How can i apply a patch on the chart using X axis values?

[Dave-12-cr]

I created a candle chart with fig, axlist = mpf.plot(df, type='candle', returnfig=True)

I'm trying to add on my chart a patch, here is what i tried:

new_patch = Rectangle(xy=(3, 9300), width=4, height=0.3, angle=0, color='orange')
ax1 = axlist[0]
ax1.add_patch(new_patch)
ax1.autoscale_view()

This code works, but the problem is that, instead of specifying the Y value with a number, i would like to specify it with dates, since my X axis has dates in the following format:

Date
2020-06-18 10:30:00   
2020-06-18 10:35:00   
2020-06-18 10:40:00   
2020-06-18 10:45:00

I tried the following: new_patch = Rectangle(xy='2020-06-18 10:30:00',9300), width=4, height=0.3, angle=0, color='orange')

But that will give me an error since X is not int. I just moved to the new library from the older one, i really like all the new features, i'm just having some troubles drawing on the chart like i did with the older one! Any kind of help is appreciated!

[DanielGoldfarb]

Hi @Dave-12-cr
There are a few things you can try. First, it is important to know that when show_nontrading=False (the default value) then the x-axis is simply the data point index number (from 0 up to N-1 where N is the number of rows in your dataframe).

1. You can try setting show_nontrading=True. This will cause the x-axis to be matplotlib dates. This may work, although you may have to convert '2020-06-18 10:30:00' into a matplotlib date first.

2. You can also try passing in the index number closest to the datetime you want. For example, if '2020-06-18 10:30:00' is on the 12th row of your dataframe, you would pass in 11.0 as the x-axis value (counting from zero to the 12th row). As another example, if '2020-06-18 10:30:00' represents a time that is halfway between the 16th row and the 17th row, then you would pass in 15.5 (thus interpolating the location).

3. If all you want is a rectangle, this can be done using the mplfinance alines kwarg to draw a short line with a large width as described here.

4. Similarly for a rectangle, or any other shape that can be used as a marker in matplotlib, you can use mpf_makeaddplot(data,type='scatter',marker=...) to place the shape (marker) anywhere on the plot. You can see some examples of doing this here and further discussion about this approach throughout Add a number or a symbol above/below a candle chart with mplfinance? #97

Personally I recommend number 3 or number 4 above as possibly being the simplest to implement; but give whichever one you want a try and let me know how it turns out for you.

[Dave-12-cr]

@DanielGoldfarb
Wow, so many ways to do this! Thank you!