Crazy idea - TR community research

This was a random idea that popped in my head yesterday and totally half-baked. But wouldn’t it be kind of fun, with all the folks participating here, if we could organize community research on different workout protocols? Lots of caveats, including, but not limited to, finding enough people willing to deviate from the TR plans to do experiments (and not to mention what TR might already be doing as far as analyzing performance outcomes). But with so many people often asking about different protocols, I thought it might be cool if the community could put different ideas to the test in a structured way. Anyhow, that’s my crazy thought of the day!


Not crazy.

You can start by taking part in the FTP Challenge.


excellent idea, especially if the data were in the public domain. Can’t you openly query past ride data on TR right now?

This is an excellent idea, hubcyclist. But it doesn’t have to be experiments: it could be based on observations of our past performance, as Amy_Gomez indicated.

This would be a way of going past the n=1 comments that people post in response to queries. We would get to amalgamate all the n=1s into big n [so to speak]. But what would be the gains? I presume that there would be some fun in trying to structure more general responses to the cries of anguish that appear regularly on this forum. We would get to know things that TR does not have the time / resources to find out [or that TR does not want us to know!]. I would keep us off the streets, wreaking havoc.

This needs several things:
1 a simple [to start] query over which someone is anguishing. An observational example: do someone’s last 2 months personal records predict whether they would be better doing a 20-minute, 8-minute or ramp test for FTP? An experimental example: what is the exact distribution of the ratios of 20-minute, 8-minute and ramp test estimates of FTP?;
2 a person with some knowledge of research design to set up the experiment / observation protocols. There are clearly some people with research experience on this forum, but they do need the time to participate;
3 participants willing to share their data and / or do the required workouts.

I don’t see how this could possibly work. Scientists try to do these tests all the time with control over key testing variables and yet there are still arguments over how the test was conducted and the implications of the results. Here, you would have virtually no control over participant’s adherence, the variables that effect results (equipment, ambient conditions, etc) or even the veracity of the data produced by each participant.

I like the concept, I just don’t see how it could be put in practice to produce anything definitive.

1 Like

I agree with this. All I would anticipate is a false sense of security. Sometimes being wrong is better than being half-right.

Hasn’t this already been accomplished by decades of athletes & coaches all over the world? :man_shrugging:

Sort of? The current state of that big n is rather vague. Some of the best coaches in the world still have different training philosophies. Especially as it relates to time crunched athletes. It seems that there’s probably better consensus on how to train elite athletes. Though even there, I’d argue that different coaches will take different approaches, and sometimes it works better than others.

We know a lot about how a specific type of training will impact the body. It seems that we aren’t as good at how the sum of training, the intensity distribution, the periodization, overall fatigue, and individual response, will affect training. A lot of the best advice I’ve seen still boils down to, “ride lots, go easy sometimes, go hard sometimes, and be mindful of letting too much fatigue accumulate.”

@hubcyclist I am presuming from your comment that you follow the TR program to the “T”?

I have no data, but I suspect that at least 1/2 of TR users deviate from the plan already at some level for a variety of reasons. I have seen here on the Forum and elsewhere that time crunched programs, like TR, allow you to achieve great gains, but not your maximum potential. So I, for example, make several modifications, including adding Tim Cusick’s fatigue resistance training protocol to the TR plan. I know that others have coaches that prescribe additional/substitute workouts.

Regarding implementing your idea in a structured way, I think @apond58 captured it correctly. The FTP Challenge mentioned earlier is on the right track, but to get meaningful results, a more scientific approach would need to be taken.

However, if just doing TR group activities outside of your normal training is what you are looking for without scientifically significant results, the the FTP challenge looks like a good one. Another suggestion is participating in Worldwide Disaster Workout Day. That was really a lot of fun!

1 Like

While what you’re saying is definitely true, many of the sports science studies I have read use relatively small populations, so they try to control as much as they can to get relevant outcomes. With such a large dataset, the distribution of outcomes could tell you something.

I think one of the biggest problems here is going to be measuring improvements from the training protocol being tested. Measuring FTP before and after the training period would be rife with problems. You just need to review the forum regarding FTP measurement uncertainty to determine that results from the trials would be unreliable.

Measuring the effect of an independent variable is trivially easy. Establish a baseline, manipulate a variable, and then re-test. Report differences in groups.

You can use any protocol you want to test. None of them are perfect. They don’t have to be. You can still run statistics on the results, and draw conclusions on the findings.

Most of the peer reviewed research is pretty narrow in scope. They’ll test things like, is 4x4 or 4x8 better. That’s rather limited. Similarly, if we put together a comparison between two training blocks, we would have a rather narrow scope in which we could say that improvement was (or was not) detected. But we should be able to come to a conclusion one way or the other without a lot of difficulty.

Isn’t this the same discussion?

Well, the biggest difference is that TR is doing their own work on the topic, to drive whatever development they have in mind. That thread talked about community side, but we got a glimpse into the TR data and direction.

What seems to be suggested here is a community driven for test design and analysis. It’s likely similar, other than the who and why. Maybe better via this discussion, with the idea of directing some training, testing and such, vs the main level of control in the TR data which is purely in analysis of workouts and potential results via FTP tests and PR charts. The “why” in that is partly linked to the training, but leaves out a TON of possible other influences.

I see this version as being tricky in it’s own way, as mentioned by others above. It’s not impossible, but the desire to have enough control to actually have any confidence in the results is quite a challenge, IMHO.