I’m interested to hear some folks best or common practices and methodologies for how they think about horizontally aligning (dynamic group in Dorico) dynamics especially in parts. I know there’s not a one single formula that can be applied here, but because one often times has to choose between horizontal alignment on a given system versus awkwardly large gaps between dynamics and notes it seems like there might be some sensible rubric others have learned engraving over the years. What makes this trickier in Dorico is that dynamic grouping cannot be layout specific; it applies to all layouts (yes, you can hand bump elements in Engrave mode per layout). Thanks for any thoughts and musings.
Can you provide an example or two to help us to better understand? I’ve not encountered issues with the horizontal alignment of grouped dynamics. So, if you are referring to the distance from the staff as a default, then that is easily made universal using Engraving Options->Dynamics->Vertical Position. If you save that as your default, then it will apply to all new projects (maybe new Flows but I’ve not checked that yet).
As for being Layout specific, no it is not and I would not necessarily expect Engraving options to be so. Most of my friends colleagues in the Engraving world tend to keep these options consistent across their work. I am open, though, to other perspectives. So, my musings await your interaction. 
There is no problem here, it’s more of a style or visual aesthetic question, and I know this is universally an “it depends” answer that has to be evaluated on a case-by-case basis. But using this as an example of such a case:
You can see on the first system how dynamics are aligned except for the decresc. in the last bar. You could obviously bring this into alignment, but then you have an excessive gap between decresc. and beam.
Similar situation with the second system. You could align the boxed dynamic with its pair, which would create a large gap, or you could align it with the larger group of dynamics towards the right.
I just show this as the types of decisions engravers have to make to get general feelings or ways to think about making the right decision, I’m not asking for everyone’s opinions on how to process this exact sample. Hopefully this helps.
Got it. Yes that does create some questions on uniformity as an engraver. And yes, “it depends.” I tend to think that when dynamic groupings are all the same distance from the staff on a given system or line of an individual instrument, that the actual distance has less importance than when we see different distances on the same system or line as you demonstrated.
My personal “rubrics”, if you will, are: 1) do one’s best to ensure that the end user (player or conductor) has clarity with the intent of the music; and 2) make things look consistent. So, in the case of your examples, this would mean that, as is, there is clarity to the end user but the placement of grouped dynamics is inconsistent.
The latter, in most cases, should likely be corrected by moving the measure(s) to the next line or system unless that simply moves the “issue” from one location to another. It would be much like adjusting the location of measures to keep from breaking a hairpin across systems or pages vs. using verbose text (i.e., crescendo or diminuendo) in place of the hairpins.
There are so many nuances to this scenario that I’m not sure that layout-specific Engraving Options would make any difference. Now, please push back and let’s keep thinking through this. I truly enjoy cordial mental gymnastics!
My rule of thumb is to keep each dynamic gesture in line, so on the top system I’d probably keep things as they are since to me the dim. is part of a different gesture. In 376 I’d probably align the >p with the p<mf, and then in the following bars keep everything between rests aligned–mp<f> is one group, p<f>mf is another group, the final mp is its own thing, etc.
I tend to do essentially what you’re suggesting, @ryanpetriello. My philosophy so far has been to first give the system the “10-foot-away” look as a performer and see if anything catches my eye as disjointed, distracting, or hidden. I then try and look for logical groupings, usually in clusters like is shown in the second system. I then try and group each cluster individually as best as possible such that the grouped cluster does not result in any excessive amount of space in any one spot that would cause the performer’s eye to have to read unnaturally low on the page.
I realize that as engravers we want to get pedantic and haul out the ruler in so many of these cases, but a) performers really don’t care; b) I don’t think it’s fun; and c) I can’t spend the rest of my life tweaking dynamics incessantly. So it’s probably not the purist way of thinking about dynamic alignment, but it seems so far to strike a healthy balance between time spent and clarity of outcome.
At the end of the day, my goal is to either make a performer go, “this looks pretty nice” or, better yet, have them say nothing at all because the music gets out of their way.