[AJM]

The increased gap between onsight and RP grades at big grades must be at least partially due to wanting big grades more: I don’t know anyone who has spent 120 days projecting a 7b.

Interesting, my assumption was more that it was based on length of time climbing (i.e. the 7b group would include more newer climbers who haven't yet stopped improving and so felt the need to project, they just think they'll come back next year and bash it out fast)... but actually if they have to have already logged 100 RP and 100 os to get in the data set maybe your interpretation is more likely to be right...

Not 120 days, and perhaps not 7b, but I wouldn't have thought it would be *that* unusual for people to be out of the "bash it out fast" category by the middle or upper 7s. Also, unless you're doing it really fast, do you necessarily have the foresight of your speed of improvement to know you'll be able to stroll it in a year?

Spending half a dozen or 10 sessions on a first 7b+ or 7c or something doesn't feel that crazy, particularly in some situations (a Londoner settling into a winter siege at the Cuttings or Brean perhaps, although I'm certain it's more widely applicable - I'm sure I have friends who've ground their way slowly upwards through Malham medium-duration redpoints in a similar way - perhaps the common feature is an obvious crag choice and a style that leans more towards redpoint, which makes it easy to keep getting back on it).

And then once you're used to spending half a dozen sessions on something, pushing it out to a dozen to get a soft 8a rather than another 7c feels justifiable for the magic grade tick?

[abarro81]

Actually yeah, the more I think about it (and you point it out) the more my original assumption looks like bollocks!

[Stu Littlefair]

Loving the geekery here; I see potential hint of a "9a effect" in the comparison of onsight max to redpoint max, similar to the 8a effect. To be confirmed in 5 years when more people have climbed 9a!

[Stabbsy]

Great post! The thing that jumped out for me was the median grades of F7b/7B - way higher than I'd have guessed. That said, maybe a bit of bias brought in due to the data source when compared to the full population of climbers? If anyone had asked me to guess, I'd had gone with medians of F6b/6C for the UK - nothing to support that other than gut instinct.

I suppose the next question would be does that affect the conclusions drawn? With a full population (i.e., with lower grades represented more fully), I'd have thought the onsight to redpoint gap might come down on average, although maybe little effect at F7b and above grade range. Any thoughts?

[jwi]

Great post! The thing that jumped out for me was the median grades of F7b/7B - way higher than I'd have guessed. That said, maybe a bit of bias brought in due to the data source when compared to the full population of climbers?

To be abundantly clear: 7b represent the highest grade ever achieved in their lifetime by the typical user of the scorecard service.

As this is a convenience sample I didn't bother to calculate p values, but I imaginee it's a pretty representative sample of climbers who wants to improve/chase grades

[jwi]

To finish off and to try to answer some questions about which grades are worth skipping grades for (7a, 7b and 8a apparently) and if the higher grades get progressively easier to break into (unlikely).

Another way to look at the width of the grade scales would be to ask, how many routes do people usually need to do at lower grades before achieving the next grade. Here again, we restrict ourselves to climbers who have logged at least one hundred routes in the data base.

If we for instance look at climbers who have a personal best of 9a, how many routes of the grade 8c+ did they do before their first 9a? Some did no 8c+ before their first 9a, a quarter of the 9a-climbers in the population did only one 8c+ or less, half did two 8c+ or less, and a quarter did three 8c+ routes or more before their first 9a. The first quartile (1 route) and the median (2 routes) is incredibly stable across the grade ranges.

The price for least “inspiring line” goes to all the 8a+ routes out there.... 🤣 (I probably should not mention outliers but one climber in the population had done 32 8a's before doing their first 8a+)





The typical climber in the population did 2 routes half a letter grade lower before they achieved their top grade and 5 routes a letter grade lower.



I am surprised that these numbers are so small, but I think it goes to show that most people do a lot of their climbing indoors.

[Fultonius]

Interesting, but I'm maybe less surprised about the les broad pyramids below the peak. I'm not along amongst my group (predominately Scotland based) who all have done a limited number of routes under their peak.

When I did my hardest it was, admittedly at the end of a period of just focussing on sport climbing and mainly indoors as I was recovering from a big knee op, so couldn't boulder and was doing less trad.

At the time I did it (soft 8a) I'd done 2 x 7c, 1 x 7c+.  I've now filled the pyramid to a whopping 4 x 7c and 2 x 7c+, so marginally less of a knife edge now.

This year I'm gonna pull the finger out and do a fair bit of sport. I hope...

[petejh]

Two routes @ 1 grade below max and four routes @ 2 grades below max is the definition of an ideal pyramid. Any more routes done at 1 or 2 grades below the max would indicate not achieving your full potential! I despair at these 8a+ onsighters who haven't yet climbed 9a

[jwi]

I guess I am surprised because I always had loads of routes at grade-1 as my max went from 5b and upwards. The only grade I ever skipped was 8a+ (but my first two 8bs are both probable 8a+s so in reality I did not skip 8a+).

[MischaHY]

I guess I am surprised because I always had loads of routes at grade-1 as my max went from 5b and upwards. The only grade I ever skipped was 8a+ (but my first two 8bs are both probable 8a+s so in reality I did not skip 8a+).

I think the thing is people don't keep a logbook of indoor routes in the same way as they do outdoor. This is even harder with bouldering because the boulders just have grade ranges and you would have to guess the individual grade. I've been meaning to start doing this with lead at least though because it gives you a real perspective on your long term volume - I know I would have easily 3-4 times as many 8a (and vastly more if we included 7th grade) if I had all the indoor stuff logged, for example - I also currently live in an area with a fairly limited amount of 8a+ that aren't hard horrible sandbagged slabs on slippy lime so this does also limit the volume that gets done outdoors.

[Bradders]

At the time I did it (soft 8a) I'd done 2 x 7c, 1 x 7c+. 

My current routes pyramid is exactly this 

I guess I am surprised because I always had loads of routes at grade-1 as my max went from 5b and upwards. The only grade I ever skipped was 8a+ (but my first two 8bs are both probable 8a+s so in reality I did not skip 8a+).

I think the thing is people don't keep a logbook of indoor routes in the same way as they do outdoor. This is even harder with bouldering because the boulders just have grade ranges and you would have to guess the individual grade. I've been meaning to start doing this with lead at least though because it gives you a real perspective on your long term volume - I know I would have easily 3-4 times as many 8a (and vastly more if we included 7th grade) if I had all the indoor stuff logged, for example - I also currently live in an area with a fairly limited amount of 8a+ that aren't hard horrible sandbagged slabs on slippy lime so this does also limit the volume that gets done outdoors.

Indoor climbing doesn't count.

[spidermonkey09]

Indoor climbing doesn't count.

This, obviously.

[andy moles]

I am surprised that these numbers are so small, but I think it goes to show that most people do a lot of their climbing indoors.

Totally anecdotal, but I'm not sure about that conclusion. When I climbed 8a I'd done one 7c+ and one 7c, but I climb very little indoors. I'd be more tempted to conclude that people are more motivated by doing a new number than by going to very nearly as much effort for a number they've already done, at least in the case of those who don't spend much time projecting.

[Will Hunt]

Also, if you're capable of climbing 8a on the main Leeds Wall then you're probably booking a trip to Flatanger to onsight Silence. I don't think I've ever done anything harder than 7a on there.

[Bradders]

I am surprised that these numbers are so small, but I think it goes to show that most people do a lot of their climbing indoors.

Totally anecdotal, but I'm not sure about that conclusion. When I climbed 8a I'd done one 7c+ and one 7c, but I climb very little indoors. I'd be more tempted to conclude that people are more motivated by doing a new number than by going to very nearly as much effort for a number they've already done, at least in the case of those who don't spend much time projecting.

Yeah definitely. Only indoor climbing I do is training in my shed (99.9% of the time).

[MischaHY]

Indoor climbing doesn't count.

Ah so you don't get fatigued or have any training effect from climbing indoors? Must be nice... 

Indoor mileage obviously very relevant as part of understanding what effort was put into progressing from one point to another. This can then inform future expectations. Part of this information gathering involves assessing intensity of volume by grade. All valuable information IMO but feel free to 'not count' it 

[spidermonkey09]

The point is surely that the data you're gathering is essentially junk for the reasons you've identified. Who knows what a grade a problem inside actually is when its in a range, which themselves bear absolutely no resemblence to outdoor climbing? Grading regimes differ hugely from wall to wall, even more so on lead walls where. Even if you think it does count, I'm sceptical how much use the information actually is. It would be useful if you were using it to solely track indoor performance, but its of limited utility crossing over to outdoor climbing. Just like my ability to tech up 7A grit slabs bears absolutely no relevance to my ability to climb inside.

I don't think I personally can discern anything useful from my climbing inside other than 'I'm going ok on the board' or 'I feel like shit on the board.' This *might* mean I can track my performance based on that feeling alone but board grades are such arrant nonsense that attempting to codify them into a usable performance tracker is pointless for me. I don't see why this would be different for plastic boulders or indoor lead routes.

[remus]

The point is surely that the data you're gathering is essentially junk for the reasons you've identified. Who knows what a grade a problem inside actually is when its in a range, which themselves bear absolutely no resemblence to outdoor climbing? Grading regimes differ hugely from wall to wall, even more so on lead walls where. Even if you think it does count, I'm sceptical how much use the information actually is. It would be useful if you were using it to solely track indoor performance, but its of limited utility crossing over to outdoor climbing. Just like my ability to tech up 7A grit slabs bears absolutely no relevance to my ability to climb inside.

I don't think I personally can discern anything useful from my climbing inside other than 'I'm going ok on the board' or 'I feel like shit on the board.' This *might* mean I can track my performance based on that feeling alone but board grades are such arrant nonsense that attempting to codify them into a usable performance tracker is pointless for me. I don't see why this would be different for plastic boulders or indoor lead routes.

I think the underlying point, that people might have weird-looking outdoor pyramids backed by significant indoor volume, remains though. The fact that cataloging that indoor volume is hard is kinda beside the point (unless your point is that it's hard to record indoor climbs in a meaningful way, which I don't think anyone is arguing about).

[teestub]

I don't think I personally can discern anything useful from my climbing inside other than 'I'm going ok on the board' or 'I feel like shit on the board.' This *might* mean I can track my performance based on that feeling alone but board grades are such arrant nonsense that attempting to codify them into a usable performance tracker is pointless for me. I don't see why this would be different for plastic boulders or indoor lead routes.

This is interesting. For me there’s a direct correlation between my indoor and outdoor climbing performance, and often before trips I’ll prioritise indoor over rock as you can get a better quality session in than you can outside, particularly on grit.

I think what grade is attached to something is really here not there, you know what’s hard for you, and you know what you couldn’t do a few weeks ago, or on the board a few months ago, and you can therefore track your progress. I think it’s also possible to have a decent stab at the actual grades of board things when you’ve climbed enough outdoor stuff at similar angles, should you want to record it more quantitatively.

[MischaHY]

I think the underlying point, that people might have weird-looking outdoor pyramids backed by significant indoor volume, remains though. The fact that cataloging that indoor volume is hard is kinda beside the point (unless your point is that it's hard to record indoor climbs in a meaningful way, which I don't think anyone is arguing about).

This was what I was getting at. Apologies for the snarky message previously.

[spidermonkey09]

I think the underlying point, that people might have weird-looking outdoor pyramids backed by significant indoor volume, remains though. The fact that cataloging that indoor volume is hard is kinda beside the point (unless your point is that it's hard to record indoor climbs in a meaningful way, which I don't think anyone is arguing about).

Yeah, they might, although I don't think its at all uncommon to be a keen outdoor sport climber but not climb at all at indoor lead walls, so I'd be dubious how likely it is. Maybe cause we don't have many really good extensive leads walls in the UK, the trend is toward good bouldering facilities?

I'm more making an open value judgement about whether its worth logging indoor lead routes or loops on a circuit board to be honest!  to me be that would pointless, taxonomy for the sake of it. I'm trying to work out why I think that and I think its because it just feels so pointless and fundamentally ephemeral data when the routes are taken down every month or whatever. Theres nothing to refer back to other than 'how it felt personally' and surely the whole point of a lot of training is that its measurable, hence the effectiveness of fingerboarding and different size edges, numbered campus rungs etc etc. Obviously theres variation in these things too but nothing like the variation in indoor grades.

Just my view obvs, nothing personal (no snark detected Mischa!)

[spidermonkey09]

you know what’s hard for you, and you know what you couldn’t do a few weeks ago, or on the board a few months ago, and you can therefore track your progress. I think it’s also possible to have a decent stab at the actual grades of board things when you’ve climbed enough outdoor stuff at similar angles, should you want to record it more quantitatively.

Fair dos! I honestly have zero idea what grade stuff on a board actually is, perhaps because I don't have your experience of steep bouldering. I can get a sense of when I'm going well but there are too many variables to know whether its because I'm actually getting better or not I find; my performance is much more up and down on a board than on other styles. Other things I find make a big difference to me on the board are when and what I've eaten that day, frequency of fingerboarding that week, and how I've slept. I don't record my sleep and diet though; maybe thats my missing puzzle piece?!

[jwi]

I don't think I personally can discern anything useful from my climbing inside other than 'I'm going ok on the board' or 'I feel like shit on the board.' This *might* mean I can track my performance based on that feeling alone but board grades are such arrant nonsense that attempting to codify them into a usable performance tracker is pointless for me. I don't see why this would be different for plastic boulders or indoor lead routes.

I think it’s also possible to have a decent stab at the actual grades of board things when you’ve climbed enough outdoor stuff at similar angles, should you want to record it more quantitatively.

I always do this. I use grades to decide how difficult training routes or boulders should be (both inside and outside). How else would I be able to know in advance what difficulty I should choose for the individual boulders in a 4x4 for example? Or how difficult a 25 move circuit should be so that it can be completed approximately 5-6 times with 8 min rest between the sets? As the grades indoors often are complete bollocks I just assign my own grades (based on how the same difficulty would feel for me outdoors).

[Fiend]

Warning: Epically long post.

Some empiricial data to ponder

Data scraped in 2017 from a popular provider of online ticklists provides some data on climbers performances. As users get “points” for climbing routes of a certain grade, they are incentivised to log routes or boulders at the highest possible grade they can get away with, regardless of what they think. Also, there is no system of signing off the ascents by belayers so not all ascents logged will have taken place. However, noisy data is better than no data.



As we can see, the best performance ever on fully worked routes has a somewhat bell-shaped distribution with median around 7b and with a premium for climbing a route graded 8a. (Much like there are lots of Marathon runners finishing just under 3 hours, but not many in just over 3 hours.)

The median best performance is surprisingly stable over time and shows signs of only slow increase over fifteen years:

Code: [Select]

year                "2003" "2004" "2005" "2006" "2007" "2008" "2009" "2010" "2011" "2012" "2013" "2014" "2015" "2016" "2017"
Median of top grade "7b"   "7b"   "7b"   "7b"   "7b"   "7b"   "7b"   "7b"   "7b"   "7b"   "7b+"  "7b+"  "7b+"  "7b+"  "7b+"

As some users never climb in the redpoint style the premium for climbing 8a is actually understated in the chart above.

If we look at the distribution of only redpoint ascents:



we see that the mode is 8a, even though the median is 7b+

Climbers who can get up an 8a with enough work are clearly more likely to put in the required effort than those who can get ut 7c+ with enough work.

This effect is slightly less pronounced if we look at ascents at a stable level. Here are the distributions for highest onsight grade and highest grade achieved “Second Go”



There is still clearly a premium for ascending 8a in faster styles as well.

For bouldering



The median of the best performance in bouldering is not improving any quicker than those of sport climbing

Code: [Select]

year                "2003" "2004" "2005" "2006" "2007" "2008" "2009" "2010" "2011" "2012" "2013" "2014" "2015" "2016"
Median of top grade "7B"   "7B"   "7B"   "7A+"  "7A+"  "7B"   "7B"   "7B"   "7B"   "7B"   "7B"   "7B"   "7B+"  "7B+"
                    [,15]
year                "2017"
Median of top grade "7B+"
To compare highest red-point grade with highest onsight grade we only look at users who have logged a reasonable amount of climbs in both styles.

Say at least both 100 red-point and onsight ascents.

Scatterplot


Grouped boxplot:


As we can see, the best redpoint and onsight grade are highly correlated ($R^2=$0.8716624). There seems to be about 2 full letter-grades between the hardest redpoint and the hardest onsight for climbers having climbed 8b or harder, as per JiBe’s old rule-of-thumb. The median best onsight grade for a climber having redpointed 8c is 8a etc. There are two and a half letter grades between the redpoint and the onsight grade for climbers having climbed from 7c to 8a+ at best. Fully half of the climbers having redpointed 8a have onsighted 7b+. For climbers having redpointed no harder than 7b+ there is just two letter grades difference.

Ideally the difference should be the same across the grades. The discrepancy could be explained by significant grade inflation in the higher grades (that is, the difference between becomes smaller the harder the routes) or that elite climbers are not pursuing onsight ascents outdoors for whatever reason.

In case you are wondering, there is no big difference is we look at the highest grade achieved redpoint vs onsight in a single year.



Relation between performances in bouldering and sport-climbing


There is a fairly strong relationship between best bouldering grade and best redpoint grade as well. This is of course unsurprising since harder routes generally have harder cruxes. For climbers who have logged at least N=100 boulders and the same amount of redpoint ascents, it seems like the typical bouldering strength required to climb an 8a route is about 7B+ whereas 25% of those who can do 7B have also done 8a. The lowest bouldering strength required for someone doing a fair amount of bouldering to do 8a is 7A.

If any thread needed Johnny Brown's involvement, this one is it 

[Johnny Brown]

Given the subjective nature of the grading scale I need to see a Bayesian analysis before commenting.