That's what I'm discussing too.
Again, if you can show me a piece of cardboard or relative-to-size chunk of wood washing down through a bowl of marbles, I'll believe you.
Or get yourself a bag of rice or beans and try to wash down some chopped onions to the bottom of a colander, without shaking the colander. The onions are more likely to move upwards as opposed to downwards. Unless you chop them so finely, that they'll pass through the colander as well; not an issue at all.
Also, if water hits a flat surface, like a horizontal piece of wood.. What would happen to the water? It'll flow off the sides, creating a gap facing downwards - if there are particles around it that are able to move - following the stream of the water. If the bark moves in that direction, it'll end up vertically. All 'follow the path of least resistance' comes into play and the bark will stay horizontal. Where it's suddenly packed between two layers of inorganic particles on both sides that keep it in place. It can't tumble over if there's no empty air pocket to fall in to. We try to prevent air pockets in bonsai as much as possible. The fact that the bark could've moved in the downwards facing gap to begin with, means there was an air pocket. That's two conditions that have to be met for a piece of bark to move to begin with.
We're not growing in bark alone. There's plenty of rocks in there too, they tend to keep things into place.
Two dimensional structures surrounded by two dimensional structures have the tendency to stack. If they're surrounded by three dimensional structures, stacking is
very unlikely. Especially if the three dimensional particles are heavier than the two dimensional ones.
If you're comparing a water filled pond, littered with bark to a bonsai pot, you're forgetting the 70-80% pebbles that should surround the bark and you're replacing those pebbles with 100% water. That's not discussing particle shapes, it's discussing water flow and its effects on free floating two dimensional structures. That alone could be the title of a research paper, but it's a different subject. Bonsai soil is not water. The movement dynamic is entirely different.
I bet your filter is sucking in water straight from the pond, and it's not passing through a pebble bed (a three dimensional maze/matrix). Otherwise the pebbles would've stopped that bark from clogging the filter. It might be an idea to install such a contraption to prevent this from happening again in the future. I know those filter systems don't come cheap.
As a matter of fact, a well designed pond should have the pump behind such a feature simply to prevent these issues from happening. There will be other issues, like stacking on top of the contraption and eventually organic waste (if left long enough) will screw up the water flow. But it seems easier to scoop up a few pounds of pebbles with organic waste and hose those down, than to take apart a water filter and having to scoop the pond every time it rained. Waste water treatments have been doing it like this for over 30+ years. Here in Europe there is a growing community that do stuff like below for swimming pools, without chlorine. The 'aggregates' is what I mean with the pebble matrix or contraption/feature.
If you would have had a aggregate bed like below, the bark would still have stacked (a mix of two dimensional objects will likely stack), but it would've stacked
on top of the aggregate instead of inside your filter. You could fix that with a shovel.