[Python] Error bars only getting caps in xerr not yerr despite there not being any xerr

Python is adding caps to non-existent error bars while completely ignoring the existing error bars.

This is my code:

data = pd.read_excel("Excel_data.xlsx", sheet_name = "avg errors")

angles = np.array(data.columns)
data = np.array(data)
averages = data[0,:]
errors = data[1,:]

fig, ax = plt.subplots(subplot_kw={'projection': 'polar'})

angles = [radians(a) for a in angles]

ax.plot(angles, averages,markers = 'o', markersize = 3)
ax.errorbar(angles, averages, fmt = 'none', yerr=errors)
ax.set_thetamin(-91)
ax.set_thetamax(91)
ax.set_theta_offset(3*np.pi/2)
ax.set_yticks([0,20,40,60,80,100])
ax.set_yticklabels([0,20,40,60,80,100])
plt.show()



Which gives me a great graph:



However, when I try to add caps to the error bars by changing the line specifying the errorbars:

ax.errorbar(angles, averages, fmt = 'none', yerr=errors, capsize = 10)


It doesn't add caps to the existing error bars. Instead, it adds error bars to the non-existent error bars in the x direction (as is visible by the slight thickening of the y-error bars near the points):

Here are arrays with approximate values for each of the datasets:

angles: [-90 -75 -60 -45 -30 30 45 60 75 90]
averages: [73 63 57 54 39 39 56 59 62 73]
errors: [ 7.9 6.4 5.2 4.3 5.6 4.0 3.7 5.5 4.6 12.0]

Continue reading...