UKC

Hi All,

I'm wanting to try some lead climbing and I'm learning about anchor building.  I'm learning using the IDEAS acronym.  In terms of gear being independent, is anyone able to ellaborate on how independent they should be?

For example if you put a couple of pieces of gear in a crack between two rocks and one of the rocks moved then both pieces of gear would fail.  I'm guessing this means they're not independent?

Thanks

You shouldn't normally be putting gear in cracks in rock that could move. But it is best practice to use independant features as part of the anchor independence when realistic

The Glenmore Lodge have some great vids on this topic, I'd go check them out. Generally when I top out I put 3 placements in separate cracks or whatever I put my pieces in. Sometimes this isn't possible so an assessment of the integrity of a crack or boulder will be needed.. Some use a 20 points method.

There's an ancronym?  Dang I'm getting old.

Yep.  Use separate cracks and boulders as much as possible.  Watch the vids - there's really good stuff out there.

And if you're sea cliff climbing and there's only grass at the top drop yer pants and clench the grass with your buttocks - you'll be a'reet.

One concept I came across on an instructors assessment is to give each piece of gear a score from 1 to 5, where 1 = marginal (e.g. a small wire) and 5 = bomber (e.g. a sling or big Hex) and aim for a total of at least 11 - this basically means you need at least 3 pieces of gear.

This may not be appropriate for all situations, but it's not the worst rule of thumb when you're just starting out - you can adapt as your experience grows.

Regarding gear between rocks, like Cheese Monkey said, try and avoid it. Otherwise, look around for clues as to whether one side is likely to move and go from there. If this is unlikely, then as long as your gear has a decent separation, then you should be ok, but try and back it up with a 3rd piece elsewhere, if possible.

Personally, I try and avoid using cams to build anchors, for obvious reasons.

What are the obvious reasons for not belaying with cams?

It's judgement two large wires or hexes in separate cracks, equalised is as close to bombproof as you need, 3 smaller wires say size 5 or less also equalised is more likely just adequate.

It's a call you need to make at the time, two size 10 wires could vary massively in different placements just by how much the rock tapers and the chance of them pulling through. Same with cams, nothing wrong with well placed cams in belay stance set ups. 

Sometimes you might have only one crack  which is in a massive face weighing hundreds of tons, if that crack widens because a 80kg person fell, then the whole mountain is coming down and you've got bigger problems. 

It's all about doing the best you can with what you face in a timely and practical manner. 

Separate features of rock, where possible. So a sling over a spike and a wire in the crack where it meets the cliff would be 2 pieces in one feature, the spike. If this failed, both placements would fail also - not independent.

Compare this to a sling over the same spike and a wire in a different crack away from the spike. If one disintegrates it should not affect the other.

I was trying not to make my post too long.

1 - partly just habit of not using them on a top rope anchor, where they could walk into an unsafe position when you're not watching them, due to rope movement. Unlikely on a belay, admittedly.

2 - if it's tied up in a belay, you won't have it for the next pitch. I've probably got 2 of most nut sizes, but only 1 of each size cam and I'm going to want that as an option on lead. If you're short of gear when you get to a stance, then use what you have and swap the cam out when your 2nd arrives with other gear.

If you can only get placements for cams, then use them.

When building a belay, you should make sure that it can handle the belayer being pulled upwards (or sideways, depending on the route) if the climber falls*. Nuts in cracks may not handle this and could pull out, whereas a cam could rotate in the crack and still hold, so some climbers suggest always using a cam in the belay if this is a risk and you haven't got a good thread/bolts/pegs (which would inherrently avoid this issue).

* - at the risk of going even further off topic, don't be afraid to adjust your anchors once the 2nd has arrived at the belay in order to better protect the leader on the next pitch.

I think the concept of 'independence' needn't be robust to the limits of counterfactual analysis. We can imagine a boulder the size and shape of a tank sitting alone on top of a crag; placing a hex under its left flank and another under its right flank would give us two outstanding placements, but the placements aren't 'independent' since both of them would be eliminated if the boulder were to move. A hex placed under its left flank and a sling or rope hitched around its girth wouldn't be independent since both of them would be eliminated if the boulder were somehow to slide off the crag. Since we can conceive an entire crag falling down, no two placements on the same crag are ever 'independent' in this extended sense; eliminating a sector of Stanage or Bosigran is sufficient to eliminate any two points on an anchor at either. Similarly, a force sufficient to bring about the entire destruction of Stanage could well have seismic repercussions in Bosigran, and its not inconceivable that whatever arrives to eliminate Stanage also dispatches the rest of the British Isles at the same time.  So for 'independent' to be a useful concept it has to be spelled out in a way that's more limited in (metaphysical) scope than 'Anchor point A and anchor point B are independent just in case were anything were to destroy or eliminate or compromise point A it wouldn't thereby destroy etc point B'. In this essay I will

Much of the time passive nuts (hexes etc) do at least as good or better job, and usually (in multipitch climbing) one prefers to save the cams for lead climber. 

In all seriousness, the trick is to not get too caught up with tight definitions of these component concepts. John Long and Bob Gaines have a good discussion in their classic book 'Climbing Anchors' about the way that the SRENE model for anchors

• Solid

• Redundant

• Equalised

• No Extension

has become bogged-down with debates about the contribution that each concept makes to our understanding of the overall anchor system, and the way that having this sort of model in mind can be counterproductive.

Since its inception SRENE has for many climbers become a sort of fixed-in-stone checklist to which every anchor must conform. This is not only a perversion of the original intention of SRENE, it is physically impossible when SRENE qualities are sought in absolute literal terms. But that hasn't kept people from trying, which over the last decade has spawned a breaking wave of commentary based on everything from Newtonian physics to blue wind. Those favoring this or that aspect of SRENE have squared off like Sumo wrestlers, charging each other with the passion of True Defenders of the Faith. But like everyone married to doctrine, they find themselves yarding on crumbly holds because when considered as inflexible doctrine, SRENE embraces mutually exclusive qualities. For instance, with hand-placed gear you can rarely if ever achieve simultaneous non-extension and equalization. And in the real world, redundancy is not always possible - though always desir­able. Mountaineer and engineer Craig Connally, in The Mountaineering Handbook, puts it this way: "If static equalization invariably boils down to redundancy rather than true equalization, and dynamic equalization doesn't lead to true shock load­ing, the old acronyms [SRENE] Ieave us adrift. What then is a reasonable, realistic approach to building anchors?"
The only "reasonable and realistic" approach to building anchors is to leave off considering SRENE as an absolute doctrine or checklist, and to use SRENE for the purpose for which it was originally devised: as an anchor evaluation methodology. (p132)

While the SRENE model doesn't explicitly include 'Independence', it does include 'Redundancy', which aims to capture pretty much the same conceptual territory albeit with a different term.

Here's what Long and Gaines have to say in starting their discussion of Redundancy (p135):

Redundancy, as a valuable quality in anchors, became widely standardized shortly after the first edition of Climbing Anchors. Since then the subject has drawn a hail­ storm of arguments and counter-arguments about A) what redundancy actually is, and B) the relationship of redundancy and security. Some argue that redundancy and security are synonymous, that one is impossible without the other. This is pas­sionately denied by others, though no one questions the overall importance of redundancy. Since you'll encounter this subject time and again, both in print and in the field, let's review the basic debate.

All anchors consist of various links of primary placements and rigging, and the strength and security of those links will always vary within one anchor matrix. To counter this, we bolster the multiple links within the anchor system through redun­dancy. Basically, redundancy demands that anchor systems be constructed of multiple components — from the primary placements, to the slings and biners we use in rigging the placements together — so that if any one component fails, the anchor will not fail. In other words, a redundant anchor never stakes our life on one piece of gear. Yet in several instances we do just that. We (almost always) climb on one rope, tied into one harness, with one belay/rappel loop, on which —when needed—  we use one locking biner connected to one belay device.

This hurls us onto a slippery slope. We can't trot out redundancy as an inviolate rule, and break it at the same time. The question becomes this: Can security only be achieved through redundancy? If so, why do we climb on one rope, one har­ness, etc. If not, where does this leave us in terms of anchors?

[Celebrated American climber Curt Shannon puts it this way: "Boeing 747 aircraft do not have two left wings in spite of the fact that if the single, non-redundant left wing comes off, everyone on board will die. Same thing with the use of a single rope. The point being that it is possible to design an element of a mechanism (such as a wing, or a rope) to a standard of quality where a lack of redundancy is a moot issue."]

Simply put, a fail-safe anchor, not redundancy per se, is the ultimate goal, and redundancy is only one important tool to achieve that goal. Stated differently, we build fail-safe anchors not to ply a rule (redundancy), but so we don't die. If redun­dancy helps us build fail-safe anchors, as it almost always does, then we use it whenever possible. But experience will eventually show us that a viable anchor can be achieved, however rarely, without letter-perfect redundancy. Furthermore, in the real world there are places where even the most experienced climbers cannot make redundant every facet of the system, and to claim and insist that you can, and must, is false and misleading.

It's a really readable book, encouraging and thought-provoking in precisely the way that maintains engaged and practical attention, and it's filled with lively examples although a lot of the examples involve things not going quite according to plan. Definitely worth tracking down a copy (*cough* lib gen dot is *cough* ) if you're wanting more tools to help you think about / build / evaluate anchors. 

I'm with you: I still think John Long's books on anchors have never been bettered. Certainly when the early editions of his instructional books came out, they were way better than any British offerings, IMHO.

My own version of SIRENE is FORTE, which means essentially the same thing. (Four, Opposition, Redundancy, Taut and Equalised.)

I suppose, four and opposite is for multi pitch anchors?

Oy, there's RENE, SRENE, SERENE, SERENEA, EARNEST, IDEAS, BEARME and god knows how many more word salads.  These don't all add up to the same set of standards, and some of the prescribed features, notably "Equalization," are not obtainable in the field and often not even in theory. The fact that the different acronyms don't always agree on what the critical features are and in some cases seem to exclude perfectly good solutions does cause one to wonder whether these oversimplifications always capture what really matters.

At the other end of the holy catechism approach is the Jim Titt Theory of Anchor Construction, to wit "place good pieces and tie them together," which, perhaps surprisingly, is actually going to work in the vast majority of cases in the field.

I suppose the word salads have their uses, although, as a superannuated climber, I and most everyone I know learned before acronyms became fashionable. We looked at our placements, ascertained the direction of the load, and "tied everything together" so that, to the extent possible, said load was distributed to each of the anchor pieces.

As for redundancy, I think Long and Gaines are perhaps tripping over their own feet, having been subjected, over many years, to a tsunami of arguments on the web and social media. You only fall victim to the contradictory thinking they mention as problematic if you try at the outset to enthrone "redundancy" or "equalization" or "independence" as commandments whose violation will immediately trigger the End of Times.  The result can be a Redundancy Taliban for whom no scenario is too implausible and all deviations are fatal.

The reality is that redundancy is a principle, not a commandment.  It is indeed important, even critical, as a hedge against uncertainty.  Curt Shannon's point about a Boeing 747 only having one left wing needs to be viewed in the context of how many other redundancies are present. The Airbus A380 flight control system has five flight control computers, although only one computer is needed to fly the aircraft. The plane can lose four of these computers and still be flyable.  The loss of a wing is too unlikely to demand redundant solutions, but the malfunction of computers is likely enough to require significant redundancy. The situation with anchors is analogous, but there we have a spectrum of judgments that no word salad is ever going to elucidate, judgments that are influenced by the situation and also the party's tolerance for risk. 

Beginners in particular and the rest of us in general have to have the humility to realize that our judgments are not always on point, and so redundancy is the best and maybe the only tool we have to manage risk intelligently. In many situations, especially when time and gear consumed are factors, a two-piece anchor is appropriate.  But when there is no time pressure, and when there is no danger of running out of gear, or when the person constructing the anchor is new to the process and so less able to judge the solidity of individual pieces, then three pieces make more sense.  it can be annoying, on, say, a two or three-pitch crag, to climb with people who insist on protocols appropriate to big routes in Patagonia.  And conversely, the kind of redundancy that is appropriate on small outings can become a dangerous time-suck on a big alpine endeavor.  The word salads don't cover this either.

So having come of age without acronyms, my advice is to forget about them.  Instead, read the anchor books, ones with lots and lots of pictures of anchors, and in each case make your own judgment before reading what the authors have to say.  This will build a far more robust experiential background then memorizing some catechisms and applying them willy-nilly.

But you haven't asked for any of this.  Your question is what "independent" means.  According to the videos and websites I checked, it means that each piece is connected to the belay point by its own strand.  It doesn't seem to refer to the idea that using different cracks might be a lot better than putting all your irons in the same fire.  It also seems to outlaw the time-tested practice of rigging pieces in series, and if so richly deserves a place on the ash heap of questionable proclamations.

I think there is clearly loads of redundancy in many peoples systems here, if you apply critical analysis, single weakest point, static system safety factor etc.. you just don't need 4 good independent anchors equalised for a climbing stance on sound rock. 

Just look at the maths, 2 large wire, a wire and a spike, wire plus a friend, add their strength then drop a third for the knot (although you can reduce this if you clip behind the knot).... anyway you're still looking at 1200kg / 12ish kn minimum, to generate those loads through your harness climbing you'd be dead anyway. 

With 3 good anchors equalised in sound rock, high strength tie offs etc to avoid losing system strength you'd lower a stretcher, a casualty and 1-2 helpers, with a back up rope attached to another 3 anchor placements of similar quality. 

The safety ratio with 3 good placements is huge. A combined strength of 30kn(ish), you're getting close to the point that industry rated pre stretched 11mm ropes (usually  36kn) will be in danger of breaking (the 9.5-10.5mm ropes will have failed below 30kn unless you're lucky), individual slings and extenders would have potentially failed at 22kn, climbing ropes possible 20kn plus, climbing harness belay loop is rated min 15kn, waist belt 12kn(uiaa minimum standards).... anyway you're see my point building anchor systems in good rock, with large gear with a combined strength pushing 3 tons is using up climbing time and causing some here to potentially carry far more gear than they need.

It's about judgement not quantity with anchors. Save the 4 anchors system for rock that has the consistency of crumbling wensleydale. 

I'm going to commit heresy now and advocate one-piece anchors.

These are sometimes appropriate - if you've got a massive (but not sharp edged) spike on a big ledge, then don't waste your time looking for that obscure second piece if there's nothing obvious and quick in front of your face.

If there's a subsequent pitch you may want to then add something for an upward pull, but for most climbs/pitches where you're just bringing up a second, the potential forces are so much smaller that your own body weight stood/sat on a secure ledge or top, is in effect a second anchor anyway.

There are even times when no anchors is ok. I can think of places where there's no obvious belay points and it's perfectly safe to walk 10 yards back and sit in the grass or heather taking in the rope - probably best to communicate what you're doing with those below before you walk that 10 yards back.

The key thing in all of this is judgement which of course is facilitated by experience. So whilst beginners should find out what makes belays safer (and conversely more dangerous), nobody should get hung up on rigid rules.

Remember what a belay is for, it's to keep the belayer on (or very near) the ground in a way that ensures they can continue to safely control the rope.

It's fine, however I have seen some use single sling anchors on the top of places like Stanage that a light breeze could move. But a single giant boulder or spike is often extremely safe. The same with dubious threads between boulders that aren't very large. 

Yeah, or just walk around any large boulder and it's like a direct belay, it's just about judgement and knowing you don't need to see your second on the pitch as you are both happy with each other abilities. 

Agreed, though not as often on UK trad.

One of the things that took me off guard as a trad climber learning in the alps for the first time was how many one piece anchors and dodgy belays high level alpinists used. 

We were guided on our JCMT course by a high level alpinist who used wiregates on bolts, single slings around spikes and occasionally a single cam for a belay. He was not being reckless, he just understood the forces at play and knew that these anchors were more than acceptable for the intended use.

As other people have said, SRENE is a useful evaluation tool, especially if you are a beginner, but the ability to respond to any belay stance and be able to create a safe enough anchor is worth far more than memorising an acronym.

Perhaps. Or it could be to hold the weight of the second while the leader escapes the system and provides / seeks assistance. Or it could be to hold the weight of the climbing team plus any dynamic forces that are created by a leader fall (hanging multi pitch belay).

Growing up climbing on Southern Sandstone before the equipping of many of the routes with bolt belays you could spot visiting proper climbers a mile off.

They would build nice three point belays using three full size pine trees.

BITD people could hire a complete climbing kit for Harrison's in the Groombridge shop: polypropylene rope, polypropylene tied sling, steel screwgate. The rope usually had to run over the rock at the top and back down to the belayer, this could sometimes be avoided by running it directly over a tree on the lip (sorry, tree). A real problem was the rope jamming in the multitude of deep ropeworn grooves through the edge at the top. No regard for the rock or future generations.

Thankfully there is now the strict sandstone code for anchors etc. Here anchors must protect the rock as well as the climber. 

I hope you are finding the debate interesting!

Personally, I try to keep it simple and aim to have 3 solid placements, ideally in separate features of rock, equalised so that if one comes out the anchor point won’t move, carefully orientated in the direction of pull so if the climber comes off the belayer won’t move either. There are various ways of rigging you could use to achieve this.

Not every belay is easy to set up like that and sometimes you may have to compensate- more placements if the rock is friable, for example.

Your best bet would be a practical session with an instructor if you can arrange one. The consequences of a belay failing could be very serious so it might be a shrewd investment.

Interestingly I think that's the only time "Taut" has been mentioned in this thread. Easy to be pulled off a stance, even with bomber placements and load sharing, if not tied on tightly.

My Uncle first took me climbing there and described what they did in his youth, just wrapping the rope round a tree at the top, no sling or carabineer at all. He said the benefit of this was that there was so much friction you could belay one handed whilst drinking (beer?).

Obviously this was one handed body belaying which I think just consisted of holding the rope with one hand.

As you say, even with the relatively tiny numbers of climbers back then, this trashed the trees and rock. 

2 - if it's tied up in a belay, you won't have it for the next pitch.

Which is why you should try to carry double reds!

I am glad you expanded on your original statement. Taken on its own it implied to beginners that there were obvious reasons why it was unsafe to belay with cams, which I thought was misleading.

I generally agree with your expanded statements but, to my mind, the circumstances that favour using cams to belay are slightly broader than you describe. I'm probably just being pedantic but for me:

1. There are many placements where cams are unlikely to walk or won't get stuck even if they do. Walking is mostly associated with runner cams and rope movement.

2. It may be obvious that some cams will not be needed on the next pitch. Examples would be big cams left over after a wide crack pitch when the next pitch obviously only needs small stuff or small cams below a hand crack).

3. Even where cams might be useful on the next pitch it may be more important to use them to ensure a safe anchor at the stance (so use them where necessary to make the anchor feel safe not just if you can only get cam placements).

4. Resisting an upward pull can be important and is often easier to arrange with cams.

My thought  process is usually this, and then considering how the persion I am climbing with will react if it is two or even one piece, the rope around a massive boulder for instance. If it is someone I don't know well I often stick in an extra piece for show. 

Edit: the above mainly applies to belaying a second to top out, not advocating this is good practice on a midway belay on a multipitch climb where the direction of pull can vary. 

Climbed with someone from UKC recently at Avon, topped out at a tree belay and ended up tying three separate slings across the two trees just so I wasn't seen as unsafe.

Bit stupid in retrospect, especially as it was on a massive ledge. 

I don't think there is any substitute for experience when it comes to building anchors. Climb with someone more experienced and observe their setups and ask.

Four points Jeremy? That’s insane

Done well the waist belay has one hand on the active rope which then passes behind the belayer and is then held by the braking hand with a twist round the wrist. Lots of friction.....the braking hand can be moved across the belayer in the event of a fall. Pretty easy to hold a bottom roped climber even without an intervening tree! Arresting a leader fall can be surprisingly easy when there are runners.....one should definitely wear grippy gloves to avoid bad rope burns ( many did not do so) and a consequential risk of dropping the climber.

Depending on the size of the trees, if on of them were to fall the rest of the belay would be ripped apart like strings. Making this kind of redundancy redundant. 

As an instructor I wouldn't advocate anything for show, just explain how strong a large wire or sling are. Once you show the strength rating on the label and start explaining it's the equivalent of 1 or 2 tons static load, they'll grasp that some thin looking wire etc.. is surprisingly strong 'if' placed well.

Completely agree. I think it was more as a failsafe in case one of the slings failed, which is stupid as the tree was more likely to fail given the two double wrapped slings. The things you do when you're scared.

One thing not mentioned about cams in cracks and boulders is the force multiplication of a cam is larger than you'd think. an 2.5kn load on the rope could put a half a tonne of force outward on the boulder. Think of sticking a car jack inside the crack and ask yourself if it might move something. 

When my brother taught me he always used to say any boulders you sling make sure they're bigger than a coffin or you're going in one..... Very grim, Make sure they don't move is always a bonus!

Try not to rely on placements next to each other or in the same feature. Having said that if you're at the top of Stanage and there is an obvious bomber crack you can get some deep nuts or cams in it will be absolutely fine. A lot of anchor building is practice and being able to recognise good and bad placements. Could be a bomber anchor but if another bit of the anchor fails under load would see you in a unsafe position etc.

I mean, why we don't just bolt Stanage I'll never know........

Agreed, but the people you're climbing with and the location and nature of the pitch/route will largely determine whether you need to be thinking about possible escape from the system when setting up a belay. Certainly unlikely to be needed at Stanage.

Hanging (or even semi-hanging) multi pitch belays, completely different ballgame 😁

MIT did a paper all about cams some time ago, unless you're on blocky loose stuff the forces generated aren't going to split normal cliffs in half, there's always the case that a cam would hold if a block crept open a little, but a wire would fall out. 

Don't overcomplicate this. Practice on super easy low grade trad, which (most rock types) generally means a good choice of independent placements.

Although there is some theory, common sense and practice/experience are the biggest factors. Just get out on the rock in safe and easy settings and practice.

Yeah, the paper is fairly straight forward, you can work things out easily 'visually' by seeing how far you need to pull a cam in order to open it up a certain distance. The ratio is the force multiplier and it's about 2-3x depending on cam design.

I was mostly talking about big boulders and leaning blocks or flake, where even if they look huge, they can still move a bit. There's some visual examples on hownot2 where they've had cames inside cracks that have three foot of rock above them (think big car sized) and when pulled, they lifted the rock fractionally. Fine if you're fairly well cammed, but an undercammed placement could well slip. I'm not suggesting anything complicated - just be aware that forces are more than you might expect.

I can think of places where there's no obvious belay points and it's perfectly safe to walk 10 yards back and sit in the grass or heather taking in the rope - probably best to communicate what you're doing with those below before you walk that 10 yards back.

Gives them a chance to run away I suppose?  

Totally agree, even the best gear has limitations if not used as they were specifically designed for or placed in less than solid rock. Those boulder fields and scree below crags only came from one place! 

Correct. Two pieces in a splitter crack are fine, two pieces in cracks around the same block are suspect. Likewise, two slings on the same spike aren't independant but the spike may be so big and solid you don't actually care.

Jk 

Yes. Of course, this can be scaled back for really good anchors, like two bolts, or one stout tree, and often the opposing piece is unnecessary. A good belay station is one you feel you could hang a car off. 

'Taut' is similar to the No Extension in SIRENE.

A quick google shows the average car weighs 1900kg, so two size 8,9 or 10 wire, equalised with a 22kn dynema sling should just do it. Given under static loads of 100kg (full laden climber), it's a static system safety factor of 20:1, some what more than any working at height regulations would require.

One anchor would be more than strong enough to cope with any of the forces involved in climbing, but and it's a massive one, you'd have no redundancy in your system, hence why you need two wires minimum. 

What has not been elaborated on yet is that rock is a natural material , in the real world  wired nuts and cams tend to rip rather than break as a failure mode.

Rock tends to be brittle, fractured, both obvious and concealed, and variable, what can possibly go wrong?

Best not to get too excited about theoretical breaking test results.

A long time ago, I think at Plas y Brenin, I was introduced to a similar system using a 1 - 4 scale. 4 was bombproof and 1 was marginal. The suggestion was they should add to 7 or more. So two bombproof anchors were fine (but not just one - in case you had made a bad assessment).  I remember being a bit worried then that this meant you could make a satisfactory belay from 7 marginal anchors. So I considered an alternative retaining the 1 - 4 scale but making them multiply to make at least 12. The outcome is much the same except that marginal (grade 1) placements never add anything to the total.

But in practice, of course, you make a judgement based on all sorts of criteria. Yes, I do occasionally accept one anchor (eg enormous trees - especially if you're then going to abseil from them). And I'll always take account of the position of anchors in relation to the direction of pull.

I don't think there's any reason not to use cams for belay anchors provided you're keeping them under tension so they don't walk in the placement (except, as you clarified later to save cams for the leader if you're multi-pitching).  I regularly use them. 

Martin

Throwing in my rules of thumbs:

• Minimum 2 solid placements
• Minimum 20 kN total
• Max 90 degrees preferably 60
• Boulders: Min 1 m^3 and not rocking (~coffin size)
• Trees: Sound and well rooted
• One tree: Diameter of a helmet
• Two trees: Diameter of a forearm

Of course. Hence why I always said good placements, if it's blocky limestone, flaky  etc then it's a judgement call, which I also said. There is no substitute for reading, learning and above all mileage, trying different gear on different rock types and so on. 

…

To the OP, Stefan makes a good point here. Your rigging angle significantly affects the force on the anchor. Counterinuitively, rigging angles >120’ transfer 100% OR MORE of the load onto each anchor. In practice, this is mediated by slack and rope stretch but you should still aim for angles to be acute where possible.

see the graphic for numbers: https://www.ropebook.com/information/vector-forces/#:~:text=As%20the%20inte....