UKC

Hysteresis on borderline grade changes.

New Topic
This topic has been archived, and won't accept reply postings.
 Offwidth 02 Nov 2009
A bit technical but I think this maybe needs looking at. If as a famous VS becomes HVS not only are all the old votes lost but a 'hysteresis' type of distortion is produced.

Old votes: HVS, hard for VS, VS easy for VS, HS
New votes: E1, hard for HVS, HVS, easy for HVS, VS

Hence with people voting for the category that suits best a noticable grade shift occurs as those voting for HS no longer can and those who would have voted E1, hard for HVS or mid HVS will now do so. The opposite happens on downgrades.

Hence part of the reason soft touches get a higher than expected vote profile (and lower than expected for tough for the grade)
 GrahamD 02 Nov 2009
In reply to Offwidth:

Surely this is only actually a problem if grades changed as a direct result of the 'voting', rather than the voting being only one input to the grade change process ?

You would know better than I but are the people who set the guidebook grades really put out by this phenomena or is it just of interest to us armchair critics ?
Removed User 02 Nov 2009
In reply to Offwidth: Interesting but the fact remains 3ps is E0
 Chris the Tall 02 Nov 2009
In reply to Offwidth:
Not sure about your theory, but I can see where you are coming from

However, been thinking about the number of grading arguments Sunset Slab, TPS and Kayak create and come up with a very simple solution

Declare these three routes to be your benchmark demarcation points for the new guide

In other words for a route to be given HVS, a majority of your grading panel must believe it to be harder than Sunset Slab, but easier than TPS

As to the gardes of the routes themselves, give these routes (and only these routes) split grades - i.e. VS/HVS etc, but state that bold climbers will feel them to be lower grade and nervous climbers the upper grade.
 CurlyStevo 02 Nov 2009
In reply to Offwidth:
Actually I don't like the voting system much myself. I think users should be able to vote on any technical and any adjectival grade for each climb. I actually think the adjectival grades should start at the easyiest move acticipated on a climb graded Easy (probably 2aish) and there should be a grading guidelines page to explain how the technical grades get applied at the lower end. (ie is a standard Easy a 2C and a standard Mod a 3a and a standarad Diff and 3b etc perhaps a very bold mod with unprotected crux would bew Mod 2a or similar).

I'm not sure how important the extra qualifications are on UKC, the naming of them is definately a bit confusing (like hard VS and HVS etc). I guess personally I maybe prefer a second bar indicating from 1-10 how difficult the tech and adj grade are for their given values.

I also think it should be possible to reapply your vote if you change your mind (showing your current vote), afterall experience or grade shifting affects votes. Throwing votes away when the grade changes doesn seem a bit of a waist to me.
In reply to Chris the Tall:

I think the split grade idea is excellent for poorly protected routes, because it then speaks for itself, re. the climber's boldness. It'll also help prevent people getting so hung-up about the adjectival grade.
 Chris the Tall 02 Nov 2009
In reply to CurlyStevo:
It would be really good if the database
a) could allow you to vote for any grade, with hard/standard/low within each grade
b) could remember how you voted (rather than simply that you have voted)
c) could retain these votes even if the grade was changed

However, I believe that this would require extensive changes and a much bigger server !!
 kareylarey 02 Nov 2009
In reply to Removed User: That's so not true!
 Only a hill 02 Nov 2009
In reply to CurlyStevo:
> (In reply to Offwidth)
> Actually I don't like the voting system much myself. I think users should be able to vote on any technical and any adjectival grade for each climb. I actually think the adjectival grades should start at the easyiest move acticipated on a climb graded Easy (probably 2aish) and there should be a grading guidelines page to explain how the technical grades get applied at the lower end. (ie is a standard Easy a 2C and a standard Mod a 3a and a standarad Diff and 3b etc perhaps a very bold mod with unprotected crux would bew Mod 2a or similar).
>
> I'm not sure how important the extra qualifications are on UKC, the naming of them is definately a bit confusing (like hard VS and HVS etc). I guess personally I maybe prefer a second bar indicating from 1-10 how difficult the tech and adj grade are for their given values.
>
> I also think it should be possible to reapply your vote if you change your mind (showing your current vote), afterall experience or grade shifting affects votes. Throwing votes away when the grade changes doesn seem a bit of a waist to me.

Good points, but the question that sprung to my mind was, how many climbers would be able to tell what a 2b move is like--or even anything less than 3c?
 ellis 02 Nov 2009
In reply to Offwidth:

Is this the reason it got upgraded from E3 to E4?
riichar 02 Nov 2009
In reply to Chris the Tall: I dont really see how this would stop the age old debates; the problem is that the style of routes varies so wildly, not that a consensus on what is harder than what of a similar style cannot be reached. In other words, for many people TPS is harder than the sloth, but for equally many the reverse is probably true.

How about an adaptation and expansion of something that got tried in the old yorkshire guide, i.e.

Sunset slab = P (for protection) 3
E (for effort) 0
Tech 5a
TPS = P2+/3, E0, 5a
The sloth = p0, E3, 5a

NB P is a scale 0-3, (0 being perfectly safe falls, 3 being no useful gear and dangerous fall/deckout potential) E is again 0 to 3 (or whatever) depending on how strenuous it is.

Thoughts on a system like this, anyone...?

riichar 02 Nov 2009
In reply to riichar:
Typo: obviously the tech grade for SS is 4b (or thereabouts), not 5a!!
 Chris the Tall 02 Nov 2009
In reply to riichar:
Dear god, save us from yet more grading systems !!!

The point about establishing a benchmark is you have something consistent to compare routes with. Yes comparing the Sloth and TPS is difficult, but so is comparing any two climbs. The best you can do is get a number of opinions and so "OK, considering all the factors, is the Sloth harder or easier for an onsight lead than TPS ?"
riichar 02 Nov 2009
In reply to Chris the Tall: LOL. To be honest I saw my grading system as taking over from all the others, not being in addition to them

One advantage would be that TPS would be undoubtedly E0....
 Si dH 02 Nov 2009
In reply to Chris the Tall:
> (In reply to Offwidth)

> As to the gardes of the routes themselves, give these routes (and only these routes) split grades - i.e. VS/HVS etc, but state that bold climbers will feel them to be lower grade and nervous climbers the upper grade.

No no no. The system already makes adequat account for the boldness of routes. there will always be grade debates between people with different strengths - this is not a problem. Should have have separate grades for every route depending on people's strengths? I mean, I assume we arent going ot grade Chequers Crack HVS/E1 dependingon your crack-climbign ability (or are we?)

Giving a route multiple grade would also devalue the grading system by making it less clear where a route supposedly lies.


PS Im not sure whether your auggestion was actually serious or not. Im also not sure abotu the person asing for more P grades. I dont like either!
 Si dH 02 Nov 2009
In reply to Offwidth:
Oh by the way, I see where your question comes from but I personally dont think its an issue as long as votes are carried over correctly when the grade of a route is updated. Yes the grade of a route affects the options available to vote on and therefore the overall result as you describe, but I dont believe this is a significant problem given the subjectivity of grading opinions anyway.
OP Offwidth 02 Nov 2009
In reply to GrahamD:

It would be worse if it worked in the opposite direction (as an upgrade would then immediatly result in pressure for a downgrade but I suspect on routes with very diverse opinions, the average impression for the grade of a route might change by close to half a grade just from the change of grade category following an upgrade (with the same people voting for the box that 'suits' best).

An example with not exactly a huge spread: 100 votes on a currently categorised tough VS that should be easy HVS. 2 think E1, 7 think tough HVS, 17 mid HVS, 30 easy HVS, 25 tough VS, 13 mid VS, 5 easy VS, 1 HS. Clearly averaging as just HVS.

In the 'best suited' voting categories for VS you get 56 'HVS' votes with a scatter of votes through VS to HS; this 'looks' very close to the border but the editor decides to upgrade. The new HVS votes are now: 2 E1, 7 tough HVS, 17 mid HVS 30 easy HVS and 44 VS; the average now looks firmly in the HVS band, higher than the true average.

Now on top of this effect firstly add the distortion of inexperienced climbers now chasing the 'easy for the grade' tick (who will almost inevitably tend to overgrade on average and who may have avoided the route with its old grade and a tough reputation). Secondly remove some votes of experienced climbers who can't be bothered to vote again. Before you know it a borderline route with a wide voting range becomes mid-grade.
 CurlyStevo 02 Nov 2009
In reply to Offwidth:
I think you also have to remember a lot of people will just vote the current grade because it felt about right afterall a lot of how a route feels is preconcieved based on the grade given.

Then there is those people that are just against upgrades full stop before they happen, once they have happened - when justified, after a period of time those people often accept the new grade.
 Mark Stevenson 02 Nov 2009
In reply to Offwidth: I see where you're coming from, but I struggle to see a situation where you will get people grading the same route easy VS and hard HVS, let alone HS/E1 unless people's ability to grade is far worst than I thought.

If there were only 3 categories this would certainly be an issue but with 5 it probably isn't something to loose sleep over.

Also, we do have to remember that if there is a contentious route the guidebook editor has probably climbed it and will use his own judgement to cast a deciding vote.
 GrahamD 02 Nov 2009
In reply to Offwidth:

I suspect that grit is 'funny' for votes for two other reasons:

The UK grading system when applied to grit is grading routes at one end of the continiuum. Since the routes are short the differnce between two grade steps could easily be the presence or otherwise of a single runner or a mat. One person's bomber cam is another's marginal.

Secondly, I suspect that more people start on grit than any other rock in the UK, so many of the sub extreme voters do not have a breadth of experience of other rock types to make fair comparisons of overall difficulty in relation to other venues. This is one reason for maybe adopting 'definitive' routes away from grit (Central Groove at HS would be a good start - if we did that I doubt that Peak grit would have quite so many VS 4bs....)
 Chris the Tall 02 Nov 2009
In reply to Si dH:
>
> No no no. The system already makes adequat account for the boldness of routes.

I think you are missing my point. Boldness is just one of a number of factors that make up the objectival grade, but for the routes I've suggested your boldness is the main determining factor in whether you feel the route deserves the higher or lower grade.

>
> Giving a route multiple grade would also devalue the grading system by making it less clear where a route supposedly lies.

I think it's ridiculous and misleading to imagine that all routes fit neatly into pigeonholes. By declaring which routes are regarded as the demarcation benchmarks you create a solid reference point.


OP Offwidth 02 Nov 2009
In reply to Si dH:

Votes are not carried over after a grade change (they are lost).

I know that normally it won't be a problem but its precisely on routes with a wide range of votes from subjectivity that do get upgraded or downgraded that it matters most (and no-one seems to have highlighted this effect before).
OP Offwidth 02 Nov 2009
In reply to Mark Stevenson:

Kayak: HVS 5b to E3 5c depending on reach and the exact line; ditto Bengal Buttress VS 4c to E1 5a.
 Chris the Tall 02 Nov 2009
In reply to Offwidth:
Have you really met anyone who genuinely believes
a) That Kayak is HVS 5b for themselves (i.e. not suggesting a hypothethical grade for other people)
b) That Kayak is E3 5c for themselves and has onsighted at least one other route of this grade

You may have point with Bengal Buttress where both crux move and the gear placement to protect it are very reach dependent
 orge 02 Nov 2009
In reply to Offwidth:

I agree that the current system of restricting the voting may create a bias on certain routes and should probably be changed.

In regard to the use of the collected data to determine more accurate consensus grades for routes, randomisation testing might be worth a look. It's a fairly simple technique which allows sample based statistics to be used to test the likelihood of a hypothesis given a certain confidence interval. I haven't thought about how you would actually go about this, but the idea would be to build up a model of opinions/votes for a certain grade and then compare this with the distribution for a particular route. Whether this is worthwhile within the context of grading debates is another matter all together though...

J
 CurlyStevo 02 Nov 2009
In reply to Chris the Tall:
some routes are just graded badly, it's not been unknow for VDiff to go to VS in one guide book writing. Also routes change due to holds breaking off and rock fall etc.

So all in I think it would be usefull to be able to vote any grade you like for a route.
 Si dH 02 Nov 2009
In reply to Chris the Tall:
> (In reply to Si dH)
> [...]
>
> I think you are missing my point. Boldness is just one of a number of factors that make up the objectival grade, but for the routes I've suggested your boldness is the main determining factor in whether you feel the route deserves the higher or lower grade.#

Im not missing your point, I understand it fully. I think youre wrong. Youve just re-made my point, which is taht boldness is one of many factors affectign a route. We shouldnt be giving two grades to bold routes because boldness is the most important factor in climbign them, any more than we should be giving two grades two steep routes where stamina is most important, to cracks where an ability to handjam is most important, or any other route for any other reason.
>
> [...]
>
> I think it's ridiculous and misleading to imagine that all routes fit neatly into pigeonholes. By declaring which routes are regarded as the demarcation benchmarks you create a solid reference point.

On the contrary, the whole point of a grading system is to force routes in to pigeon holes. That's what its all about - making a judgement that enables us to put routes of continuously varying (and subjective) difficulty into digitised categories. If you admit defeat and refuse to pigeon-hole a particular route, then its grade is worthless.
 Si dH 02 Nov 2009
In reply to Si dH: PS apologies for the abysmal grammar and spelling in some parts of that last post!
 GrahamD 03 Nov 2009
In reply to Si dH:

Actually, I thought it was one of your better ones I fully agree with your point about not treating boldness as a special case in the grade.
 Chris the Tall 03 Nov 2009
In reply to GrahamD:
> (In reply to Si dH)
>
> I fully agree with your point about not treating boldness as a special case in the grade.

And another one missing the point - I am not suggesting treating boldness as a special case.

I merely suggested that three routes - Sunset Slab, TPS and Kayak - be designated demarcation points for the grades VS to E2. These three routes are suitable because
a) they are popular
b) they are similar
c) they are fairly close together
d) the generate a lot of debate regarding their grades

Now point d is due to their boldness, which is probably the factor in grades that is most down to personal preference. So you could say it would be better to pick well-protected routes as your benchmarks, but I can't think of any that are either as popular or contentious as these three

Just to repeat, I am not suggesting anymore than one route per grade per guide is given a split grade. OK you could acheive the same by saying that these routes represent the absolute limit of their grades, but I reckon the split grade is better.
OP Offwidth 03 Nov 2009
In reply to Si dH:

The point was clear enough, and correct

In reply to Chris the Tall

Yes to both on the Kayak point, its a doddle to tall climbers who (unlike you) don't mind the next move (HVS 5a at most to the break). Some short climbers find full-on 5c smearing at that height E3 (harder for them than say 4 Pebble slab).
 GrahamD 03 Nov 2009
In reply to Chris the Tall:

OK, if you want to denote 'borderline' routes essentially as benchmarked, then pick something that can be related to on a national level not statistical outliers. (we are trying to fit within a UK wide grading system, not just a grit wide one). In that case, maybe something like Demo route is borderline HS 4b / VS 4b; Red Edge is borderline HVS 5a / E1 5a etc.
OP Offwidth 03 Nov 2009
In reply to GrahamD:

I'd love to see some national lines in the sand.. as it is, all sorts of genuine drifts are starting to come in. Munich on Tryffan, say, is now given HVS where the crux would be standard grit VS (unless the gear placements have exploded).

As for Chris's grit standards problems: I find Kayak the least serious of the three on a beta based lead, despite having the hardest move and in a groundfall position; as an onsight if you reach or you don't makes a huge difference... hence it's a poor grade standard. I think Sunset Slab is comfortable HVS (around the lower quartile dividing mark) and to me 3 Pebble is either clear HVS or clear E1 depending on if you finish left or pad up the right slab.
 Chris the Tall 03 Nov 2009
In reply to GrahamD:
> OK, if you want to denote 'borderline' routes essentially as benchmarked, then pick something that can be related to on a national level not statistical outliers.

You mean there's climbing beyond the peak district ????

Like it or not, the Peak is the most popular climbing area in the UK and based on UKC stats ( a large enough sample to be credible) TPS and Sunset slab get twice as many ascents as Demo Route, and 25 times as many as Red Edge

But that's not the point. Mr Offwidth is the editor of the forthcoming Froggatt guide, so it's reasonable to assume that was in his mind in starting this thread. Comparing a short gritstone HVS with a lakes multi-pitch is never going to be easy, but it's far more important that the grades are consistent within the guide
 Chris the Tall 03 Nov 2009
In reply to Offwidth:
Admittedly my experience on Kayak is limited to three visits over 20 years and rather clouded by the fall I took on the second.

But having soloed TPS and Sunset Slab more times than I care to remember I have great difficulty accepting that they are the same grade. VS and HVS, HVS and E1, don't really care, but not both HVS.
 Al Evans 03 Nov 2009
In reply to Offwidth: You are twisting the definition of Hysterisis, which I wrote a green paper about for the CS way back in 1966.
In hysterisis an object/theory/whatever is stressed to its limit.
In the case of a climbs grade this would be by numerous experiments, e.g people trying and suceeding or failing on it. After the tests it is returned to its original relaxed strength.
Hysterisis occurs if its new strength is greater than its original, e.g if it 'remembers'the stress and compensates for it. This was particularly significant in the technology of strain guages, which would give false readings if previously used, then applied to previously unstrained objects (the basis of my paper).
Thus my argument is that your applying hysterisis to climbing grades can only apply to previously stressed guages, in this case people who have some prior knowledge of the climb rather than first aquaintences with it, My conclusion is that we have to grade climbs for an on sight first ascent.
OP Offwidth 03 Nov 2009
In reply to Chris the Tall:

As regular solos they should feel the equivalent of two 'grades' apart. Onsight, the gear on 3PS is a big comfort and with the lh finish its just over a metre below your feet on the last tricky 4c smear, this to me is tough boldish HVS (The rh finish is a much bolder and longer sequence of 4c padding and would be fine for me at E1). Sunset has a 4b move with a nasty groundfall in prospect: easy unprotected HVS to me. Hence I find no difficulty in accepting that they might well be opposite ends of HVS (a wide old grade band afterall ?
OP Offwidth 03 Nov 2009
In reply to Al Evans:

I thought hyteresis was commonly where the forward and backward graphs of dependant variables enclose a loop indicating the dissipation of energy (magnetic, torsional, dielectric, etc). This can be even in the small scale... you dont have to be 'stressing things to their limit'. In this particular case grading energy is being wasted and the climber does indeed get hardened by the unneccesary path of the grading loop, even if the climber's stress limit hasn't been exceeeded ?
 Si dH 03 Nov 2009
In reply to Offwidth:
You are correct, the reason it is thought of as a stretching things to their limit by some is probably due to the fact that the most common example given is that of elastic followed by inelastic stretching of something under strain. But I do think applying it to climbing grades is stretching the definition somewhat
 Ropeboy 03 Nov 2009
In reply to Offwidth:
>> Some short climbers find full-on 5c smearing at that height E3 (harder for them than say 4 Pebble slab).

lol, as a bit of a shorty I'd rather do Kayak a dozen times over than do 4PS again!

Which reminds me of that reachy E3 at the left hand end of the crag....Parallel Piped, I did it with a tall mate who basically did from the starting footholds whilst I had to smear my way up the chuffing thing!

Don't change grades, they are so morpho dependant it's almost pointless changing them.

J

New Topic
This topic has been archived, and won't accept reply postings.
Loading Notifications...