CrossFit Discussion Board

CrossFit Discussion Board (http://board.crossfit.com/index.php)
-   Community (http://board.crossfit.com/forumdisplay.php?f=25)
-   -   T-Nation, Shugart and the Truth About CrossFit (http://board.crossfit.com/showthread.php?t=38880)

John Filippini 11-17-2008 03:28 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by Robert Callahan (Post 448733)
The entire purpose of ALL science is to provide logical, reasonable mechanisms to explain why things we observe happen. So if we are all observing something in reality that goes against what the current "science" tells us should be happening, it is the science that is WRONG, not the observations. Just because a study has never been done to verify something does not mean that it is not true.

You're quite right, but incomplete. The part of me that studied psychology also wants to add that the job of science is also to verify the validity of our observations, which can be imperfect and biased. If our studies don't line up with our experiences, the next job is to find out why there is a discrepancy. There may be an error in our study, or there may be a bias in our casual observations. Another study is needed to find out. That's how science continues to grow and change. We only let it rest when the body of evidence, made up both of studies and casual observation, reaches consistency.

Besides, I never even argued that without a supporting study the community was wrong. I don't happen to have that position. I merely get confused when the following discussion occurs and it degrades to CFers get ****ed.

CFer: "Dude, this CF stuff is awesome, you should try it."

Dude: "Why should I believe you? Where's the evidence?"

CFer: "Look at all these people it worked for!"

Dude: "Yeah, but I like what Plan X has been doing for me, and there's no organized science to make me think CF will work better. Without that, I might just be wasting time I could be training with Plan X to roll the dice on an untested program."

CFer: "**** you then, you're an idiot."

Clearly this is an exaggeration, I don't think all CFers are like this, or your average Dude doing Plan X. Chances are the dude will try at least one or two WODs, that's how I got hooked a year and a half ago. But there's nothing wrong with the above argument, or trainers recommending to others not to try CF on those same grounds. It's a situation that should be easily and happily remedied, instead of the final response being the summary of the reaction on these boards.

Daniel Schmieding 11-17-2008 04:16 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Phil (you don't mind if I call you Phil, do you?),

My "deadlift" analogy was the waterbottles being lifted properly. Not some set of barbell-loaded lifts pre-race.

Don't be an ***.

Robert D Taylor Jr 11-17-2008 04:19 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
How do you propose to run a valid study with controls? Anecdotal evidence is not science, but people knew gravity worked well before Newton.

Derek Maffett 11-17-2008 04:33 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by Robert D Taylor Jr (Post 448742)
So even though there are hundreds of people with quantifiable increases instrength, speed, endurance (sometimes all three) because it wasn't done in a "study" it's not valid um...OK

Quote:

Originally Posted by Phillip Garrisonq (Post 448747)
Anecdotal evidence is not science. Thousands of people have claimed to see bigfoot that doesn't make it evidence.

Phillip, that is a straw man argument and you know it.

Shane Skowron 11-17-2008 04:46 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by Robert D Taylor Jr (Post 448791)
How do you propose to run a valid study with controls? Anecdotal evidence is not science, but people knew gravity worked well before Newton.

Right on.

How could you possibly have a control group with a study on elite athletes? Elites, by definition, possess capabilities beyond that of the standard control population.

Tate Rivera 11-17-2008 05:31 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by Phillip Garrisonq (Post 448747)
Anecdotal evidence is not science. Thousands of people have claimed to see bigfoot that doesn't make it evidence.

Really? Bigfoot? I mean c'mon. There is a slight difference between 'claims' of bigfoot and hardcore evidence that many people, through CrossFit, are fitter now than they were before.

Brandon Oto 11-17-2008 05:34 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
I really doubt that Phillip is arguing that CrossFit doesn't make anyone more fit. The issue is how much more fit, in what ways, and particularly whether it's better at those things than other programs -- since these are the questions you want to know when telling someone whether CF is right for them.

Anyone, especially a beginner, will get more fit doing almost anything, from hula hooping to CrossFit. That's not an interesting point to debate.

David Wood 11-17-2008 06:30 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
I wonder if people will ever get tired of this topic (or, at least, this thread)?


The ideal "scientific" experiment would be something like this:

a) take a "large" "population" (more on this in a moment),

b) randomly split them into two groups,

c) subject them to "enough" training (3 months? 6 months? a year? 3 years?) under two different regimens: CrossFit and whatever you want to compare it to (traditional bodybuiding? (whatever that is, everyone seems to have their own definition) . . . or whatever the CSCS-approved regimen is these days (if there is such a thing)?)

d) then, compare the two groups on previously-agreed up "tests".

So, as soon as you contemplate doing this, a whole lot of problems (and decisions) show up.

What population?

Newbies? Untrained? College-age students (probably something like 90% of all studies are done on untrained college-age students, 'cause there are a lot of them, and they're relatively easy for academic researchers to get access to). (And remember, 87.3% of all statistics quoted on the internet are made up on the spot.)
Athletes (already trained)? "Elite" athletes? Middle-aged frumps like myself? Senior citizens?

The results you get for any program may vary quite a bit depending on which population you want to work with.

How large? Well, you'd like to have enough to provide a good sample size and some statistical reliability . . . I'd suggest at least 100 or so in each arm of the study (each training group). You're going to have to account for dropouts, especially if you're planning on a year or more for the training.


Ok, then, you have to agree on what you're going to test on. Frankly, the only test that would make sense to me is to agree on the following: make up a LONG list of various physical tasks . . .
a) run 100 yards
b) run 400 yards
c) run a mile
d) run 10 miles
e) climb a 20-foot rope
f) shovel 1 ton of gravel into a wheelbarrow and move it 75 yards (multiple trips allowed)
g) shoulder a sandbag equal to your bodyweight and carry it 50 yards (then put it down, pick it up, and carry it back)
d) etc . . .

The tests should (collectively) stress every possible metabolic pathway, and every dimension of fitness (strength, speed, accuracy, endurance, etc.). I'd want to make them as low "skill" as possible . . . no olympic lifts here.

Ideally, in my version of this test, the list would have at least 40 items, along with clear rules for how they would be scored.

Then, finally, in my version, on the testing day, only 10 of them would be drawn at random and tested. No one would know until the day of the exam what the test would be (because it wouldn't be determined yet).

(This is the way the CF Games should be run, too, but do you think they ask me? No . . . . . )


If you haven't fallen asleep yet, you can begin to see why this is a fairly difficult and expensive study to run . . . large population needed, long-term training needed, a very richly-supplied testing setup needed (have to be ready to test all 40 items).

However, I submit that this is the test that really reflects what CF is striving toward . . . enhanced work capacity on wide, wide variety of real-world requirements.


The costs of doing such a study are what lead researchers to try and test simpler things, that are easier to measure:

vertical jump
leg extensor strength
VO2 max
treadmill run times

and then use the fact that these tests appear to have some degree of correlation to what they really want to get at to justify using them. (Actually, even the notion that they have a simple idea of "what they really want to get at" is probably mistaken . . . athletic superiority? "health?" longevity? a winning basketball season? . .. . what?)

This testing of simple measurements (which are correlates of something more complex) is what usually "scientists" are talking about, and which (I think) Phillip is suggesting should be done.


I'm not opposed to it . . . I just doubt that it would ever be as satisfactory as the test I describe above.

Tim Donahey 11-17-2008 06:30 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by Tate Rivera (Post 448826)
Really? Bigfoot? I mean c'mon. There is a slight difference between 'claims' of bigfoot and hardcore evidence that many people, through CrossFit, are fitter now than they were before.

No, I think he's claiming that Bigfoot does Crossfit.

WOD
Three rounds, 21-15- and 9 reps, for time of:
Footprints in Alaska, California and Washington

3-2-1-GO!

Mike Prevost 11-17-2008 07:21 PM

Re: T-Nation, Shugart and the Truth About CrossFit
 
Quote:

Originally Posted by David Wood (Post 448874)
I wonder if people will ever get tired of this topic (or, at least, this thread)?


The ideal "scientific" experiment would be something like this:

a) take a "large" "population" (more on this in a moment),

b) randomly split them into two groups,

c) subject them to "enough" training (3 months? 6 months? a year? 3 years?) under two different regimens: CrossFit and whatever you want to compare it to (traditional bodybuiding? (whatever that is, everyone seems to have their own definition) . . . or whatever the CSCS-approved regimen is these days (if there is such a thing)?)

d) then, compare the two groups on previously-agreed up "tests".

So, as soon as you contemplate doing this, a whole lot of problems (and decisions) show up.

What population?

Newbies? Untrained? College-age students (probably something like 90% of all studies are done on untrained college-age students, 'cause there are a lot of them, and they're relatively easy for academic researchers to get access to). (And remember, 87.3% of all statistics quoted on the internet are made up on the spot.)
Athletes (already trained)? "Elite" athletes? Middle-aged frumps like myself? Senior citizens?

The results you get for any program may vary quite a bit depending on which population you want to work with.

How large? Well, you'd like to have enough to provide a good sample size and some statistical reliability . . . I'd suggest at least 100 or so in each arm of the study (each training group). You're going to have to account for dropouts, especially if you're planning on a year or more for the training.


Ok, then, you have to agree on what you're going to test on. Frankly, the only test that would make sense to me is to agree on the following: make up a LONG list of various physical tasks . . .
a) run 100 yards
b) run 400 yards
c) run a mile
d) run 10 miles
e) climb a 20-foot rope
f) shovel 1 ton of gravel into a wheelbarrow and move it 75 yards (multiple trips allowed)
g) shoulder a sandbag equal to your bodyweight and carry it 50 yards (then put it down, pick it up, and carry it back)
d) etc . . .

The tests should (collectively) stress every possible metabolic pathway, and every dimension of fitness (strength, speed, accuracy, endurance, etc.). I'd want to make them as low "skill" as possible . . . no olympic lifts here.

Ideally, in my version of this test, the list would have at least 40 items, along with clear rules for how they would be scored.

Then, finally, in my version, on the testing day, only 10 of them would be drawn at random and tested. No one would know until the day of the exam what the test would be (because it wouldn't be determined yet).

(This is the way the CF Games should be run, too, but do you think they ask me? No . . . . . )


If you haven't fallen asleep yet, you can begin to see why this is a fairly difficult and expensive study to run . . . large population needed, long-term training needed, a very richly-supplied testing setup needed (have to be ready to test all 40 items).

However, I submit that this is the test that really reflects what CF is striving toward . . . enhanced work capacity on wide, wide variety of real-world requirements.


The costs of doing such a study are what lead researchers to try and test simpler things, that are easier to measure:

vertical jump
leg extensor strength
VO2 max
treadmill run times

and then use the fact that these tests appear to have some degree of correlation to what they really want to get at to justify using them. (Actually, even the notion that they have a simple idea of "what they really want to get at" is probably mistaken . . . athletic superiority? "health?" longevity? a winning basketball season? . .. . what?)

This testing of simple measurements (which are correlates of something more complex) is what usually "scientists" are talking about, and which (I think) Phillip is suggesting should be done.


I'm not opposed to it . . . I just doubt that it would ever be as satisfactory as the test I describe above.


Hi David

I learned pretty early on as a grad student in exercise physiology that the science lags behind the coaches. The usual order of things is that coaches figure out what works and then much later (years and sometimes decades) scientists prove that it works. If coaches wait around for science, they will be way behind the cutting edge. Coaches should never ignore the science, but they must forge ahead if they want to gain an advantage. The same is true for athletes. If you wait for the science, you will be behind the power curve. Sometimes you have to experiment or look around at what other athletes are doing successfully. Successful athletes have always been ahead of the scientists from glycogen loading, to periodization to tapering etc...

The other big issue is that there are lots of good studies to be done but nobody to pay for them. Research is not free. This is particularly problematic for strength training research. There is alot more research out there on endurance training due to it's connection to cardiovascular health. There isn't much money out there to fund strength training research. I can't imagine who would pay for a Crossfit study except maybe Coach and I can't see what he would gain by doing so. The study you lay out would be good but expensive....

Then there are all of the issues you mentioned. Having done some human subjects research myself I can tell you that rats make MUCH better research subjects. Humans don't follow the protocol, they fail to show up for testing, they don't follow instructions, they drop out of your study without telling you and figure out many ways to mess up your study. If you want 10 good tests, you need to recruit 20 subjects. It is a tough business. Gives me a headache just thinking about it.

Mike


All times are GMT -7. The time now is 11:24 PM.

Powered by vBulletin® Version 3.6.8
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
CrossFit is a registered trademark of CrossFit Inc.