Indeed, there is a bug, but it isn't quite like what you think is happening. The collection object gets a single zorder value in order to determine its drawing order in relation to other artists, but it is up to the collection object to determine the draw order for its own elements. This is the best we can do given Matplotlib's layering engine for rendering.

The Poly3DCollection has some sophisticated logic to try and properly layer its elements on the fly, and that is why you see it work fine in the example. This was necessary because incorrectly layered polygons are very obvious.

Looking over the code for Line3DCollection, it does not have any of this sophistication. I am presuming that it is because incorrectly layered lines are not nearly as visually obvious compared to polygons. I didn't even realize that one can make something like a polygon collection out of a line collection! @mdboom , Is that even a supported feature of LineCollections, or is that an accident of design?

If that is a supported feature of LineCollections, then I will see if I can utilize the same code in Poly3DCollection in Line3DCollection, which should fix this problem.

Thanks for your rapid reply.
Supported or not, the LineCollection3D is a fantastic visualization tool, that I've used in a few publications already, and have passed on to researchers in other fields for visualizing their own complex data.
Your work on this is greatly appreciated!

Are there any tests you'd like me to try to help you out?

Confirming this still exists in 1.5.3.

@WeatherGod did you end up getting any guidance from @mdboom?

Has there been any progress on this?
FYI I should point out that LineCollection allows you to plot arbitrary Data (that's what we use it for), while PolygonColection isn't really intended for that.

We use it in our research papers, for example see Fig 2 in this paper:
http://praevium.com/wptest_v1/wp-content/uploads/2015/08/2015-JLT-JohnJayaraman-Wideband-Electrically-Pumped-1050-nm-MEMS-Tunable-VCSEL-for-Ophthalmic-Imaging.pdf

This issue has been marked "inactive" because it has been 365 days since the last comment. If this issue is still present in recent Matplotlib releases, or the feature request is still wanted, please leave a comment and this label will be removed. If there are no updates in another 30 days, this issue will be automatically closed, but you are free to re-open or create a new issue if needed. We value issue reports, and this procedure is meant to help us resurface and prioritize issues that have not been addressed yet, not make them disappear. Thanks for your help!

Checking this - bug is still present, in Anaconda / Python 3.11, Spyder 5.4.3, matplotlib 3.7.1

(Unrelated) - the demo script now produces an error near the start;
demo script in docs page:

fig = plt.figure()
ax = fig.gca(projection='3d')

Result (python 3.11):
TypeError: FigureBase.gca() got an unexpected keyword argument 'projection'

Had to replace line with ax = fig.add_subplot(projection='3d') as per this hint.

Here is the modified demo script to reproduce this problem:

"""
=============================================
Generate polygons to fill under 3D line graph
=============================================

Demonstrate how to create polygons which fill the space under a line
graph. In this example polygons are semi-transparent, creating a sort
of 'jagged stained glass' effect.
"""

from mpl_toolkits.mplot3d import Axes3D
from matplotlib.collections import PolyCollection, LineCollection
import matplotlib.pyplot as plt
from matplotlib import colors as mcolors
import numpy as np


fig = plt.figure()
ax = fig.add_subplot(projection='3d')

#ax = fig.gca(projection='3d')
#fig, ax = plt.subplots(projection='3d')

def cc(arg):
    return mcolors.to_rgba(arg, alpha=1.0)

xs = np.arange(0, 10, 0.4)
verts = []
zs = [0.0, 1.0, 2.0, 3.0]
for z in zs:
    ys = np.random.rand(len(xs))
    ys[0], ys[-1] = 0, 0
    verts.append(list(zip(xs, ys)))

#poly = PolyCollection(verts, facecolors=[cc('r'), cc('g'), cc('b'), cc('y')])
poly = LineCollection(verts, facecolors=[cc('r'), cc('g'), cc('b'), cc('y')])
    
poly.set_alpha(1.0)
ax.add_collection3d(poly, zs=zs, zdir='y')

ax.set_xlabel('X')
ax.set_xlim3d(0, 10)
ax.set_ylabel('Y')
ax.set_ylim3d(-1, 4)
ax.set_zlabel('Z')
ax.set_zlim3d(0, 1)

plt.show()