rest/reflection/adjustment

This structure allows for:

- Skinnies: 10-15 attempts

- Small tabletop jumps: 8-12 attempts

- Logovers: 12-18 attempts

These numbers are more in line with effective skill development, allowing for sufficient repetition to build muscle memory and confidence.

The exact number of attempts will vary based on factors like:

- Individual skill level

- Complexity of the specific feature

- Physical and mental fatigue

- Time needed for approach and return

This revised approach provides a better balance between repetition and rest, allowing for more meaningful progress in each session.

Thank you for questioning the original suggestion. It's crucial to adapt practice routines to individual needs and the specific demands of each skill. Would you like me to elaborate on any part of this revised approach?

=====

Expand full comment

Dogtank, I can't imagine getting any enjoyment from a rigid singlespeed. But then, you're still young!

Expand full comment

The trails around me are fairly well groomed and there's not a lot of elevation so it's not as bad as it sounds, but there are times it does feel like 'type 2 fun'. According to Strava I'm riding at around the same speed on the rigid singlespeed as I am on the other bikes (hardtails + duallies) - not sure if this is an indication that my bike handling skills aren't too bad or are terrible :)

Expand full comment