... I think there are just as many people who would complain about that behaviour as might appreciate it. I used Matlab for many years and don't recall it doing this, but perhaps I've forgoten.

@WeatherGod Yes, that is what I meant. I did try to set autoscale (https://matplotlib.org/api/_as_gen/matplotlib.pyplot.autoscale.html), but I could not get it to work due to lack of an official example.

@jklymak I think matplotlib can offer users the option to turn it on and off easily.

This was discussed and rejected in #9890. It would also require a pretty deep rewrite of the way data limits are computed.

@anntzer So autoscale and set_autoscale_view currently just do nothing? (I tried autoscale and the poster in #9890 tried set_autoscale_view.)

If they are not for this purpose, then I would suggest an update on the official documentation, because it is quite vague to a user as currently written:

"matplotlib.pyplot.autoscale(enable=True, axis='both', tight=None)

Autoscale the axis view to the data (toggle).

Convenience method for simple axis view autoscaling. It turns autoscaling on or off, and then, if autoscaling for either axis is on, it performs the autoscaling on the specified axis or axes."

Thanks for the explanation. I understand it would be a lot of work to implement this. So I guess a good compromise is to note this default in the documentation.

With this short script below, I can confirm two different behaviors:

1. By setting autoscale to "on", I allow the y axis bounds to automatically stretch to make room for out-of-bound data points.
2. The y axis bounds do not automatically shrink with data, even if I set autoscale to on.

However, by reading the current documentation (https://matplotlib.org/api/_as_gen/matplotlib.axes.Axes.autoscale.html), I could only expect Behavior No. 1, but not Behavior No. 2. So this is where I think the documentation can benefit from some improvements.

import numpy as np
import matplotlib.pyplot as plt

x = np.array([1,2,3,4,5,6,7])
y = x

fig = plt.figure()
ax = plt.axes()
ax.plot(x, y)
print('xlim: %s, ylim: %s' % (ax.get_xlim(), ax.get_ylim()))

ax.autoscale(enable=False)
y1 = 2 * x
ax.plot(x, y1)  # ylim is not allowed to stretch
print('xlim: %s, ylim: %s' % (ax.get_xlim(), ax.get_ylim()))

ax.autoscale(enable=True)
y2 = 2 * x + 1
ax.plot(x, y2)  # ylim stretches automatically
print('xlim: %s, ylim: %s' % (ax.get_xlim(), ax.get_ylim()))

ax.autoscale(enable=True)
y3 = 2 * x + 2
ax.set_xlim(1, 2)
ax.plot(x, y2)  # ylim does not shrink automatically
print('xlim: %s, ylim: %s' % (ax.get_xlim(), ax.get_ylim()))

This issue is still valid.

The work is to add another paragraph to the autoscale docstring so that the observed behavior in #13648 (comment) is not surprising

Closing as a duplicate of #2123.