What would be the best visualization for this data set?

Question

I have a data frame that look like this:

    Team    T1  s_T1    T2  s_T2    varK_T1     varK_T2     Accuracy_T1    Accuracy_T2 
    team1    5       1       6       0.5       68          60            82            89
      .      .       .       .        .         .           .             .             .
      .      .       .       .        .         .           .             .             .
      .      .       .       .        .         .           .             .             .
    teamN    7       2       8        1        70          54            70            81

The Team col is also my index. This exemplary table is the result of simulations by me. T1 and T2 are times and s_T1, s_T2 are the errors on the times in minutes. Than we have variation coefficiants and accuracy for the times.

I want to visualize this data but I don't really know what would be the best representation? People should see on the visualization that for the teams the times T2 are better than T1. Especially the problem is the length of the table. What you see above until the row TeamN is only a little example. The real table has more than 100 Thousand rows.

What kind of plot what you use for best understandind?

Thanks and greetings

Answer 1

If you actually want to plot the information of more than 100.000 teams I would only plot the T1 and T2 time for each team on a density plot. So put T1 and T2 on the x and y axis, separate both into bins (like 1 min intervals) and the plot on each square of the grid the number of teams in that square coded via a color or grey scale variation. If on average the T2 times are shorter than the T1 times, you should see more density in the bottom right of the plot and less in the top left.

If you only plot a handful of teams you could make use of a lot more of the columns in your data and completely different plot types would be appropriate.