If possible can you simplify this to Matplotlib with no dates, no pandas, and maybe the latest release? I think fill-between may have had some work recently. Thanks!

For instance if I do:

import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure(1)
ax = fig.add_subplot(111)
x = np.arange(30)
np.random.seed(1)
y = np.cumsum(np.random.randn(30))
y = y - np.mean(y)
ax.plot(x, y, drawstyle='steps-post')
yy = y
yy[y<0] = 0
ax.fill_between(x, yy, step='post')
plt.show()

I get what I think you are trying to do, so I'm guessing this is actually to do with how you are using pandas (but I could certainly be wrong):

Yes this works with the extra code you've provided: yy = y; yy[y<0] = 0.

Thanks for the push to simplify the bug. It is an NaN bug, and I think this should work with NaNs, although replacing NaNs with 0 is not a big burden.

import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure(1)
ax = fig.add_subplot(111)

x = np.arange(1,11)
y = np.array([1,2,3,-4,-5,-6,7,8,9,-10])
yn = np.where(y > 0, y, np.nan)

ax.plot(x, y, drawstyle='steps-post')
ax.fill_between(x, yn, step='post')
plt.show()

The above code generates the first graphic: https://user-images.githubusercontent.com/145117/106748570-46cc4480-65da-11eb-8e38-83e2d4e02fc2.png

See related: #15065 and #16635

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!