|
Optimistic
·
Mar 9, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
Up to 58 replies now which is great...just a bump to see if we can get a few more responses, and tomorrow I'll work up what's come in and post it up. Thanks again to those that wrote in.
|
|
Doug Hemken
·
Mar 9, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
I looked at this question a few years ago with some data collected by Hans Florine - his data was not a random sample, but collected mostly at trade shows. Your data will not be a random sample of the climbing community, either. I found some evidence of a quadratic relationship between bmi and climbing ability. And, to my dismay, some evidence of a gender gap.
|
|
Josh Kornish
·
Mar 9, 2014
·
Whitefish, MT
· Joined Sep 2009
· Points: 800
You aren't going to be able to conclude any correlation between the two. It's much more between your ears than BMI.
|
|
Doug Hemken
·
Mar 10, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Edit of the above: mis-remembered that. Quadratic relation with height, but linear relation with bmi and hand strength. Hand strength is related to gender. R2 was around 0.34, N=101.
|
|
Optimistic
·
Mar 10, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
Doug Hemken wrote:Edit of the above: mis-remembered that. Quadratic relation with height, but linear relation with bmi and hand strength. Hand strength is related to gender. R2 was around 0.34, N=101. I think I'm going to have a tough time developing a training regime to optimize my height.
|
|
Doug Hemken
·
Mar 10, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Too bad we can't control everything related to ability, eh? If I recall correctly (obviously I often don't, the last time I looked at Hans' data was 6 years ago!) I was pursuing the observation that many great climbers are short. Hans' main interest was in hand strength, and he spent a great deal of time getting people to squeeze a dynamometer. The question of "hand strength versus technique [meaning balance, quickness, and precision?]" was a debate that raged in the 70s.
|
|
Ryan Watts
·
Mar 10, 2014
·
Bishop, CA
· Joined Apr 2013
· Points: 25
Unfortunately climbing grades aren't really comparable across areas/styles. Take a 5.11 face climb, a 5.11 OW crack, and a 5.11 run out slab. How many self-described "5.11 climbers" are equally solid on all three? Are all 3 likely to be equally correlated with BMI? I would expect a stronger correlation between bmi and ability in say, crimpy slightly overhanging face climbing than something like Offwidth crack climbing. Hope you find a way to make it work though, interesting idea in any case!
|
|
Doug Hemken
·
Mar 10, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Lack of precise measurements does not make it impossible to detect a trend, just harder to find it in a small sample.
|
|
Ryan Watts
·
Mar 10, 2014
·
Bishop, CA
· Joined Apr 2013
· Points: 25
Doug Hemken wrote:Lack of precise measurements does not make it impossible to detect a trend, just harder to find it in a small sample. It's not a matter of precision -- they aren't even measuring the same thing. 5.11 face climbing ability vs. 5.11 OW ability - it's not like 5.12 face = 5.9 OW or something. They're entirely different. It's like comparing "baseball ability". You could assign some arbitrary "skill" value to each player but can you really compare a pitcher to a DH in any meaningful sense?
|
|
Tony B
·
Mar 10, 2014
·
Around Boulder, CO
· Joined Jan 2001
· Points: 24,665
Jon Zucco wrote:You are going to get a lot of different results that may or may not tell you anything of value. I'd wager that onsight ability correlates much more with experience level than bmi. Yeah- so it will show you a poor correlation. I filled it out, but I did so for my pre-injury numbers not today's, as I am currently not climbing and will not be for a few more months (ACL replacement)
|
|
Tony B
·
Mar 10, 2014
·
Around Boulder, CO
· Joined Jan 2001
· Points: 24,665
Ryan Watts wrote:Unfortunately climbing grades aren't really comparable across areas/styles. Take a 5.11 face climb, a 5.11 OW crack, and a 5.11 run out slab. How many self-described "5.11 climbers" are equally solid on all three? Are all 3 likely to be equally correlated with BMI? I would expect a stronger correlation between bmi and ability in say, crimpy slightly overhanging face climbing than something like Offwidth crack climbing. He accounts for some of that when he splits sport and trad. I am not to worried about the anomoly of OW, since <<<< 10% of trad climbs are effected by the crux being OW, and the number of people who pursue them in concentrated quantities sufficient to change the mix significantly are small. As well, overhanging crimpy face climbs generally don't have so much gear and so are generally sport... not trad, so I think you'll see that separated out to a degree as well, such that sport OS will probably correlate more to BMI than will trad. And they you have another effect - the skinny sport types might focus more on hard routes, and might be the younger climbers, giving the appearance that BMI correlates to (but does not necessarily cause) the signal of BMI. In any case, it is a curiosity to which the answer may appeal to the interest of many climbers, yet result only in a collective shrug. That's the case for me, anyway. Best of luck with the data processing. Enjoy, and share when done!
|
|
Optimistic
·
Mar 10, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
Should be able to post something by later this evening...Excel is not my forte, to say the least. My thought was to try a scatter plot of BMI against Onsight grade and see what that looked like, but I'm having trouble specifying the axes correctly. I'll get my wife to give me a hand with it this evening. For the moment, I can tell you that the average respondent onsights trad a hair harder than 5.10b and has a BMI of 23.2. I can also tell you just based on visual inspection of the data that, assuming everyone who wrote in is being honest, some very solidly constructed folks are climbing very hard! If anyone is a) just sitting around and b) is a total Excel and stats whiz (actually just "vaguely competent" would be an improvement over my skills), feel free to get in touch.
|
|
Tony B
·
Mar 10, 2014
·
Around Boulder, CO
· Joined Jan 2001
· Points: 24,665
Optimistic wrote:I'll get my wife to give me a hand with it this evening. I am similarly optimistic, but not always successful. (snark).
|
|
Optimistic
·
Mar 10, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
Tony B wrote: I am similarly optimistic, but not always successful. (snark). I had a feeling there was a comeback in that vein heading my way... :)
|
|
Optimistic
·
Mar 10, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
102 surveys in now, awesome! An MP member (he can out himself if he wants to) with some stats knowledge has offered to help, so we can get more into statistical significance and p-values and stuff (wahoo! Don Whillans would be totally wowed!). I'll send today's results to him in the morning and we'll see what we can figure out.
|
|
Optimistic
·
Mar 13, 2014
·
New Paltz
· Joined Aug 2007
· Points: 450
Hi all: We're working on the data, and currently we don't have enough responses from women to achieve any kind of statistical significance. It would be great if we could get 5 (or more!) more women to respond. Cheers, David
|
|
Doug Hemken
·
Mar 13, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Updated below - this is only part of the data BMI and Climbing Ability An overview of who has responded to the survey, so far (excluding the 4-5 women)
Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- trad | 90 9.886111 1.393283 5 12 sport | 96 10.95573 1.053812 7 13 boulder | 92 3.5 1.593255 1 7 bmi | 99 23.21818 2.609846 17.2 32.7 age | 99 30.52525 8.754408 16 64
Trad and sport are measured on the YDS scale, using just the decimal portion, and with letter qualifiers converted to 1/4 grade intervals. Bouldering is measured on the V scale. In preliminary analyses, age does not have a measurable effect on any of the climbing outcomes. Gender does appear to have a substantial effect, but without a few more women responding including them in this analysis just reduces our already low precision.
|
|
Doug Hemken
·
Mar 13, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Updated below -- this is only part of the data BMI and Trad BMI does not explain a lot of the variation in trad onsight ability, but is explains some (R^2 is about 9%). From these data, we would predict a person of average BMI (around 23) would climb about 5.10. One unit of change in BMI predicts about 0.16 full grade change in trad onsight ability. (Note, BMI has been centered for the following analyses.) Source | SS df MS Number of obs = 90 -------------+------------------------------ F( 1, 88) = 9.33 Model | 16.5675423 1 16.5675423 Prob > F = 0.0030 Residual | 156.202597 88 1.77502951 R-squared = 0.0959 -------------+------------------------------ Adj R-squared = 0.0856 Total | 172.770139 89 1.94123752 Root MSE = 1.3323 ------------------------------------------------------------------------------ tradn | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- bmic | -.1644358 .0538233 -3.06 0.003 -.2713982 -.0574733 _cons | 9.936081 .1413863 70.28 0.000 9.655106 10.21706 ------------------------------------------------------------------------------
|
|
Doug Hemken
·
Mar 13, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Updated below -- this is only part of the data BMI and Sport Results are similar to trad. BMI does a better job of predicting sport ability, with an R^2 of 17%. (So the correlation is stronger.) The effect, however, is about the same. One unit change in BMI predicts about 0.16 full grade change in sport ability, but here the person of average BMI is predicted to be climbing about 5.11. As I recall the numbers (see way above, not too well!) these results for sport and trad are consistent with the data Hans collected among more elite climbers, a few years back. Source | SS df MS Number of obs = 96 -------------+------------------------------ F( 1, 94) = 19.24 Model | 17.9259482 1 17.9259482 Prob > F = 0.0000 Residual | 87.5734008 94 .931631923 R-squared = 0.1699 -------------+------------------------------ Adj R-squared = 0.1611 Total | 105.499349 95 1.11051946 Root MSE = .96521 ------------------------------------------------------------------------------ sportn | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- bmic | -.1650248 .037621 -4.39 0.000 -.2397222 -.0903274 _cons | 10.99234 .0988644 111.19 0.000 10.79605 11.18864 ------------------------------------------------------------------------------
|
|
Doug Hemken
·
Mar 13, 2014
·
Madison, WI
· Joined Oct 2004
· Points: 13,680
Updated below - this is only partial data BMI and Bouldering I won't elaborate on these. Keep in mind the response scale is the V scale, and not the YDS scale in the previous analyses. Source | SS df MS Number of obs = 92 -------------+------------------------------ F( 1, 90) = 12.18 Model | 27.5403822 1 27.5403822 Prob > F = 0.0007 Residual | 203.459618 90 2.26066242 R-squared = 0.1192 -------------+------------------------------ Adj R-squared = 0.1094 Total | 231 91 2.53846154 Root MSE = 1.5035 ------------------------------------------------------------------------------ boulderconv | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- bmic | -.2116049 .060626 -3.49 0.001 -.332049 -.0911609 _cons | 3.528521 .1569687 22.48 0.000 3.216675 3.840366 ------------------------------------------------------------------------------
|