/ Grading in selected route guidebooks.

bpmclimb on 06 Jan 2017
I recently bought and have started to use the new Rockfax South Wales Sport guidebook, and I'm disappointed already to see some gradings which seem way off, both to myself and to many other climbers (judging by the UKC logbook notes and voting). As I've only perused three or four crags in detail, it doesn't seem unlikely that the book contains considerably more existing but erroneous grades.

It occurs to me that Rockfax (like the CC) may have a strict policy with their select guides of not changing any grades from the latest definitive. Is that the case?
In reply to bpmclimb:

> I recently bought and have started to use the new Rockfax South Wales Sport guidebook, and I'm disappointed already to see some gradings which seem way off, both to myself and to many other climbers (judging by the UKC logbook notes and voting). As I've only perused three or four crags in detail, it doesn't seem unlikely that the book contains considerably more existing but erroneous grades.

It would be great to have some examples here. We had a lot of local input into this guide, as you can see by the names on the cover, so I would be surprised if there were problems with grades. A few cases where you disagree would be useful for us to look at.

> It occurs to me that Rockfax (like the CC) may have a strict policy with their select guides of not changing any grades from the latest definitive. Is that the case?

We have absolutely no such policy and never have. We do take other guidebook grades into consideration when writing new books, or editions, but we would never not change a grade just because it was something else in another book. In fact we have done pretty much exactly the opposite for the last 25 years and have put in lots of work updating many crags that have persistently hung onto stern or soft grades. That said, we are extremely sensitive to grade creep hence we do take a lot of convincing before upgrading routes since there is a tendency for this to happen if systems are left unchecked.

As an illustration of our policy on reviewing grades with new books - this is actually the only time we consider changing a grade nowadays . Routes in the UKC Logbook database, that are in Rockfax guides, are now locked at their grades UNTIL we do a new book.

Alan
@ndyM@rsh@ll - on 06 Jan 2017
In reply to Alan James - UKC and UKH:

>Routes in the UKC Logbook database, that are in Rockfax guides, are now locked at their grades UNTIL we do a new book.

That explains lots of the Lakes bouldering grades then :P could you consider relaxing this for where guides are getting pretty old or, due to resources likes LaksBloc, are unlikely ever to be redone?

bpmclimb on 06 Jan 2017
In reply to Alan James - UKC and UKH:

> A few cases where you disagree would be useful for us to look at.

Well there are only a few, but as I say, so far I've only looked closely at a small portion of the book. Check out You've Had Your Chips at Tirpentwys: almost certainly three grades easier than the given grade, and with nearly 200 logged ascents there's plenty of opinion/voting - it's a popular route which has been gagging for a regrade for years. Another missed upgrade, and a pretty obvious sandbag (although admittedly with fewer votes, so a less cast-iron case) is Baldy Walks to Ponty at Trehafod.


Routes in the UKC Logbook database, that are in Rockfax guides, are now locked at their grades UNTIL we do a new book.

That's interesting - I didn't know that.



drysori - on 06 Jan 2017
In reply to Alan James - UKC and UKH:

> A few cases where you disagree would be useful for us to look at.

Salmon Running, Bear Cunning at the Gap is in the online database at 7b. I've not seen in the book to know if it's the same, but this was a 7a which I thought was regarded as soft (I onsighted this shortly after my first 7a redpoint).

The guide looks great, and South Wales is long overdue a decent guidebook. My only issue is that trad climbing in south Wales seems to be so out of fashion these days!
In reply to @ndyM@rsh@ll:

> That explains lots of the Lakes bouldering grades then :P could you consider relaxing this for where guides are getting pretty old or, due to resources likes LaksBloc, are unlikely ever to be redone?

We can do that since we built in a system to break the link on certain crags. If you email me with some crags then I can sort that.

Alan
In reply to bpmclimb:

> Well there are only a few, but as I say, so far I've only looked closely at a small portion of the book. Check out You've Had Your Chips at Tirpentwys: almost certainly three grades easier than the given grade, and with nearly 200 logged ascents there's plenty of opinion/voting - it's a popular route which has been gagging for a regrade for years. Another missed upgrade, and a pretty obvious sandbag (although admittedly with fewer votes, so a less cast-iron case) is Baldy Walks to Ponty at Trehafod.

Ok, well could I ask you not to judge the entire guidebook until you have looked at more than a couple of examples please.

The two routes you mention are not miles out. You've Had Your Chicks (6a) is in at 6b and votes indicate it might be 6a+ although 1/4 of the votes are still for 6b, so not miles out, and certainly not "three grades out" as you say. Baldy Walks To Ponty (6a+) does look like it might need an upgrade in future but it will need more than 10 votes to indicate that.

Please keep in mind that many of these routes are in a book for the first time, and the text was largely put together by the people who put the routes up. It sometimes takes years for grades to settle.

Alan

PS. If you reference any more routes can you type them into the route box above the posting box so that they link directly - thanks.
John Willson - on 06 Jan 2017
In reply to bpmclimb:

The CC has no such policy. We largely applied it in preparing South West Climbs vol1 – FCQ excepted – where for a variety of reasons it was appropriate to do so, but not elsewhere.
In reply to drysori:

> Salmon Running, Bear Cunning at the Gap is in the online database at 7b. I've not seen in the book to know if it's the same, but this was a 7a which I thought was regarded as soft (I onsighted this shortly after my first 7a redpoint).

That is an odd one which does look like it is overgraded at 7b. The rest of The Gap looks okay though. Routes like Controlled Emission (6c) look overgraded but I think there was rockfall there so the votes are off.

> The guide looks great, and South Wales is long overdue a decent guidebook.

Thanks.
wurzelinzummerset on 06 Jan 2017
In reply to Alan James - UKC and UKH:

> The two routes you mention are not miles out. You've Had Your Chicks (6a) is in at 6b and votes indicate it might be 6a+ although 1/4 of the votes are still for 6b, so not miles out...

Actually, that route is obviously graded incorrectly at 6b. I've climbed it and its difficulty doesn't correlate at all with other routes at the crag. It probably is several grades out. With the UKC voting, you can only vote as low as "low" 6a+ for a 6b, so obviously it's not really reasonable to use that line of argument to suggest it's not miles out. Anyway, I'm just trying to be constructive, as I do like the Rockfax guides, and recognise getting the grading spot on for everything is impossible. And as for the cluster of votes for 6b, I think human psychology would account for that.
drysori - on 06 Jan 2017
In reply to Alan James - UKC and UKH:

Yes, it stood out for me because it bumps up my best UK onsight... The rest of the Gap looks ok.

There has been a surprising amount of development in south Wales in recent years, resulting in a lot of minor-but-useful crags being discovered (or literally unearthed). My impression is that while they have attracted some experienced climbers, they have also been frequented by the growing new sport climbing community in the area, aided by the new walls. So as well as being new a lot of the people making comments on grades may be relatively inexperienced, hence it will take a lot of time to settle. It's definitely helped by having a clear and appealing guide so that more people go! As to be honest as long as grades are within a few letters either way it doesn't really matter on a sport route to my mind.
bpmclimb on 06 Jan 2017
In reply to Alan James - UKC and UKH:

> Ok, well could I ask you not to judge the entire guidebook until you have looked at more than a couple of examples please.

Actually, I was careful not to do that, I said that I'd perused only three or four crags in detail. If I pick up on two mistakes in 3 or 4 crags, then it doesn't seem unlikely that there will be more. It is, of course, possible that I've been unlucky enough to pick up on the only such examples in the entire book, in which case I'll be pleasantly surprised. Time will tell.


> The two routes you mention are not miles out. You've Had Your Chicks (6a) is in at 6b and votes indicate it might be 6a+ although 1/4 of the votes are still for 6b, so not miles out, and certainly not "three grades out" as you say.

Sorry, but You've Had Your Chicks is miles out. In fact, to repeat myself, almost certainly 3 grades out. You would immediately know it yourself if you climbed the route. Check the logbook comments: there are 8 suggestions of 5+ and another three for soft 6a (the lowest available grade in the voting is 6a+).
bpmclimb on 06 Jan 2017
In reply to John Willson:

> The CC has no such policy. We largely applied it in preparing South West Climbs vol1 – FCQ excepted – where for a variety of reasons it was appropriate to do so, but not elsewhere.

I always thought it was a general policy - thanks for the clarification. Brian
The Ex-Engineer - on 07 Jan 2017
In reply to bpmclimb: Really great to see that we are finally getting the sort of feedback I was looking for when I posted or the review thread http://www.ukclimbing.com/forums/t.php?t=655573 "How accurate is the grading? Is it soft, hard, consistent, highly variable?"

I'm not in the least surprised that some grades are way out. That is just how things tend to be with newer crags especially when predominantly developed by individuals or small groups.

Again, it doesn't detract from all the hard work that has gone into the new guidebook, but there is no point anyone thinking that the first ever edition of a guidebook will be as accurate as the 3rd or 4th edition as we have with Dorset.
The Ex-Engineer - on 07 Jan 2017
In reply to Alan James - UKC and UKH:

> Routes in the UKC Logbook database, that are in Rockfax guides, are now locked at their grades UNTIL we do a new book.

In this case, with lots of newly described crags, that seem likely to be a poor approach.

Also, I thought one of the main advantages to climbers from potentially moving to 'Apps' was to get away from the all or nothing approach that producing new hardcopy guidebooks requires.

A review in late 2017 with the view to publishing a formal errata (as you did with PGE and possibly other guides) and updating the database and any App would seem to be a much better response.

Nobody is not going to buy the shiny new guidebook just because a few grades out of 1700+ are utterly out, but down playing the issue (which is how your comments come across) or ignoring it for several years doesn't seem sensible.

It is certainly not going to encourage people to move to the new world of the Rockfax App over paper guides if you're going to actively ignore direct feedback about routes for years on end and not take advantage of the ability to occasionally update things on an ongoing basis.

Steve Perry - on 07 Jan 2017
In reply to Alan James - UKC and UKH:

Three Pebble Slab (HVS 5a)

When does TPS get upgraded, the majority of votes say E1, not HVS like the guide ;-)
Offwidth - on 07 Jan 2017
In reply to The Ex-Engineer:
I'd agree with all of that. UKC do themselves no favours with the perceived independance of the routes database and the Rockfax brand reputation if they lock in clearly incorrect grades against the wishes of most climbers who have climbed the route (not the same as those who have logged it). If its clearly a wrong grade please update the database and the Ap. Its not just bad initial grades either: routes can change a lot over a period of years with broken holds and polish.
Post edited at 11:40
AJM - on 07 Jan 2017
In reply to Alan James - UKC and UKH:

> That is an odd one which does look like it is overgraded at 7b

Without question.
Offwidth - on 07 Jan 2017
In reply to Steve Perry:

The route is HVS or E1 as an onsight, depending on the line taken on the finish so voting is always going to be odd on this route. It was strange that Rockfax chose to change the grade back to HVS where the evidence at the time was mainly for the higher grade in the logbooks; especially as the line in Rockfax takes the 'true' bold padding finish rather than the easier HVS variation trending left. 4c padding moves with a ground-fall of over 10m in prospect is E1 if we use the UK grading system properly.

The route gets way too much discussion due to a simple failure to be clear what is what about the line and how UK trad grades work; not helped by some influential old fogies stuck in the sandbag culture of the late 70s.
SDM on 07 Jan 2017
In reply to Alan James - UKC and UKH:

> The two routes you mention are not miles out. You've Had Your Chicks (6a) is in at 6b and votes indicate it might be 6a+ although 1/4 of the votes are still for 6b, so not miles out, and certainly not "three grades out" as you say.

This is the problem with the voting system. You can only vote for one grade either side of the database grade which makes it a lot harder to judge when the grade is wrong unless you have a huge sample size.

Looking at You've had your chicks, there are 6 comments that the route is no harder than a 5 yet those people can't vote any lower than 6a+ so looking at the votes alone, it looks like the grade might not be far off but reading the comments makes it sound like the difference may be much larger.

You then have a load of people who have probably never climbed much harder than a 5 (reference all of the comments about it being people's first '6b') who are never going to vote lower than 6b because it would diminish what feels like a new milestone in their climbing.

You will never get rid of the confirmation bias entirely but I think the results of the grade voting would be far more useful if you introduced an open ended voting system via a dropdown box offering every grade from the very bottom all the way up to 9b+. Sure, you'll get occasional very obvious jokers but it wouldn't be hard for people to discount the opinions of the obvious outliers.

The open-ended system would make it far clearer when the grades are way out (which can be quite common at newish crags/areas or at certain well known holiday destinations) and would be much better for routes that have altered significantly due to broken holds etc that can change a route by a lot more than one plus grade.

While we are on the subject of grade voting, I would also like to see the option to amend your vote in the future. If I do a route that is a new grade for me, I might vote on my estimation of the grade based on my experience up to that point. 12 months down the line, I may have climbed a bunch of other routes of a similar grade and style that may make it very clear that my original vote was wrong but it is set in stone forever, I cannot amend my vote.
bpmclimb on 07 Jan 2017
In reply to The Ex-Engineer:

> I'm not in the least surprised that some grades are way out. That is just how things tend to be with newer crags especially when predominantly developed by individuals or small groups.

> Again, it doesn't detract from all the hard work that has gone into the new guidebook, but there is no point anyone thinking that the first ever edition of a guidebook will be as accurate as the 3rd or 4th edition as we have with Dorset.

> I'm not in the least surprised that some grades are way out. That is just how things tend to be with newer crags especially when predominantly developed by individuals or small groups.

> there is no point anyone thinking that the first ever edition of a guidebook will be as accurate as the 3rd or 4th edition as we have with Dorset.


And of course (to add to years elapsed and number of editions) there's a third significant variable: how thoroughly the research has been done. In this regard there's a wide spectrum among climbing guidebooks: at one end you can picture a dedicated team, each member of which responsible for their own small area, checking all online sources, consulting widely with local activists, and checking by physically reclimbing the majority (if not all) routes; at the other end a much more quickly produced product which is essentially a repackaging of existing information. The former can achieve many years worth of grade reappraisal and consolidation at a stroke; the latter may have very little effect, and simply pass on anomalies to the next guidebook.

To be clear, I'm not stating (yet) exactly where on the above spectrum I think this new Rockfax book lies; I'm suspending judgement until I've read and used it a lot more. I simply expressed disappointment to have already picked up on at least one major grading error, despite having read only a small portion of the book. The route in question is such a clear-cut (and actually pretty well-known) case of overgrading: a popular route at a popular crag. How did they miss it?

While we're at it, how could it (apparently) need pointing out to a RF high-up that the UKC voting system is ineffective at dealing with large grade anomalies? How do RF reliably detect errors in grading? Were, in fact, any grades reassessed and amended for this book at all? I'm starting to wonder. Also, after confidently and categorically telling me that I'm wrong about You've Had Your Chicks (6a), despite clearly having no knowledge of the route, will Alan come back on here and concede his mistake? Questions ..........
The Ivanator - on 07 Jan 2017
In reply to bpmclimb:

In defence of the new Rockfax guide there has been some regrading from the originally proposed grades at Mermaid Cove Rhossili Bay Scarfish (5c) downgraded to 5c and The Naughty Step (6a+) upgraded to 6a+, having done both these the new grades are more representative of what you should expect (i.e. the routes are at least 2 grades apart).
So difficult to judge what this says about overall grade accuracy in the guide as a whole. I would completely agree that You've Had Your Chicks (6a) is 3 grades out, done it twice and standard 5+/5c is what it is.
In reply to The Ex-Engineer:

> In this case, with lots of newly described crags, that seem likely to be a poor approach.

Actually I was hasty when I wrote my reply yesterday.

When dealing with three parallel version of the same information - RF Print book, RF App and UKC Logbook - we had to introduce a system to lock grades to maintain consistency. Initially our plan 2 years ago was to keep the print book as the master copy, hence only change the grades through the system when we upgrade that. In fact the way it has panned out since then means that the master copy is now likely to be the App version.

So when I said we would only change it with a new print book, in actual fact now we will do it with a App upgrade. That isn't to say that we will run an entire upgrade to change a single grade, but it can be a yearly operation if required. It is unlikely to be that for every book though since it is a surprising amount of work to run and update.

In the case of South Wales Sport Climbs there probably will be a need for new developments and grade changes to be done in the next 12 months or so.

It should be remembered that, ideally, a grade should only need to be changed once or, better still, not at all. This is why locking grades became necessary since we had certain routes changing all the time. By restraining the possibility to change grades it avoids too much over compensation swing.

Alan
In reply to SDM:

> This is the problem with the voting system. You can only vote for one grade either side of the database grade which makes it a lot harder to judge when the grade is wrong unless you have a huge sample size.

We are aware of this and plans are afoot to open the system up a bit. It is ridiculously complicated though and should probably be re-written from scratch but that would take even longer.

We are also aware of confirmation bias especially in terms of grade creep which is why we don't change grades on a whim. Before upgrading a route a grade would need a substantial vote in favour of the upgrade.

Alan
In reply to bpmclimb:

> And of course (to add to years elapsed and number of editions) there's a third significant variable: how thoroughly the research has been done. In this regard there's a wide spectrum among climbing guidebooks: at one end you can picture a dedicated team, each member of which responsible for their own small area, checking all online sources, consulting widely with local activists, and checking by physically reclimbing the majority (if not all) routes; at the other end a much more quickly produced product which is essentially a repackaging of existing information. The former can achieve many years worth of grade reappraisal and consolidation at a stroke; the latter may have very little effect, and simply pass on anomalies to the next guidebook.

These are all popular misconceptions that get trotted out whenever we get into guidebook debates. How can something that has never been described in a guidebook before be 'repackaged'? Is a guidebook that covers all the sport routes in an area 'selective' in the old-fashioned sense of the word? Why do 'definitive' guides often have identical descriptions, grades and stars, across multiple editions, for entire crags, if they are being so extensively reassessed each time?

Even if there were any guidebooks that were ever produced in the idyllic way you lay out (something I doubt), I can assure you that they would still have grades you disagreed with in them.

Have you ever done any guidebook work?

> While we're at it, how could it (apparently) need pointing out to a RF high-up that the UKC voting system is ineffective at dealing with large grade anomalies? How do RF reliably detect errors in grading?

The UKC / Rockfax voting system is the most effective and democratic grading system ever built for routes in this country. It allows everyone to have a say and has been a major contributor to helping establish better grades since it was set up 13 years ago. It is used by all Rockfax guidebook writers and most sensible guidebook writers for other books. It isn't perfect, and adding the possibility to vote on a wider range would be an improvement, but it is still the best thing we have got and, as with most things we do, a work in progress.

> Also, after confidently and categorically telling me that I'm wrong about You've Had Your Chicks (6a), despite clearly having no knowledge of the route, will Alan come back on here and concede his mistake? Questions ..........

Well based on the votes shown, the route should be 6a+. The fact you can't vote less than easy 6a+ is a problem, but the actual votes we can look at indicate mid-to-high 6a+. Judging by the comments it does seem that it is easier and I have a reputation as a very stern grader so I am sure if I climbed the route I would be likely to give it a harsh appraisal. My suspicion here is that, this section was written by Gary Gibson, and this is the grade he gave the route when he first put it up (which it might well have been) and what he still gives it on his SportsClimbs.co.uk topo. As editors, Mark and myself do look at grades in detail but with many thousands of routes it isn't a surprise if some slip through the net.

Alan
Samuel P - on 07 Jan 2017
In reply to Alan James - UKC and UKH:
If/when you update the voting system, could we have the voted grade stated next to the username in the public logbooks section? This is surely going to make it more accurate and have people think twice before agreeing with the grade to 'bump up' their stats.

In regards to the flaw of You've Had Your Chicks (6a) could it be possible that lots of people simply don't vote as they can't select 5+....
Post edited at 16:20
phil456 on 07 Jan 2017
In reply to wurzelinzummerset:

> With the UKC voting, you can only vote as low as "low" 6a+ for a 6b, so obviously it's not really reasonable to use that line of argument to suggest it's not miles out.

What are you supposed to vote when you think is a sandbag .?
Only happened to me once and I though it was a programming bug.

guy xavier percival - on 07 Jan 2017
In reply to Alan James - UKC and UKH:



> Have you ever done any guidebook work?

http://www.cordee.co.uk/Fairy-Cave-Quarry-det-15-96-290-9614.html

Looks like he has.
Michael Gordon - on 07 Jan 2017
In reply to Alan James - UKC and UKH:

Some might argue it is quite hard to read any sort of consensus on grades if you don't provide enough voting options! It's not so much of a problem with trad routes as most routes graded E1 are unlikely to fall out of the HVS-E2 range in different climbers' opinions. But with sport grades being narrower I would advocate changing the voting to encompass at least 2 grades either side of the recorded grade, hence for a 6b route have 6a,6a+,6b,6b+,6c. This would give a good range of options while still pinning folk down to voting sensibly. And if the 6b route really is 5+ then a 6a consensus doesn't seem too far out. On the voting it seems less necessary to me to split the 6b or 6a+ grades further (into easy/mid/hard).
The Ex-Engineer - on 07 Jan 2017
In reply to Alan James - UKC and UKH:
Thank you for the more detailed reply.

I'm sure that in another decade time we will all be using guidebook apps, however I am slightly dubious in some regards about how it will all work out.

The expectations of climbers always tend towards being unreasonably high (or that might just be those climbers who have been encouraged to post their opinions online by someone setting up a climbing website!).

Are Apps going to be a massive improvement? or in twenty years are people going to be engaged in online debates/complaints/rants about out of date Apps as they do now about out of print or delayed guidebooks?
In reply to The Ex-Engineer:

> Are Apps going to be a massive improvement? or in twenty years are people going to be engaged in online debates/complaints/rants about out of date Apps as they do now about out of print or delayed guidebooks?

Whenever I read comments like that I think back to 20 years ago, in the 90s. Largely text-based guidebooks with grades decided by small groups of experienced and usually very capable climbers, with no real accountability or idea of the difference between a Diff and a VDiff. Guidebook texts which were often just reprints of the previous edition with new routes stuck in. Guidebooks which paid more attention to the history and first ascents than describing where the crag was and how you should get there. Few illustrations or worthwhile maps. No online forums for discussion. No online databases for public voting.

We have come a long way in the last 20 years, but it is easy to miss it.

Alan
bpmclimb on 07 Jan 2017
In reply to Alan James - UKC and UKH:

> These are all popular misconceptions that get trotted out whenever we get into guidebook debates. How can something that has never been described in a guidebook before be 'repackaged'?

They are not misconceptions, they are facts: it is undeniable that there is considerable variation in how well researched guidebooks are, and I know from experience the difference that makes to the job of subsequent guidebook writers (assuming they care about getting things right). Much of the information needed for S Wales Sport has existed for some time in electronic form that can easily be "repackaged" (or "packaged", if you prefer that word); e.g. the online SWMC pages. Hypothetically, a guidebook producer could use all the grades exactly as they appear there, and save a lot of time and trouble by not questioning any of them.


Is a guidebook that covers all the sport routes in an area 'selective' in the old-fashioned sense of the word? Why do 'definitive' guides often have identical descriptions, grades and stars, across multiple editions, for entire crags, if they are being so extensively reassessed each time?

You're putting a lot of words in my mouth. I didn't say that all guidebooks of any particular type were well researched; I simply said that some guidebooks are better researched than others. I certainly didn't say - or imply - that the better researched ones were the definitives. You came up with that notion - in order to disagree with it, seemingly. Not sure what's going on there.


> Even if there were any guidebooks that were ever produced in the idyllic way you lay out (something I doubt), I can assure you that they would still have grades you disagreed with in them.

Like most sensible people, I would say that perfection is an impossible ideal. However, I have been lucky enough to work with some (largely volunteer) guidebook teams who have been very conscientious about providing accurate information, and willing to put the time in to achieve that. And regarding grades I disagree with: of course I would very likely have a few residual preferences about minor grade differences; that's not what we're talking about here.


> Have you ever done any guidebook work?

Yes. I get the impression that you assumed the answer to be no, although a quick look at my profile would have answered your question. Do you really lock horns with someone over the course of a long thread and not have a quick look at their profile? Apparently you would rather guess, and then get assertive (and actually rather patronising) based on your guesswork. Much like your procedure with the grade anomaly we have been discussing, in fact (see below).


> The UKC / Rockfax voting system is the most effective and democratic grading system ever built for routes in this country. It allows everyone to have a say and has been a major contributor to helping establish better grades since it was set up 13 years ago. It is used by all Rockfax guidebook writers and most sensible guidebook writers for other books. It isn't perfect, and adding the possibility to vote on a wider range would be an improvement, but it is still the best thing we have got and, as with most things we do, a work in progress.

Sounds grand, but this is pure sales talk. The voting system has deficiencies, the most serious of which is that the system breaks down, pretty much completely, in the case of rogue grades - those which are two or more grades out. It's not that it becomes more unreliable; it becomes completely useless. The value of seeing a gentle distribution of a few votes around 6a+/6b when you're not seeing a load of votes at 5+ is exactly zero. The system's value, by the way, is further reduced by the fact that many climbers who use the logbooks (and leave notes) don't bother with the voting. That's why it's crucial for guidebook researchers always to read the logbook notes (I'm still amazed that you needed this to be pointed out). Once the grade has then been flagged as potentially rogue, someone in the guidebook team has to go and climb the route to check it. Perhaps you call that impractical - I call it doing the job properly.


> Well based on the votes shown, the route should be 6a+. The fact you can't vote less than easy 6a+ is a problem, but the actual votes we can look at indicate mid-to-high 6a+.

Meaningless, as discussed above. Logically, you can't take a portion of data, some of which is missing, and use it to indicate anything - that's ridiculous.
Steve Perry - on 07 Jan 2017
In reply to bpmclimb

> The voting system has deficiencies, the most serious of which is that the system breaks down, pretty much completely, in the case of rogue grades - those which are two or more grades out. It's not that it becomes more unreliable; it becomes completely useless. The value of seeing a gentle distribution of a few votes around 6a+/6b when you're not seeing a load of votes at 5+ is exactly zero. The system's value, by the way, is further reduced by the fact that many climbers who use the logbooks (and leave notes) don't bother with the voting. That's why it's crucial for guidebook researchers always to read the logbook notes.

I agree with that, a rare occurrence but does happen, in fact it happened today on a sport route we did in Costa Blanca.

.

Chris Craggs - on 07 Jan 2017
In reply to Steve Perry:

> I agree with that, a rare occurrence but does happen, in fact it happened today on a sport route we did in Costa Blanca.

Thanks for the comment, I'll act on it next time round. I have done the route and don't recall any issues, but it was a quite few years back,

Chris
Steve Perry - on 07 Jan 2017
In reply to Chris Craggs:

> Thanks for the comment, I'll act on it next time round. I have done the route and don't recall any issues, but it was a quite few years back,

> Chris

You must have been going well Chris! We weren't the only ones who struggled on the start of it today. Some folk were going left onto Ruda and then coming back to it. Cheers

bpmclimb on 07 Jan 2017
In reply to Alan James - UKC and UKH:

Anyway, enough of the long posts from me!

On a positive note, a new sport guidebook to the area is very welcome, and I'm sure my copy will get a lot of use. It certainly beats numerous separate SWMC printouts.
Offwidth - on 07 Jan 2017
In reply to Alan James - UKC and UKH:
Thats good news.. Please don't fiddle with the borderline routes, like 3PS, as it's routes that are more than a full grade off that matter. You could deal with it with a combined misgrade (sandbag or a soft touch) button which you would only need on routes with evidence of probems or on rarely climbed or newer routes. If enough people click this just ask a few people you trust to check it.

Two good opposing examples of modern definitives getting things very wrong on famous routes, despite oodles of expertise are Lands End Long Climb at VD (solid HS 4a in my view, albeit climbable at VD avoiding most of the descibed route) and Bowfell Butrress given HS 4b and which is in reality about HVD 4a.
Post edited at 23:25
In reply to bpmclimb:

You are right Brian. My responses were poor. I jumped to conclusions and didn't look at who I was dealing with. The system is flawed and anomalies like this show those up. I do think that overall the UKC logbook system is pretty good though and I do get a bit defensive over it.

My lesson (and partial excuse) is don't rush replies on a Saturday when busy doing other non-work things.

Apologies.

Alan
Michael Gordon - on 08 Jan 2017
In reply to Offwidth:

> You could deal with it with a combined misgrade (sandbag or a soft touch) button which you would only need on routes with evidence of probems or on rarely climbed or newer routes. If enough people click this just ask a few people you trust to check it.
>

Sounds like a rather cumbersome way of doing things. So you've got to wait for folk to make comments that the grade of a random route is miles out, secondly you've got to spot that, then you put in a button allowing some to re-iterate that the grade is miles out but not allow them to suggest what they think it should be (why?). I don't see the purpose of the button at all - all it says is that the grade may be wrong (something you already know from the logbooks) without providing any further information.

Your examples will surely be already known about?
bpmclimb on 08 Jan 2017
In reply to Alan James - UKC and UKH:

Thanks very much for that, that's big of you. For my part, I know I tend to get overly argumentative when I get the bit between my teeth - sorry for that. Anyway, with this thread perhaps some issues have usefully been brought into sharper focus.

Cheers, Brian
Offwidth - on 08 Jan 2017
In reply to Michael Gordon:
Why is just voting the grade is very wrong cumbersome?. It stops this situation that distorts grade votes as you don't produce bad data by your vote. The rest is exactly the same as you can comment if you wish.

Personally, I prefer the Mountain Project system where actual grades are attributed to the voter, so you can follow trusted graders without the noise of the clueless and people voting especially dumbly or wrong on purpose are exposed. They are also retained if the grade changes.

You say these examples are known but LELC was left at the wrong grade again very recently and the BB misgrade is fairly new. I've seen this happen a few times, the former is either lack of care or pig-headedness, the latter probably caring too much (over obsession on risk or over-attention to complaints).

On borderline routes, editors could help themselves by following Alan's advice... don't change classics unless you know something is clearly wrong (there should be no noticeable difference between an easiest E1 and a top end HVS). I'd add, if you think as an editor the change is big, be as conservative in that change as possible then we would maybe have have BB at S 4b which I'd be much more OK with.... same for all those starred VDs at Birchen, Chris Craggs updated from VD to VS that are now settled around HS.

We simply shouldn't have well established starred routes that are 2 (or more) grades out with modern databases providing good input.
Post edited at 11:18
The Ivanator - on 08 Jan 2017
In reply to bpmclimb:

Great to have a heated thread about Rock Climbing issues, there's been too few on the Forums lately. I looked again at the Tirpentwys pages in the new guide and was interested to find that Mental Mantles (5c) got a downgrade from its SWMC wiki 6a, which is probably justifiable, but a far less obvious case for reclassifying than You've Had Your Chicks (6a). Slightly odd that some thought about regrading the crag took place without dealing with the obvious outlier.
There are a number of routes elsewhere in the book that I am familiar with which have been thoughtfully regraded so it seems "You've had..." may be more the exception than the rule. It is a very welcome guide that I know I'll get lots of use from.
Michael Gordon - on 08 Jan 2017
In reply to Offwidth:

> Why is just voting the grade is very wrong cumbersome?. It stops this situation that distorts grade votes as you don't produce bad data by your vote. The rest is exactly the same as you can comment if you wish.


I guess there's nothing wrong with it, it just seemed odd to only have that option on routes where it was already suspected to be a problem (in which case you already know about it). Although presumably there's a reason why only a few options are allowed in the voting (stopping folk voting v-diff for E9s etc), I don't think it would do any harm to give a couple more either side, particularly for sport routes.
Chris Craggs - on 08 Jan 2017
In reply to Michael Gordon:
> I guess there's nothing wrong with it, it just seemed odd to only have that option on routes where it was already suspected to be a problem (in which case you already know about it). Although presumably there's a reason why only a few options are allowed in the voting (stopping folk voting v-diff for E9s etc), I don't think it would do any harm to give a couple more either side, particularly for sport routes.

I'm not sure how easy it would be to implement - but I guess to boxes 'higher' and 'lower' (cue Play Your Cards Right chants) could be added - a few ticks in either one of them would flag up an issue.


Chris
Post edited at 15:41
Duncan Bourne - on 08 Jan 2017
In reply to Offwidth:

I think that in the blurry area between grades (E1-HVS for example) things will always be subjective. TPS being one. I have always felt it to be E1 purely due to the padding run out on the upper slab, a fall from which would see you deck. Others feel that it is steady enough, like Sunset Slab. Time also changes routes. When I first did TPS it was hard to get anything to stick in the hole, which now seems "improved" . Gear changes, footwear changes, routes become polished and worn, climbers get taller and shorter. I would say that people are significantly taller than they were 50 years ago so if the grade consensus is among a bunch of 6' ers rather than a bunch of 5' 6" climbers then that will also make a difference
Offwidth - on 08 Jan 2017
In reply to Duncan Bourne:
I think its E1 for the same reasons as you. Once upgraded it should certainly have stayed at E1. As a maxim, borderline and safe I prefer at the lower grade, borderline and bold I prefer the higher. I despair with those climbers who seem to be grading more for the technicality, as what's the point of the wonderful UK trad system if we don't use it. As well as grading for the easier line, I think there is also a bit of grading for the repeat, or the headpoint and some old climbers thinking back to when comparitively it was HVS (grades have drifted on grit). Setting off up that delicate top for the first time, even in 2017 with fabulous sticky rubber, is very much heart in mouth for someone near thier limit on that grade border.

As for the slot... its was good enough in the early 90's so I've seen no significant change in protectability since I've known it.
Post edited at 18:25
Duncan Bourne - on 08 Jan 2017
In reply to Offwidth:

My own favourite head slapping moment was when I heard someone say that it should be VS because they had top-roped it and didn't find it that hard! Sheeesh!

I may be miss remembering the slot. Either that or I had crap gear when I first led it.
Coel Hellier - on 08 Jan 2017
In reply to Offwidth:

> As for the slot... its was good enough in the early 90's so I've seen no significant change in protectability since I've known it.

Are you convinced that the gear in that slot will hold? It's always seemed so smooth and polished that a cam might just spit straight out. You can fiddle a wire in also, but it's a hard placement to judge and I wouldn't call it bomb-proof either.

Whenever I've led it I've always thought that the uncertainty in the grade of the thing derives from uncertainty in whether the gear is reliable.
Offwidth - on 09 Jan 2017
In reply to Coel Hellier:

I've never described it as bomb proof, I said good enough (with nut and cam). Careful is the definitive adjective but its irrelevant at the E1 bit at the top of the slab anyhow: which "... can easily reduce brainless strongmen to tears and raise muttered calls for top ropes." I agree with the the definitive that its only tough HVS if you sneak off left at the top.
Coel Hellier - on 09 Jan 2017
In reply to Offwidth:

> ... but its irrelevant at the E1 bit at the top of the slab anyhow

But if the gear in the slot can't be trusted (and personally I don't fully trust it) then the crux step rightwards is E1 in itself.
Offwidth - on 09 Jan 2017
In reply to Coel Hellier:

I accept that's a reasonable view... it's just not quite mine. In the end the routes that worry me most are whole grades out, not borderline arguments like the exact protectability of a move on a well travelled classic like 3PS. I have plenty left to discover on the grit venues I've not got to yet, let alone the rest of the UK. At one time (late 1980s definitive grit guides) about a third of the grades below VS were obviously very inconsistent; even with all the efforts made in the modern guides by those who care about lower grades, things are still a tad stiffer in some guidebooks than others (the average difference may be more than that between our views on this particular sequence).
In reply to bpmclimb:
To redirect this thread away from the all-consuming question of Three Pebble Slab,....

I realised over the weekend that we hadn't actually exported the App version of South Wales Sport Climbs hence I have been able to modify the grade of You've Had Your Chicks (6a) in the App document. I gave it 6a for the time being but you can now vote on 5c if you wish.

The South Wales Sport Climbs app version has been delayed due to us waiting for Version 2.0 of the App which should be released in the next week or so. This is a major upgrade which I'll write about elsewhere. However, the reason we had to delay it is to put in steps to take account of the old App guide Gower Sport Climbs. This has had to be withdrawn and we have had to put in some code to enable those who purchased it to have a free upgrade path to the Gower package on the new app version.

Alan
Post edited at 09:48
In reply to drysori:

> Salmon Running, Bear Cunning at the Gap is in the online database at 7b. I've not seen in the book to know if it's the same, but this was a 7a which I thought was regarded as soft (I onsighted this shortly after my first 7a redpoint).

Salmon Running, Bear Cunning - Right-hand Finish (7a) is the old version which finished up the 6c.
Salmon Running, Bear Cunning (7b) includes the tricky direct finish and is the way the route is now described.

Alan
freelunchprovider - on 09 Jan 2017
In reply to Alan James - UKC and UKH:

Beat me to it Alan,I was just about to point out the same. Salmon Cunning as described with the obvious (very tricky)direct finish is 7b any day. Glad to see that you have upped the grade of the micro arete The Yolks on You from my original wikki 4+.
With over 1700 routes there's always going to be some grades that folk quibble about but Mark's done a good job with such a lot of widespread crags.
Just Another Dave - on 09 Jan 2017
In reply to Coel Hellier:

> Are you convinced that the gear in that slot will hold? It's always seemed so smooth and polished that a cam might just spit straight out. You can fiddle a wire in also, but it's a hard placement to judge and I wouldn't call it bomb-proof either.

Just out of interest, I fell off my first attempt in the early/mid 90's: had a sideways nut in the slot (didn't have cams) which I was convinced was bomber. And it was. Didn't feel unreasonable for HVS. Got back on and managed the rockover 2nd go, but needed talking through the top slab, which felt worth every penny of E1.

Sorry Alan, take the thread back again!
bpmclimb on 09 Jan 2017
In reply to all:

There's another rogue grade that comes to mind: The Rotten Corner (HS) at Wyndcliffe Quarry. My opinion and that of various climbers I trust is that the grade is OTT by at least one grade, but probably more; the most one could reasonably give it is (low) HS 4b, and that's being generous. It's really only Severe 4b.

The voting doesn't reflect that, though. Severe isn't available to the voter (which I for one would have gone for), but low HS is - and even that isn't particularly well-represented. The majority seem to have gone for high HS and high 4b. I think there's psychology at work here: pretty much every climber recognises that the route is a soft touch, but there's a reluctance to believe that an "official" grade could be that far out. Put it this way: had the route been given S 4b from the outset I don't believe anyone would be crying sandbag.

Given that there's a line in the 2012 CC Wye Valley Supplement saying that the route is "considered overgraded", would it be acceptable for me (I moderate the crag) to lower the grade, and hence restart the voting? Maybe conservatively (for now), down one notch to HS 4b?




The Ivanator - on 09 Jan 2017
In reply to bpmclimb:

I think that many of the "high in the grade below" votes are a legacy from the old UKC voting system - when it was revamped (in early 2013) I recall all the legacy votes from the simpler old system (grade below, given grade, grade above) were converted into the grade closest to the given grade in the new system. So with The Rotten Corner (HS) (Severe 4a IMO) all the votes for HS (the lowest option available were changed into "high HS" when many were probably voting as low as they could and were of the opinion that the climb was easier than the options available. A few extracts from a thread from around the time of the changes ...probably explains it better than I just have:

Franco Cookson - on 22 Mar 2013
In reply to Alan James - UKC and UKH:
One problem with the new system: It makes the old votes the lower end of what the maximum grade used to be. So an E1, that had loads of voted for 'E2 or harder', now has a load of votes for low in the grade E2, when in actuality all those people thought it E3...

The Ivanator - on 22 Mar 2013
In reply to Franco Cookson: ...and obviously works the same for soft touches. When I voted HS for Wrecker's Slab I was thinking S/HS borderline not HS/VS borderline.
Over time I guess this will iron out and the real sandbags and soft touches will emerge.

Alan James - UKC and UKH - on 22 Mar 2013
In reply to Franco Cookson:
It is still an improvement though. We didn't have data for the grade below beyond it being 'the grade below'. We allocated all these votes to the top of the grade for want of a better place to put them. I can't think of any other way we could have dealt with them.

As said above, this will iron out as more votes are registered.

Alan
Simon Caldwell - on 10 Jan 2017
In reply to The Ivanator:

It would be great if we could (a) see what we'd voted, and (b) be able to change our votes.
Offwidth - on 10 Jan 2017
In reply to Simon Caldwell:
This would be nice from a personal perspective but even nicer for the future of grading. Ivanator raises another example above: I also though Wreckers was certainly nothing like hard enough to be close to VS, taking all the adjectival factors into account. Actual fixed votes (as in Mountain Project: example here for a famous route: https://www.mountainproject.com/scripts/ShowObjectStats.php?id=105836362 ) means data doesn't keep getting lost on grade changes and we can all be accountable and the experienced trusted climbers can be followed by other climbers or guidebook editors. It's a prime factor in why I don't use the logbooks ... I just don't have time to invest my vast experience in a broken system.

I also think if UKC don't do something like this soon, someone else will (either through extending a platform like Mountain Project or by producing a new database system... either way giving something that fixes user names to grades and is clearly not overly influenced by any guidebook organisation). Web changes happen fast in the modern world.
Post edited at 10:43
bpmclimb on 10 Jan 2017
In reply to The Ivanator:

> Over time I guess this will iron out and the real sandbags and soft touches will emerge.


This is a comforting idea, which many climbers would casually subscribe to, but I'm sceptical - in the absence of a guidebook team making executive decisions, the mechanism by which such a process might work isn't clear to me; nor do I think it actually happens reliably. Sometimes grades just sit there, wrong but unchanging.

Take my previous example of The Rotten Corner (HS): first ascent in 1999 (probably given VS 4c, not sure though), and appeared in 2007 LWV at that grade. The only other guidebook entry I referred to earlier, a brief note in the 2012 LWV supplement: "considered overgraded" No new grade specified.

AFAIK there are no plans for a new LWV definitive, and the route won't be in the upcoming LWV Sport. The UKC voting doesn't reflect the actual grade - but even if it did, moderators are supposed to stick to the latest guidebook grade. In the absence of a new definitive, why would the (rogue) grade of Rotten Corner ever be corrected?









TonyB - on 10 Jan 2017
In reply to Alan James - UKC and UKH:

> Routes in the UKC Logbook database, that are in Rockfax guides, are now locked at their grades UNTIL we do a new book.

I'm quite surprised that you don't do this when you do an online update to the guidebooks. For example, the online update that you did to El Chorro in 2010 has not made it across to the logbooks. Will there be a new book coming out at some stage and the changes are waiting for that?

Cheers, Tony

The Ivanator - on 11 Jan 2017
In reply to bpmclimb:

It was some time ago that I made that comment, but think I was referring to the UKC voting indicating whether something was a real sandbag/soft touch. What I was getting at was that with time the (sometimes misleading) placement of the legacy votes from the old system would be outweighed by the weight of votes cast under the new system, taking The Rotten Corner (HS) as an example all the votes for mid or low HS must have been cast since March 2013 and you would expect the voting at that end of the scale to continue rising until it is clear from the voting alone that the climb is wrongly graded.
Of course simply re-grading it would solve all that at a stroke. In my opinion with clear cases like this it would be a responsible bit of crag moderation to revise the grade - you wouldn't want an inexperienced climber to lead The Rotten Corner (HS) then nip over to Wyndcliff and jump on Fibre (VS 4c) thinking they were in for a similar challenge!
Offwidth - on 11 Jan 2017
In reply to The Ivanator:
Except who defines a responsible moderator. I think many UKC votes tend to get close-ish to a grade (unless it's below Severe or its so far off the confirmation bias and lack of proper evidence nobbles things). Moderators are sometimes going to be plain wrong in the same way that people in guidebook teams are sometimes plain wrong; yet a team can consult more widely and fix the mistakes before publication. So, I'd trust the guidebooks on average over moderators anyday.

The best circumstance is fixed UKC voting with identifiable expertise at the grade, which is what I want. Good data linked to many named individuals. Current UKC votes start to fail on more exact grades: as an example on some of the most popular climbs in the UK, the Stanage Popular low end VS classics, all come out as approaching mid grade. In contrast, some of the tough and lonely grit high end VS classics, almost (or sometimes more than) a grade harder, but normally only climbed by the adventurous or experienced, can end up less than half a grade away on the votes. If you looked at the average votes from the exact same people I strongly suspect the grade difference would be near exact.
Post edited at 01:16
In reply to TonyB:

> I'm quite surprised that you don't do this when you do an online update to the guidebooks. For example, the online update that you did to El Chorro in 2010 has not made it across to the logbooks. Will there be a new book coming out at some stage and the changes are waiting for that?

Hi Tony.

If you read further down I actually corrected my own mistake. It had been our original plan to lock the grades to a guidebook master document, but we have now found that the App version has become the master document hence the grades are locked to this, which does enable us to change grades between print editions. As mentioned though, this will not be a frequent occurrence since updates are quite time consuming to release, and people tend not to actually update the data anyway, unless you force them to.

The El Chorro information is too old to have got appified but we are working on a new edition which should be out in 2018. I will make a point of making sure that UKC logbook gets updated as we produce this guidebook/app package.

Alan
TonyB - on 11 Jan 2017
In reply to Alan James - UKC and UKH:

Hi Alan,
Thank you for the explanation. Exciting times in 2018!
CHeers, Tony
bpmclimb on 11 Jan 2017
In reply to The Ivanator:
> with clear cases like this it would be a responsible bit of crag moderation to revise the grade - you wouldn't want an inexperienced climber to lead The Rotten Corner (HS) then nip over to Wyndcliff and jump on Fibre (VS 4c) thinking they were in for a similar challenge!

I'm inclined to agree.

Regraded to HS 4b. Voting now available from low S to high VS (FWIW I plumped for S 4b).
Post edited at 10:39
The Ivanator - on 11 Jan 2017
In reply to Offwidth:
Widespread regrading by moderators with mixed experience and ability is clearly a bad idea. Regrading widely recognised mistakes in guidebooks such as The Rotten Corner (HS) or You've Had Your Chicks (6a) seems to me a different kettle of fish, these two routes are both 2+ grades offbeam with plenty of comment from experienced/respected climbers to justify a change.
There are moderators who could sensibly and very selectively regrade routes (and Brian is certainly one of them), but I appreciate this potentially opens the door to less responsible use of this facility.
Post edited at 10:56
The Ivanator - on 11 Jan 2017
In reply to bpmclimb:
Re. The Rotten Corner (HS)
I'd go for S 4a if the insitu threads are in good nick, the bleached shoelaces that were there last summer might persuade me that HS 4a is a more appropriate grade!
bpmclimb on 11 Jan 2017
In reply to The Ivanator:

Those threads could certainly do with replacing; I'll do that next visit, it's just up the road. Less easily fixed are the placements - they're chunky enough, but they're concretions so could possibly snap if fallen on. I seem to remember someone I know questioning them a while back (might have been Mark, or Rick perhaps).
Offwidth - on 11 Jan 2017
In reply to The Ivanator:
I think its fraught with potential to make things worse. Present the evidence and let UKC decide if to make the change. There won't be many examples like this.

I've been thinking on how to help the current database on lower grade stuff (where most guidebook checkers really don't have the sensitivity and the HD and HVD bands are too narrow to subdivide) ... maybe sub severe just reduce each grade band to 2 voting values (or alternately make HD one value and HVD 1 or 2 values ) then Lands End Long Climb which currently has no HD votes can get the severe votes it deserves, judging from some sensible logbook comments. I still think the top wall is HS ... there is a recentiish break visible which might have removed a good jug and/or some protection. This climb was reviewed for the new guide by several people as good as the best UKC moderator in experience levels and its still clearly a wrong grade.... ditto for the Bowfell Buttress upgrade to HS.

UKC could also centre the votes for obvious anomalies on the best average within the grade given (in which case LELC, which averages HVD could be centred on high VD and the vote for Severe comes in with no structural change and no loss of low HD data.
Post edited at 15:04
The Ivanator - on 11 Jan 2017
In reply to Offwidth:
I think we're singing from the same hymn sheet. Rogue grades are becoming quite a rarity (although rogue interpretations of grades will always be out there ...Zimpara's opinion on Flying Buttress (HVD 4a) for example). So the leave well alone mantra is largely the best approach.
Although with the two climbs discussed at length in this thread that have both been downgraded as a consequence "You've Had your Chicks" 6b > 6a & "The Rotten Corner" VS 4c > HS 4b I am 100% confident the right thing has been done.
Billg - on 11 Jan 2017
In reply to AJM:
Salmon Running.... is 7b if you climb to the top belay. Its soft 7a or 6c+ to the first belay and has had extra bolts added over the last few years
Post edited at 20:08
AJM - on 11 Jan 2017
In reply to Billg:

There's a top belay? It was several years ago now but I'm surprised I didn't try it if it was a 7a with a 7b extension.
freelunchprovider - on 12 Jan 2017
In reply to AJM:

The John West Finish has been incorporated into the route description going directly up and passing the belay of the 6c groove on the L. The original wandered up R to a grubby finish.
bettypastie - on 21 Mar 2017
In reply to Alan James - UKC and UKH:


> Salmon Running, Bear Cunning - Right-hand Finish (7a) is the old version which finished up the 6c.
> Salmon Running, Bear Cunning (7b) includes the tricky direct finish and is the way the route is now described.

Why didn't you chose to describe the well-known 7a version? The new 7b version has been put in the Top 50 list in the guidebook which I also don't really understand considering they are chosen based on 'popularity' and 'UKC logbook votes'...