UKC

Before I start this long post: this is NOT intended as Lattice-bashing thread, but rather an enquiry as to what limitations there may be associated with how their data-gathering and training programmes operate (limitations, I should think, which will be of inherent interest to Lattice themselves, and which I'd welcome hearing from any of those guys directly about). I'm asking purely as a matter of curiosity.

--

First up, let me say that I think Lattice are really great. The sports science that they are employing, their drive to help everyone from punters like me to top pros improve, and their obvious pure and simple love for climbing, are all awesome.

I recently undertook a Lattice assessment, in conjunction with hiring a coach for a couple of sessions for the first time, and the process has been a revelation. I had no idea how weak my fingers were relative to where I need them to be to climb harder; my shoulder stabilisation is non-existent and that is generating all sorts of problems; my core is so much weaker than I anticipated, etc etc. These are all things that I wouldn't have found out otherwise, and even better I now know exactly what I need to work on to improve. I'm psyched.

However, something I've been thinking about is that when my numbers got crunched, what they basically spat out was the conclusion that my climbing grade is outstripping what would be expected given my actual physical capacities.

For information: in the past year I've onsighted a 7a, redpointed a bunch more (about 10, I think) and redpointed third go a single 7a+ (soft at the grade, but whatever: Poison Ivy (7a+)). My scores on finger strength especially, but also functionality and first-burn aerobic capacity are well below what Lattice expect for the grades I've managed. The report puts this down to generic explanations to do with terrain, technique, and tactics compensating for my poor physical capacities.

Another factor however is that I have massive aerobic base fitness and capacity for recovery without getting fatigued. My coach hypothesised that this is why I can get up routes I "shouldn't" be able to.

But this has gotten me thinking about the data Lattice have, and whether it is being inherently limited by e.g. the kinds of people who do Lattice assessments, giving a somewhat incomplete snapshot of what it 'takes' to climb certain grades. The following seem pertinent issues:

- My recovery and base fitness come from a mix of a lot of trad climbing and a lot of ARC and circuit training on my own in the past (which I did so I could get better at holding on and placing gear). This has translated into me being able to climb (some) 7a sport, despite being weaker than most 7a sport climbers...

- ...however my guess is that not a lot of primarily trad climbers do Lattice assessments (trad ethos, an anti-training mentality that is not that uncommon etc.)...

- ...my guess is also that most people who do Lattice assessments are competitive type-A sport climbers and/or boulderers who are almost by definition indoor wall wads who may also climb outside. But accordingly, the data Lattice is collecting is heavily skewed towards this sort of climber, with not much going into the system to reflect 'my' sort of climber.

- Not sure about this one, but a friend who has done a few Lattice assessments notes that in the past he never really bothered to do much red-pointing, as he had neither the mental skills nor desire to really work projects. So when he did his Lattics assessment, he put down "7a+" as his best red-point to date. Now his scores are significantly higher than mine in terms of what Lattice assess - but he's discovered more recently that actually if he really tries to redpoint, he can climb 7c. Now, if a lot of people aren't really seriously redpointing but are putting down a "red-point" grade when they tell Lattice what they climb, and which is significantly lower than what they actually COULD redpoint, then Lattice may be having their data skewed accordingly.

Anyway, what I'm trying to say is: yes I try bloody hard and yes I'm prepared to run it out even when I'm gassed out of my head, but I'm wondering if there aren't more people like me out there, and hence it's a bit mis-leading for me to present on the Lattice model as a bit of an outlier. Maybe I'm an outlier on the Lattice model - but not in the wider world of climbing, and is that something to think about (from a purely intellectual, curiosity, point of view)?

OK as part of the occasional salad but nowhere near the amount of protein you'll be needing.

My favourite sort of lattice is on a fruit pie...

I've had 1 full assessment and 2 remote assessments over the last 18 months and i'm similar to you.

I can sometimes onsight 7a and have redpointed 7c+ (3 sessions ish), 7c's (2 sessions ish), 7b and 7b+'s (2-4 goes normally).

yet my finger strength is apparently of a typical 7a climber. which seems... wrong, epecially as my finger strength is spot on for  v7 boulder (which I am ish).

my strength is my ability to do lots of hard moves near my limit (power endurance/endurance routes).

my background is 8+ years as an average trad climber

I think a lot of the people assessed are boulderers who haven't really maxed put their redpoint potential or pump out after a few sustained moves

anyway.. I pay much more attention to the FS v Bouldering grades scores which seem spot on

If an effective measure of tenacity existed it might rebalance the data*... With a higher tenacity score you need less physical attributes.

I suspect that in the "lower" grades people ability to climb rock and headgame influences their RP grade as much as physical ability.

*Although the aerobic testing does test this to an extent

I'm sorry but you have lost me. You mention lattice for poison ivy , but ivy does not require lattice support!

Funny, that's always been miles out for me whereas for route redpoint it's been closer the mark. 

3 thoughts:

- before Lattice went big more widely with the remote testing and boards and so on you had to go do it in Toms garden, so users will have either been keen and happy to travel, strong sheffielders, or both. It was also only really the medium overhang level (27deg?) which put a floor on the level you could usefully use it at. Whilst it may have expanded a lot more recently, I wouldn't personally assume that it's skewed by a lot of strong indoor folk who don't do trad. 

- some people don't really get on with the shape of the lattice edge. One of my friends had nearly a full number grade between the grade the %bw one arm hang he could do and actual perfoances on rock. [Edit, should have said, he also saw a big disparity between the lattice edge and similar size but different shapes]

- also it depends on style, right? If your lattice results are right, then whilst you may be able to do some 7a+s that play to your strengths something that's very bouldery or very intense power endurance might be beyond you. It can only ever tell you so much...

That does seem odd. Must be different ends of the range, surely? (I would have to check, but I thought the last results I had said that 8a and V8/9 were in a similar finger strength bracket).

Doesn't matter so much if they get on with it or not as long as it's consistent (and it's rounded to prevent overloading via friction I presume so you can only really hang it if your muscles can hang it).

Results wouldn't be valid for a different shape edge. 

Tim

Yep, having trained on a Beastmaker (though it turns out, not really effectively) I didn’t get on with the different shape and angle of the Lattice crimp rung at all. 

I’m not actually sure if it made that much of a difference - there is no doubt that I’m weaker than I should aim to be - but it was a bit off putting. 

You might find this interesting. A thread on UKB led to me asking if the Lattice assessment has true predictive power, or, since they ask for your details beforehand, there is a degree of post-hoc fudging of their conclusions.

Tom R answered the call and assessed me with no fore-knowledge and tried to guess my grades and strengths / weaknesses (as judged by those who know me).

The result, in short, they under-estimated me a bit. They overestimated how much my strengths (finger strength, lank) would be mitigated by my weaknesses (everything else, but mainly power, burl, core). Perhaps insufficient reckoning for low cunning and tenacity - how an imbalanced climber can often find methods (or entire projects) that suit.

https://www.ukbouldering.com/board/index.php?topic=29636.0

I'm not 100% sure I follow, but in terms of measuring your own progress I agree a consistent edge is key.

In terms of results not being valid as "lattice edge" results then also yes of course.

But the fact that your strength is benchmarked on that one particular edge means that if you struggle for some reason with the shape of that edge then it may lead to the lattice model significantly misrepresenting your abilities. I mean, ultimately what the test is trying to predict is what you can achieve; if what it predicts is woefully short of what you do achieve then it isn't you at fault! In an ideal world, the lowest result from the various bits of test data should broadly tie up to your observed level, explaining why you are where you are and what your most significant weakness is.

I don't mean this to do the lattice stuff down, incidentally, I have found it a valuable tool in the past and I like the overall training framework. All these sorts of models will have limitations.

I mean, more strength is always good. But if your profile is accurate, your 6C boulder and route goals are in vaguely similar styles (and the 6C was a boulder problem rather than being some sort of quasi route length affair) and there's no "one was a siege one was second go" sort of mismatch going on with your best grades, then on the face of it that doesn't sound like a terrible mismatch?

Not sure I follow?

if you’re going off my logbook, that 6C boulder is totally soft! The best I’ve managed apart from that is technical (ie not powerful) V4s outdoors  

i’ve climbed 6C/V5 in the gym but I trust gym grades so little that I don’t consider that as meaning anything in particular. 

Thanks, that was really interesting!

Sorry. What I meant was if your profile 6C was 

- a genuine 6C (you may have answered this!)

- in a similar climbing style to your route ambitions (i.e. not a slopey compression prow when you dream of crimpy limestone walls)

- was a length that makes it likely the moves were hard (rather than a 30 move traverse or something that's probably better assessed with a route grade)

- and was done in a similar sort of length of time to your best routes (not a 10 year siege versus a casual second go tick)

....then 6C boulder strength wouldn't feel that badly mismatched with 7a+ best redpoint - personally I'd say it would let you get a bit further than that still on sport, but it doesn't feel like a ridiculous mismatch. That's what I meant.

Having not really read your post properly, this might be relevant:

I did some lattice-style assessment, using the same theory and data (I think) and I thought the results were bunk.

Apparently, my fingers are so weak that I should be climbing V3 and redpointing 7a. But I flash a bunch of V5s every time I boulder indoors and can onsight some 7as indoors and out. And I would have said that finger strength is probably a relative strength of mine, since I do best on crimpy horrors and am shite at roofs. 

I think the reason for this bunk result is that the data they've got has come from people who are crap, like me, but unlike me, they're crap because they've only been climbing a couple of years and spent most of that time doing lattice-style training. Consequently, they've got strong fingers and shit technique, so the read across for me, having been climbing for yonks but not very hard, is poor. At higher grades, the data will come from good, experienced climbers and is more likely to have better read across since there'll be less disparity in technique.

Also, I'm probably not very good at going up and down campus boards or statically hanging finger board edges because I've never done it before. I find that when I train, I basically just get good at doing that specific thing.

Doesn't really matter though, the coach I did the assessment with was good and gave me loads of useful info for structuring sessions. I've given up all hope of improving because there are so many interacting factors that militate against it for me, that so long as I go to the wall regularly in winter and get out climbing regularly in the summer, I'll occasionally climb a route I find worthwhile. That's now about as far as my aspiration stretches - an obsessive training freak I am not.

Ah, i see. 

yeah I think I tick all those boxes. Doesn’t feel like a ridiculous mismatch to me either - it’s more that the Lattice profile seems to indicate I shouldn’t be able to do either. But I can (admittedly not all of them, and not all the time - but sometimes).

I suspect there is an element of truth in what you are suggesting. Any data set is only as good as the available inputs. I haven't done a Lattice assessment yet but I wonder if I would end up as a similar 'outlier', again coming from a trad background. I certainly tend to climb sport routes like a trad climber!

Another explanation is that 7a is a grade which you can get up with good technique and lots of stamina but only moderate strength/power. I suspect there would be fewer such 'outliers' at say 8a, where I imagine you need lots more strength, power endurance and so on.

Yeah that sounds strange, unless I am misunderstanding. Surely Lattice aren't suggesting that a 7a climber (even onsight 7a) has the same finger strength as a V7 boulderer? I generally do 7a first redpoint or onsight but can't touch V7 and V6 is the living end (this is indoor grades as I don't do outdoor bouldering - but I suspect that's not unusual for 7a sport climbers).

Even technical V4 vs redpointing 7a and soft 7a+ is hardly a mismatch.

I think you may be right there Jon. Lattice was centred around the very good and very strong, since it has marketed itself to the mainstream, its focus has shifted towards the YouTube generation. These two data sets compare poorly with talented/experienced older climbers. You can't measure experience on a finger board however with sufficient data on a climbers history, it may be possible to model an experience factor. 

I am unsure as to the value and likelihood of this modelling as I suspect the return on investment would be poor. 

Particularly since relevant experience is measured in pitches/moves, not years. Somebody who has been puntering about for decades, like me, might sound like they've been climbing a long time, but in terms of really trying hard moves on rock, I'm sure there are plenty of people who have been climbing for five years or less who have way more *actual* experience.

6C/7a certainly sounds consistent to me. I've bouldered 6C a few times - Frankenjura grades so relevant style & unlikely to be soft - and 7A once, and I redpoint around 7a/b. One of the two 7b's I've done was pure bouldering on a rope and was about 6C, so I'm assuming it's unlikely there'd be many 7b's with crux sequences any harder than that.

My experience parallels yours'. Back in the day, I onsighted over 30 7as. Although not a boulderer, I found V6 pretty much my limit. If I managed to 'fall up' anything harder, the wind must have been blowing in the right direction!

Mick

Try 7b at Rubicon...at least one is a hard-for-the-grade 7A to start. 

My initial reaction to the thread was that I disagreed that there would be a bias of the type suggested because I think most people who are keen enough to want to do an assessment are all fairly obsessed and tend to have graduated to sport or bouldering from trad. However on reflection I can see Presley whippets point that the usergroup has undoubtedly changed over time. This will be less pronounced if the model is just based off those who pay £50 pm? for a premium plan rather than everyone who pays for a lite plan or basic assessment though. 

Two questions would help determine how significant this is : do lattice continuously update their model with all the data or are they far more selective about what they use, and how much use do lattice make of data on the climbing experience people have when they set up the models? I believe they have always collected this information so if it's recorded then it's just a case of modelling different groups separately. 

Showing your age I fear. I'm sure there are many sport climbers & boulderers these days who've never placed a wire in their lives. 

All their data is useful and interesting  for sure. But much more useful are the testing protocols providing you with your benchmarks, so you can  measure your training progress. 

There will always be exceptions, and I did have in mind Tor-style nasties when writing my post earlier which provide an obvious counterpoint. That does feel bonkers mind you. I can think of 7cs that would be considered pretty bouldery that would have that sort of crux grade

To be able to do *all* 7bs then sure, you likely need a Font grade that's not very far below your french grade (7b to 7A/+ in your example, and I'm sure you can find parallels at other grades, there's a super short New Cuttings 7a that gets 6C as a boulder for example). I was thinking as a rule of thumb for more "middle of the spectrum" stuff.

I've never done any sort of lattice training or assessment but I've heard of it.

Could part of the mismatch be because it only really measures strength and power endurance which are only the limiting factor on a certain style of route / boulder problem?

For example if I'm trying to pad my way up tricky grit stone slab no amount of one a dead hangs is going to help there, most of my strength is useless.

In my case I'm quite strong but crap on slabs but if off vertical routes are your thing (as not much trad is overhanging) and you have great techy skills on slabs and a good head then I could see someone being able to climb reasonable grades while still being comparably physically weak.

(a) The Lattice data are about sport climbing grades, I suspect Tom R is very well aware that grit is different 

(b) being elderly, I can remember having conversations with a young & fit Johnny Dawes who was very emphatic about the importance of pebble-squeezing finger strength for grit slabs.

Is this the right place for this discussion? 

Would it not be better placed on one of the lattice discussion boards that they tout on their YouTube videos? 

Just to chip in, I did a full Lattice Assesment around a year ago and it was spot on with my (outdoor) bouldering grade.  They gave me a potential Sport Grade too, but as I do very little Sport Climbing, I'm a mile below my potential.  The assesment helped me to identifiy my biggest strengths and weakness, compared to their data.  It has always been up to me to apply that to the types of climbs I want to get up.  It definately helped me and now I have a line in the sand, should I wish to be assessed in future.

If you have climbed harder than their prediction, I'd imagine it's because your quality of movement, route reading, mental game, projecting skill, experience in choosing the best conditions and so on, more than offset your strength.  There's a positive here - if you can improve your strength further, that would likely open up (some) harder climbs.

I bet there are plenty of climbers in the opposite situation.  They're super strong from loads of training but their technique and experience restricts them to climbs where strength is dominant factor.

As I understand it Lattice's reference database is made up from climbers it has assessed in the past.  I suspect that most of these climbers will be physically motivated an used to training, and have developed their strength quicker than their technique.  Therefor the database and hence metric is not so relevant to many of us who are weak but have developed outstanding levels of climbing cunning over the decades.  I suspect that many Lattice assessed rock athletes could up their grade even further by shadowing older weaker climbers, maybe even swapping leads on some greasy LLiwedd chimneys?

This last part pretty much exactly describes me!

Problem there is likely only to get hardcore Lattice users, thus likely replicating the potential limitation I’m interested in! 
 

also I didn’t know they had a forum...

Their FB "forum" is full of all sorts of people, loads of whom have never bought a Lattice product / service (along with plenty of the "hardcore"!). There's a lot of good info on there, and some tripe as well.

It's been interesting following this thread. I did an assessment last week, mostly out of curiosity rather than as a prelude to a training plan (although I think there will be some suggestions and recommendations as to how I can proceed/improve my approach). I don't do any 'training' or indoors climbing, instead just bouldering regularly on the grit. I haven't received my results yet, so my comments are on what I remember we discussed at the time. 

I was having a bad day (I'm not a very consistent climber) and it wasn't surprising that my performance was very poor. My endurance was notably bad, as I only climb for short periods, trying hard until I'm tired, then going home. Definite room for improvement there. My finger strength lined up with about 6B+ boulders. My flexibility was reasonable, my core strength not bad for someone who doesn't train core. Overall, I was not surprised to find that I am a weak punter. 

How did that align with my performance on real rock? Well, last year I regularly climbed 7A boulders, usually in a session, with a handful of 7A+ and a few 7Bs (in different styles - from short and crimpy to longer and burly). I think it shows that it's possible, with some technique and persistence, to climb moderately difficult boulders without having to be very strong or fit. My goal for this year is a 7C so it'll be interesting to see if that can be achieved with persistence and getting specifically strong by trying a particular problem lots, or whether I'll have to start actually using my Beastmaker.  I suspect the later. Optimism and effort only go so far. Either way, in relation to their data, I suppose I fit in the 'not fit or strong, but wily, experienced and a careful picker of battles' camp, more than the 'very strong but can't translate that into performance on rock' camp.  

I should say that the whole assessment process was interesting and enjoyable. 

There is definitely a subset of climbers who are not particularly strong but who make up for it through experience and cunning. Often these climbers have a predominately trad background. It could well be that these climbers represent a relatively small minority of the Lattice data set, at least at the lower end of the Lattice grade range (onsight/redpoint low to mid f7s). At the same time, it could well be that the majority of the Lattice data set at these grades comprises people who are relatively strong because they train properly but who perhaps haven't been climbing as long, so don't have as much experience and cunning. Hence the Lattice results of the former group look like outliers.

However I wonder whether at higher grades this discrepancy disappears. Once you get into high f7s and f8s, you can't just rely on experience and cunning as some strength, you actually need to be relatively strong and do at least some training. On the other hand, you can't get up those grades with strength alone as you need good technique and tactics as well. So I wonder whether at these higher grades the data set is more homogenous. There will still be some outliers of course but these will be 'genuine' outliers rather than representatives of a distinct subset within the data set.

We agree.  I may be tempted to do a Lattice assessment out of curiosity, I'd be more temped if Lattice qualified their data and presumably training plans for the old and weak but talented

Yeah I'd like to do it out of curiosity if nothing else, just a matter of finding a time when I'm going reasonably well but not climbing every available weekend!

The whole outliers thing seems the wrong way of looking at it to me - the strength vs sport grade graph is probably a bell curve with more people at the central 'average' but plenty sitting at the 'weak for grade' or 'strong for the grade' ends.  Certainly don't see any real difference in the high 7s, low 8s- I'm a pretty classic weak sport climber (rp 8a, boulder 7b slab once 20 years ago) and while not average I don't think I'm that unusual either.  What does happen as you move up the grades is that the choice of routes for the weaker climber reduces - there's lots of 7c+/8as I'm never going to be able to do unless I get a good bit stronger.

This picks up on a related point - should the data be based on someone's best grade or the grade they are fairly solid at (different styles etc)? Using the best grade can in itself distort the results.

Having thought about this, I reckon that lattice data suits lattice users who are their own subset within climbing. 

Outliers such as oldies who get by using brain rather than brawn don't fit in. 

The data for those above could be easily harvested by offering freebies to the memberships of the CC & FRCC. (don't take this the wrong way guys, I am trying to get you summat for nowt).