Forums: Climbing Information: Climbing History & Trivia:
Ratings Progression - Linear or Logarithmic?
RSS FeedRSS Feeds for Climbing History & Trivia

Premier Sponsor:

 


tecais


Jun 18, 2004, 9:31 PM
Post #1 of 15 (3100 views)
Shortcut

Registered: Jul 18, 2002
Posts: 58

Ratings Progression - Linear or Logarithmic?
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

In a different topic petsfed speculated ratings have a logarithmic progression
http://www.rockclimbing.com/...opic=63726&forum=23:

In reply to:
And I've found the inverse to be true. The harder it gets, the more finely we try to slice it. So a two finger "thank god" pocket and a one finger "thank god" pocket could be the difference between 5.14d and 5.15a, if such things could be found on climbs of that difficulty. One would hope for a linear progression, but its more logarithmic.

Performance in many athletic events can be quantified, e.g. time for running, weight for lifting, distance for jumping and throwing etc. In contrast, climbing difficulty ratings are subjective and qualitative, a bit like the scores given for figure skating, gymnastics and similar complex routines, even though a numerical scale is attached. I haven't checked the statistics for various athletic events but suspect historical performance trends against a linear time axis might well be logarithmic on a large enough time scale (years), approaching some asymptotic limit defined by human capability. Occasionally there are freak breakthroughs, like Bob Beamon's 29'+ broad jump, but the average improvement should show a smoothly diminishing trend. Based on this analogy to other sports I would agree with petsfed.


Partner climbinginchico


Jun 21, 2004, 11:44 PM
Post #2 of 15 (3100 views)
Shortcut

Registered: Mar 24, 2004
Posts: 3032

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

sorry to split hairs, but Mike Powell holds the Long Jump world record. Beamon did have the breakthrough. But now that it has been matched, and surpassed, we know that the limits of human achievement are slooowly expanding. Obviously- look at the progression climbing difficulty has made over time.


nagatana


Jun 22, 2004, 12:06 AM
Post #3 of 15 (3100 views)
Shortcut

Registered: Sep 28, 2003
Posts: 425

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

We know there is progress; the nerds are just debating whether it's logarithmic or linear.


sandbag


Jun 22, 2004, 12:12 AM
Post #4 of 15 (3100 views)
Shortcut

Registered: Jan 12, 2003
Posts: 1443

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

You forgot one overlooked and often amazing contribution to the discussion: youre dealing with a living system, i.e. the human. as with any biological organism, they will have random events that you can neither predict nor quantify on a scale other than the history behind it. that being said, there is a limit to human achivement, and the ratings are subjective enough to warrant some latitude because all the humans climbing them are similar but different. will some one eventually put up a 5.16? maybe.


valeberga


Jun 22, 2004, 12:23 AM
Post #5 of 15 (3100 views)
Shortcut

Registered: Feb 2, 2003
Posts: 434

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Of course humans will put up a 5.16. It's our nature (at least in Western Civilization) to make sure that all aspects of our imaginary realities are "increasing." :roll: There will be a 5.16 because we want there to be a 5.16. But will 5.16 actually exist? No, there will only be rock. But does the rock exist... ok that is a different discussion... :x

Any attempt to assign mathematics to a completely subjective aspect of humanity is inherently flawed. We only know that 5.11d is harder than 5.11c, and then only because we say so. It's counting, but it's not mathematical. Now if we were to actually assign ratings on a mathematical basis, now that would be something. But that would require intelligence, forethought, and for someone to give a s#$t.


chuffer


Jun 22, 2004, 12:30 AM
Post #6 of 15 (3100 views)
Shortcut

Registered: Jun 3, 2004
Posts: 135

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

In Star Trek V Captain Kirk, who we all know does not have time to climb regularly, is free soloing El Cap. If a no talent portly hack like Kirk can do that then we can logically assume that ratings will progress in a linear fashion coinciding with the linear progression of human ability...

J


sandbag


Jun 22, 2004, 12:40 AM
Post #7 of 15 (3100 views)
Shortcut

Registered: Jan 12, 2003
Posts: 1443

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

In reply to:
In Star Trek V Captain Kirk, who we all know does not have time to climb regularly, is free soloing El Cap. If a no talent portly hack like Kirk can do that then we can logically assume that ratings will progress in a linear fashion coinciding with the linear progression of human ability...

J

good point, but then again, he was also climbing 700 years from now too. :shock:


chuffer


Jun 22, 2004, 12:56 AM
Post #8 of 15 (3100 views)
Shortcut

Registered: Jun 3, 2004
Posts: 135

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

In reply to:
good point, but then again, he was also climbing 700 years from now too.

I think that was my point about human progression, although I can't be sure. Incidently, at the risk of revealing the extent of my geekitude, I think it's only 300 and some odd years in the future. :oops:

j


whitefingers


Jun 22, 2004, 1:28 AM
Post #9 of 15 (3100 views)
Shortcut

Registered: Feb 12, 2003
Posts: 124

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

When comparing climbs in the same area, the routes at the lower end of the scale can usually be graded by the average weekend hack. i.e. "yeah, that was a 5.8" As the routes get more difficult, the increase in the degree of difficulty becomes less and less tangible. This would be indicative of a logarithmic progression right?


musicheck


Jun 22, 2004, 2:12 AM
Post #10 of 15 (3100 views)
Shortcut

Registered: May 3, 2003
Posts: 34

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

personally, the way i see it is that the difficulty progresses in a linear function, but people progress in a log function. its a scientific fact that beginners advance faster than advanced people cause theyre farther from their potential. If someone who climbs 5.11 tries a 5.10, its as hard for them as someone who climbs 5.12 trying a 5.11, but it takes a shorter time to progress from 5.10 to 5.11 as 5.11 to 5.12. For example, if you look at the 80's grades advanced alot faster than they do now.


jgill


Jun 29, 2004, 2:35 AM
Post #11 of 15 (3100 views)
Shortcut

Registered: May 18, 2002
Posts: 653

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Here's a little divertissement for any of you who are math minded. It shows the absurd lengths mathematicians will go to to model something, and it is related to the question posed.

Lets assume that "Joe", a sports climber, has been climbing a short time and can do 5.10 and some 5.11 moves. He wants to take up bouldering in a well established area where most agree on the assigned V-scale ratings. On his first time out, with no training, he is successful in the V2 range. He then "trains" (sorry, that's classified - he wants to patent the process!) for 16 days (not necessarily successive), immediately thereafter finding he can stick problems in the V3 range.

We'll assume that Joe's bouldering skill (S) in terms of the V-scale is a function of the number of "training" days (t) he's put in (e.g., S(16) = 3) and that his rate of improvement immediately after t training days is inversely proportional to the standard (i.e., skill level) at which he boulders at that time. Thus, as he ascends the V-scale improvement takes more and more effort.

Questions: (1) After 64 training days, at what V-level is Joe climbing (round off)? (2) Approximately how many training days will it take Joe to reach V10 ? (3) How many days until Joe breaks into the stratosphere with V16 ? (4) What's your opinion of this "mathematical model" ?

You can assume all variables are continuous, so that this might be a good problem for first semester calculus students.
:wink:


Partner angry


Jun 29, 2004, 11:07 PM
Post #12 of 15 (3100 views)
Shortcut

Registered: Jul 22, 2003
Posts: 8405

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

The comment about the living organism was correct. The other comment that people are now matching the breakthrough jump is also correct. Now we need to add to this the idea that most humans have the ability to endure beyond what an animal would just on instinct. (yes I know a horse will kill itself with lactic acidosis, but not without prodding).

So we are alive, we can adapt. We are able to overcome pain, and eventually teach ourselves to overcome an enormous amount of pain. Finally, we see what is possible with others' breakthrough performances, now it is possible to the mind and we all succeed. This is best illustrated in Roger Banisters sub-4 minute mile. It had never been done, once he did it, nearly every elite miler in the world followed suit within the year.

Humans will get to be far better climbers, but the populace will not be that much better. Look to a 5k run, the best runners have gone from 16, to 15,...now 12 and change. Yet the average guy still struggles to break 20. I'm pretty sure the average climber will still stuggle on 5.10 and 5.11, even though some sicko is sending 5.19


deafears


Jun 29, 2004, 11:36 PM
Post #13 of 15 (3100 views)
Shortcut

Registered: Dec 10, 2003
Posts: 103

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

In reply to:
This is best illustrated in Roger Banisters sub-4 minute mile. It had never been done, once he did it, nearly every elite miler in the world followed suit within the year.

I'm pretty sure it was just two or three other guys who ducked under 4:00 in the same calendar year as Sir Roger. No more than four, certainly.

Point is, truly elite levels are not just mental barriers. In climbing, I'd plug this at consistantly sending 5.13+. It's a level where sheer force of will is not enough -- you have to have some genetic gifts to climb this hard, and that fact will not change for future generations.

Lastily, just because climbing uses a numeric grading scale does not mean that the difficulty of a route is strictly quantifiable. Grades are basically just a shorthand way of encapsulating a relative difference in difficulty -- for most people, most of the time, a 5.13 is harder than 5.12. But, for certain individuals (the very tall and the very short make the most obvious examples) the grading scale can get way out of whack, at least on certain moves/routes. Grading is more art than science, and mathmatical models are not very effective for describing all the nuances of how grades are developed and applied.


unabonger


Jul 1, 2004, 12:34 PM
Post #14 of 15 (3100 views)
Shortcut

Registered: Aug 8, 2003
Posts: 2689

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Eric Horst discusses the limits of climbing difficulty in his "Training for Climbing".

He points out that Gullich's "Action Directe" at 14d was done over 13 years ago and only now we are barely establishing 15a. This suggests that like in other sports --running--long jumping--etc. we made large jumps in difficulty because of technique, training, and equipment improvements, but that now, the rate of improvement is much slower.

Increases will occur, mostly because of the large numbers of climbers leading to higher number of genetically superior statistical outliers being identified and groomed in the sport.

He also points out that we may be reaching the limits of the YDS as difficulty becomes a matter of even more subjectivity due to anatomical differences, and suggests that we need a more appropriate rating system based on the number of people who are able to do a route, which is the same concept as Mr. Gill's "B" system.

UB


zozo


Jul 1, 2004, 12:37 PM
Post #15 of 15 (3100 views)
Shortcut

Registered: Feb 3, 2004
Posts: 3431

Re: Ratings Progression - Linear or Logarithmic? [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

http://alpha.fdu.edu/psychology/beeker2.jpg


Forums : Climbing Information : Climbing History & Trivia

 


Search for (options)

Log In:

Username:
Password: Remember me:

Go Register
Go Lost Password?



Follow us on Twiter Become a Fan on Facebook