Bringing up a second, Belay styles

I’m trying to educate myself on belay styles when bringing up a second climber. I’m still learning so please correct me if I’m wrong. When bringing up second on steep alpine route what belay systems do you use?

Indirect belay (Being clove hitched in or otherwise attached to anchors and belaying second up directly from belay loop.) Seems most simple.

Redirect (Belay second up from belay loop redirected through anchors/master point.)

Direct belay (Belay on anchors or master point. Example: Belay with atc in guide mode.) Seems most safe in event of belayer accident.

What are the pros and cons? Does your technique change when climbs get steeper or more complex? How do you back up these systems?

I almost always use an ATC in guide mode directly off the shelf of the masterpoint. This isolates you so that if your follower falls, you're not taking the impact to your harness and frees you to move and adjust while belaying. Belaying off the harness is useful for marginal gear anchors where the belayer becomes an additional point in the system to absorb some fall force (should not be terribly high while bringing the second up, but if they immediately lead from there).

A redirect offers much of the benefit of a guide mode belay in that the belayer can move slightly more freely, but will still be taking impact from any falls.

Hope this helps some

Not "alpine" specific but...

belaying off harness - i like this when i am able to sit on a ledge with my legs hanging off or some similar situation (or if rigging a direct belay is more of a PITA than it's worth). I often turn the belay device (atc style) backwards so i am feeding the brake strand up and into a pile beside me if I am doing this method. if you flip the device, one thing is to remember to pull up to brake instead of down. Squishing your soft parts can be a real problem with this setup.

redirect - i don't like this one in general cause it tends to pull me towards the anchor, thereby un-weighting the anchor or cycling the load on it more than i want. Sure, it works well sometimes but i try to avoid it. [x[Maybe in a hanging belay situation this would be good for helping reduce the (negligible) TR fall-forces as the belayer gets pulled toward the anchor?]x] -- ((edit this isn't true, pay no attn to me here. Thanks Kyle, Wivanof, etc. - jm))

direct off anchor - this is my preferred method, esp with the atc-guide style device. generally easy to pull rope through, allows independent adjustment of the belayer and the device (to some extent) and provides some level of braking assist in a fall (ie, the climber strand applies friction to the brake strand). The drawback is that it can be slightly difficult to feed slack through the device if the climber is weighting the rope.

I don't climb too steep (ie, i top out with 10s) so I cannot speak to that portion of your question. Good luck out there. Keep learning!

Advantages of Indirect: compatible with low anchors, compatible with crappy anchors, easier to give slack
Advantages of Redirect: easier to create tension to assist the follower, eliminates belay transition if switching leads, easier to give slack
Advantages of Direct: arguably safer, easier to take photos, easier to escape, easier to convert to 3:1

Serge Smirnov wrote: Advantages of Indirect: compatible with low anchors, compatible with crappy anchors, easier to give slack
Advantages of Redirect: easier to create tension to assist the follower, eliminates belay transition if switching leads, easier to give slack
Advantages of Direct: arguably safer, easier to take photos, easier to escape, easier to convert to 3:1

Except when its' not arguably safer, right?  For example, low anchors, crappy anchors (or no anchors), two concurrent followers but ropes come out in different directions, etc..

Only ways I can think of it as being safer is either a poor belayer, belayer fatigue or injured belayer (e.g., gear dropped by another party above).

Great responses everyone. It is giving me a great and straight forward pro and con list.

What about when you use half ropes? This is what I normally use. Often, I will clip one strand to one anchor (clove hitch), clip the other stand to the other anchor (clove hitch), and adjust for comfort and equalization. Then belay off my harness (indirect) Is there any issue with this?

Michael Tilden wrote:... half ropes? ... clip one strand to one anchor (clove hitch), clip the other stand to the other anchor (clove hitch), and adjust for comfort and equalization. Then belay off my harness (indirect) Is there any issue with this?

For simply belaying the second, no problem.  But towards continuing the next lead ...

If by "one anchor" you mean a bolt:  I'm a gumbi who's team usually has a couple cordalettes.  And often by default we try to do the usual SRENE or EARNEST anchors.  So my first thought was, the second comes up and clips into ... what?  They might do the same as you with their rope ends if the bolt flanges are big enough for two biners.  But some aren't.  Perhaps a non-gumbi anti-cordalette type will chime in.  :)

If by "one anchor" you mean to one piece of gear:  There's usually more flexibility to squeeze in another biner.  However, I'd get a little nervous with needing to fiddle with something in close proximity to a piece of gear when all I have is two pieces for the both of us.

If the anchor is solid, tie off the second’s rope to it and have them follow toprope solo on a couple progress capture devices. While they do that, haul the pack, eat, drink, take photos, masterbate, or do whatever it is you feel like doing while they climb. Don’t belay them

https://www.climbing.com/skills/advanced-techniques-follow-on-toprope-solo/

https://youtu.be/QO3jic7GQ7Q

Belaying off the anchor is a great way to go. Using a Petzl Grigri or a Kong gigi works fantastic. The atc guide has quite a bit of friction in guide mode

stolo wrote: The atc guide has quite a bit of friction in guide mode

This may be a matter of rope selection?

I find double 9mm's to work like a dream with ATC Guide in guide mode.  Our usual usage is with two concurrently climbing followers.

I have found the friction with two larger ropes (9.5 or 10.0 mm?) in guide mode to be a pain in the a$$

Kyle Tarry wrote:

Redirected belays double (ish) the force on the anchor because of the so-called "pulley effect" (i.e. there are two weighted strands pulling on the anchor instead of one).

That's true.  

But, so what?  He's essentially top roping here. And don't you build your anchors to (potentially) take a FF2?

Kyle Tarry wrote: I'm not making any statements about anchor strength or fall forces, just that what was posted about a redirect belay reducing force on the anchor is fundamentally wrong.

Yeah, sorry. I should have read the comment you were replying to instead of just yours.

JFM, you're wrong. A redirect does not reduce the TR fall forces in a hanging belay or otherwise. The redirect increases the load as Kyle stated. Even so, it doesn't mean much. Basically, same as a TR.

If you're climbing where the anchors are sound bolts, consider this: https://vimeo.com/44869774. My wife and I use this method, with Munters, for swapping lead on multipitch. You can belay your second up and she can continue up the next pitch on lead. Not suitable for trad.

Kyle Tarry wrote:

Redirected belays double (ish) the force on the anchor because of the so-called "pulley effect" (i.e. there are two weighted strands pulling on the anchor instead of one).

 Not exactly.  This is a misconception or misunderstanding of the pulley effect in this situation.

Sure, if the belayer is not hanging on the anchor at all, standing on a large ledge and then has to use his body weight to counter the redirect, then it is true, approximately double the force on the anchor.  The important point here is if your anchor is not good enough for double the force of a top rope fall, get the fuck out of climbing immediately.  Go straight to golf.  

But...

If your belayer is already hanging on anchor,  then, follower falls and hangs on anchor, there is no magical pulley effect that somehow yields double the force since the total load on the anchor is the belayers weight and the followers.  In this case the force on the anchor will be approximately the same with a redirect as a direct belay.  

To the OP, all three options have their place.  You should learn them all and understand which is best for each unique situation.  

So many options...and, for me it kind of depends on circumstances--I don´t always do the same thing. Some examples:

--Trad - I usually have three good anchors with a cordelette and master point (cordelette material is thinner dyneema sling rather than cord, for bulk issues). I clove into the master point with a locker. If using double ropes I do the same thing. The Yosemite anchor is another good option if just using the rope: https://www.climbing.com/skills/learn-this-build-a-climbing-rope-anchor/ 
--Bolted anchors, party of three - I use a pre-prepared quad on the bolts and will clove into the quad with both ropes (or one rope and PAS), belay the followers up simultaneously in guide mode (double ropes). I'll carry the PAS to facilitate any untying for lead swapping.
--Bolted anchors, party of two, not swapping leads - Same as previous--the quad. If I think the follower will fly up the pitch no problem and there is no rockfall potentil--redirect/belay off of harness. If the follower might be hanging or there is a possibility I might get knocked out, then guide mode.
--Bolted anchors, party of two, swapping leads, one rope  - I clove into one bolt, leave a loop of slack, clove into the second bolt. In the loop I'll tie a figure-8 if wanting a master point and/or for guide mode mode. In this situation of swapping lead, with two good bolts, I prefer a redirect/harness belay since it makes the transition to the next pitch really fast. Often, here in Catalunya, there will even be third massive bolt nicely placed for a redirect off of a quick draw. Sometimes, I may take the rope coming off the bottom of the second bolt back to my belay loop so as to have both bolts to hang directly from.
--Bolted anchors, party of two, swapping leads, double ropes - I clove one rope to first bolt, the second rope to the second bolt. Then I tie an overhand or figure-8 in the two ropes coming off the bottom of the bolts to make a master point if using guide mode. Again, redirect to ease the belay transition unless the second will be dangling or there is serios rockfall potential.
--Speed/alpine belay...if the terrain is easy, when speed is important, the second will be moving fast, and the only kind of fall I am likely to catch is a second who slightly loses his balance for a moment, I'll use a body belay...usually anchored, but sometimes merely with a big immovable boulder between my legs, taking care to be sure of the direction of pull/possible pendulum, etc in case of a slip.
--And, yes, sitting on a good ledge and watching a follower is nice sometimes...anchor runs to my belay loop and I belay off my harness trying not to get a leg caught under the rope going to the second...straight line and no slack from anchor to belay loop to follower so I don't get pulled anywhere and the anchor takes most of the weight of any fall.
--Another option if the anchor is way back behind the cliff edge...munter off of the anchor, lower myself to the cliff edge where I want to belay, tie a BFK in both lines of the rope and use that as a master point for either guide mode or a redirect or even to clip into my belay loop for belaying directly off my harness.

Anyway, I am still learning and still trying to refine all of these techniques. If you ask again in a year, I may be doing things differently. I guess the point might be to have a number of ways to skin the cat then applying whatever is most efficient/safest in the right circumstance...and be open to new techniques.

Here's my outlier's perspective on the three methods.

Indirect belay. For me this is the standard belay and I use it 90% of the time with trad anchors.  I use a modified method more common in the UK than in the US, in which the belayer's tie-in is rigged with no slack at all and the belay device is clipped to the rope tie-in loop, not the harness belay loop.  This has two advantages. (1) Main point: the load from the second is transmitted directly to the anchor.  No pinching or twisting load is applied to the harness, making it comfortable to hold a hanging second for as long as necessary, especially if the belayer is using an assisted-braking device (which I do). (2) Minor point: the anchor is to some extent "protected" from high loads by the shock-absorbing effect of the tie-in.  

The rope-loop placement makes the indirect belay more comparable to the direct belay in terms of comfort, while preserving all the other advantages of the indirect belay (distinctly easier taking-in of the ropes, ease of lowering with none of the complicated work-arounds required by guide plates, none of the dangers of mishandling a guide plate lower, and none of the occasional but vexing problems caused by a guide plate jamming in a corner or on some other feature).  By the way, I have no problem belaying two seconds climbing simultaneously with an indirect belay.

Redirect belay. I rarely use one. Usually, the belay stance would have to be fully hanging in order for the advantages of the face-in position to outweigh the disadvantages.  Also, if the second is traversing horizontally to the belay and facing, basically, a leader fall, a redirect is better than either of the other options.  Otherwise, you get the friction taking in the rope of a guide plate, you get the pinching load on the harness in a fall, and you get pulled into the wall and are pinned there if the second is hanging.  Some self-rescue transitions are easier, but I think that, considering the rarity of any kind of self-rescue need, choosing an otherwise inferior method to modestly facilitate a very remote possible application is poor strategy.  

Direct belay (with guide plate). I use one maybe 10% of the time on trad anchors.  The main advantage is, of course, belayer comfort and convenience, since the belayer is independent of the belay load. This is bought at the expense of continually pulling on the second by most of the belayers I've experienced, and epicondylitis for the belayer from hauling ropes through the devices.  (Because of their frequency of use, guides are particularly susceptible to this overuse injury.)  

In addition to the well-known lowering work-arounds and dangers, the guide plates foster inattention (typically celebrated as a major advantage) and so increase the likelihood of a second experiencing a leader fall while the belayer is rooting around for their sandwich in the pack or putting on their wind shell.  

The plates also seem to me to influence choices of anchor and belayer placement that are adapted to the convenient positioning of the plate rather than meeting the demands, especially the communication demands, of the environment.  So, for example, the belayer uses a plate on a tree back from the edge and ends up with little or no communication with the second, when with the same anchor and an indirect belay, they could sit at the edge and see and hear their climber.  

When all these are taken together, the direct belay has the most, not the least potential dangers, and this is without mentioning various gotchas, eg that some devices (Jul and Gigi) with a single rope need the braking carabiner in a special position to keep the strands from swapping places in the device slot. (As an aside, the idea that one can quickly rig a 3:1 hoist from a direct belay is misguided, as the friction through the plate and final carabiner reduces the mechanical advantage to about 1.7 : 1).  

 All that said, direct belays are very nice on stances that have been convenience-bolted for that application, and more an more climbs are being equipped in this way.

Manual belay off the hip, from above, is a pain in the ass and can be quite sapping of energy because of the ergonomics.

Kyle Tarry wrote: This seems like an odd critique.  There is zero reason that the belayer needs to stand adjacent to the anchor with a direct belay; you could set it up exactly like you suggested, and then sit at the edge and belay in guide mode (or Grigri, or whatever your choice of direct belay device is).  You can easily do your belaying 15 feet from the anchor with a typical setup.

Maybe it is odd, but I see it regularly---it isn't something I made up.  Moreover, if you leave the plate on the anchor and position yourself 15 feet away, you've given up all ability to pay out slack (without running back to the anchor). So you can observe what the second is doing but have lost half of your capacity to respond.

I think that this is a very subjective claim.  The only significant risk that you list is those associated with lowering, which I agree are worth consideration.  The rest are things like inattention and poor rope management, which can happen with any belay system, and are a function of the belayers behavior and not the type of system being used.

It is subjective, but no more subjective then the claim that the indirect belay is safer.  And saying that inattention and poor rope management can happen with any system overlooks the fact that poor rope management is much easier with a guide plate and inattention is one of the main promoted "advantages."  As a second, you tend to to get a dose of both: excess slack because the belayer is busy with non-belay tasks followed by too much tension as the belayer tries to recover.

Sure, but the reduction from "ideal" ratios is true for pretty much all systems using carabiners (instead of pulleys) and typical progress capture devices (for example, this is a well-known limitation of crevasse rescue haul systems).  The direct belay system still allows for setting up a haul significant faster and easier than any harness belay system.  I do agree this is not that significant of an advantage, really, because these scenarios are uncommon enough to be akin to the self-rescue applications you pointed out.

The friction reduction is worse with a guide plate than with just carabiners, about 1.7:1 vs. 2.1:1.

In the end, the reality is that all of these systems are totally and completely safe and acceptable when employed properly and appropriately, and some nuances of personal preference may often be the most significant factor as to what is chosen.

Yup, I was just airing my nuances.