I recently came across the numbers of convictions and executions in the latter half of 18^th century England. The number of executions varied and it wasn’t immediately clear if there were any larger trends.

	Convictions	Executions
1749-58	527	365
1759-68	372	206
1769-78	787	357
1779-88	1152	531
1789-98	770	191
1799-1808	804	126

But I found that once you look at the number of executions relative to convictions, the growing distaste in executions in Georgian England becomes far more apparent.

	Convictions	Executions	Percentage
1749-58	527	365	69.3
1759-68	372	206	55.4
1769-78	787	357	45.4
1779-88	1152	531	46.1
1789-98	770	191	24.8
1799-1808	804	126	15.7

I graphed change in the percentage against time but also included the total number of executions by varying the line thickness. I think it’s interesting because it shows the decline in the percentage of executions while reminding us that there were still a tremendous number of executions. And while the overall trend is clear and good, there are major setbacks along the way.

x = np.array(range(6))
y = df['Percentage']
widths = df['Executions'].values / 100

import matplotlib.patches as patch

fig, ax = plt.subplots(1, 1, figsize=(10, 10))
ax.set_xlim(-.5,5.5)
ax.set_ylim(10,80)

new_w = np.array(widths)

for i in range(len(x)-1):
    c = [[x[i], y[i]+new_w[i]/2.],
         [x[i+1], y[i+1]+new_w[i+1]/2.],
         [x[i+1], y[i+1]-new_w[i+1]/2.],
         [x[i], y[i]-new_w[i]/2.]
        ]
    p = patch.Polygon(c)
    ax.add_patch(p)
plt.xlabel("Decade", fontsize=16)
plt.ylabel("Execution Percentage", fontsize=16)
plt.title("Declining appetite for executions in 18th century England", fontsize=18)
plt.xticks(x, decade)
plt.show()