It seems to come from this:

See #2941 for a bit of background why that happens.

Note: It's almost never a good idea to iterate over and modify individual ticks. Ticks are volatile and can change throughout the lifecycle of a figure. The correct approach to aboves code is to use tick_params:

fig, ax = plt.subplots()
ax.plot([1, 2, 3])
ax.tick_params(axis='y', labelcolor='r')
ax.spines.left.set_position(("outward", 10))
plt.show()

More details are discussed in #22856.

Thank @timhoffm . But tick_params can not change only part of the ticks, correct?

@y9c If you mean "some of the ticks", yes, tick_params changes the global defaults for ticks. But that's the point. Ticks can change and if you set them individually, you have always the risk that your individual settings get lost. If you need to operate on individual ticks, I advise to do this as late as possible in your figure creation.

Hi @timhoffm , I tried to set tick label color after all plotting function, but still don't work. It seem that once set_position function is called, the color will always be reset.

fig, ax = plt.subplots()
ax.plot([1, 2, 3])
ax.spines.left.set_position(("outward", 10))
for lab in ax.get_yticklabels():
    lab.set_color("r")
plt.show()

works for me:

In https://matplotlib.org/stable/tutorials/intermediate/artists.html it is shown that one can iterate over the ticks and tick labels. This should probably be updated.

Also, the approach there will not affect the tick color as expected (not at all...).