I want to have two different charts, where in the first I can choose an artist in a legend and in the second chart I want to have a dropdown, that filters a year. In my database I have artists, sales and years (for published albums). The charts work fine by itself, but if I want to show both with a vertical concatenation I get a problem. The first chart works as expected but the second graph is completely empty on selecting a year. import altair as alt import pandas as pd # Sample Data top = pd.DataFrame({ 'Artist': ['Artist1', 'Artist2', 'Artist3', 'Artist1', 'Artist2', 'Artist3'], 'Sales': [100, 200, 300, 150, 250, 350], 'Year': ['2020', '2020', '2020', '2021', '2021', '2021'], }) # Selection for the legend select_artist_legend = alt.selection_point(fields=["Artist"], bind="legend") # First plot (cumulative sales over years) line = alt.Chart(top).transform_window( cumulative_sales='sum(Sales)', sort=[{'field': 'Year'}], groupby=['Artist'] ).mark_line().encode( alt.X("Year:T"), alt.Y("cumulative_sales:Q"), alt.Color("Artist:N"), opacity=alt.condition(select_artist_legend, alt.value(1.0), alt.value(0.1)) ).add_params(select_artist_legend) # Dropdown selection for the year year_dd = alt.binding_select(options=sorted(top.Year.unique()), name="Year") year_select = alt.selection_point(fields=["Year"], bind=year_dd, name='year_select') # Third plot (sales by artist for selected year) bar = alt.Chart(top).mark_bar().encode( alt.X("Artist:N"), alt.Y("Sales:Q"), alt.Color("Artist:N") ).add_params(year_select).transform_filter(year_select) # Combine the charts combined_chart = line & bar combined_chart Note: I'm using VS Code and Python Notebooks I tried debugging and trying to separate the two charts more, didn't seem to help. I also tried stacking bar twice instead of stacking line and bar, but there everything works, so stacking by itself didn't seem to be the problem. Continue reading...