Irelands alcohol consumption over time

america.py

import pandas as pd

url = "https://sdmx.oecd.org/public/rest/data/OECD.ELS.HD,DSD_HEALTH_LVNG@DF_HEALTH_LVNG_AC,1.0/.A.....?dimensionAtObservation=AllDimensions&format=csvfilewithlabels&startPeriod=1960"

df = pd.read_csv(
    url,
    storage_options={
        "User-Agent": "Mozilla/5.0"
    }
)

# Keep only USA rows (country code column is usually REF_AREA)
usa = df[df["REF_AREA"] == "USA"].copy()

usa.head()


import matplotlib.pyplot as plt

# --- Define columns for OECD dataframe ---
time_col = "TIME_PERIOD"
value_col = "OBS_VALUE"

# --- Make sure year/value are numeric and sorted ---
usa = usa.copy()
usa[time_col] = pd.to_numeric(usa[time_col], errors="coerce")
usa[value_col] = pd.to_numeric(usa[value_col], errors="coerce")
usa = usa.dropna(subset=[time_col, value_col]).sort_values(time_col)

# Optional: cast year to int for cleaner matching
usa[time_col] = usa[time_col].astype(int)

# --- Color Palette (Modern "Journalist" Look) ---
LINE_COLOR = "#1B4F72"   # Deep Navy
HIGHLIGHT_COLOR = "#CB4335" # Muted Red for events
GRID_COLOR = "#EAECEE"

match_year = 1987
latest_year = int(usa[time_col].iloc[-1])
y_match = float(usa.loc[usa[time_col] == match_year, value_col].iloc[0])
y_latest = float(usa.loc[usa[time_col] == latest_year, value_col].iloc[0])

fig, ax = plt.subplots(figsize=(11, 6), facecolor='white')

# 1. Comparison Line (1987 level)
#ax.hlines(y=y_match, xmin=1987, xmax=latest_year,
#          colors='gray', linestyles='--', alpha=0.3, linewidth=1, zorder=1)

# 2. Main Line
ax.plot(usa[time_col], usa[value_col], color=LINE_COLOR,
        linewidth=3.5, solid_capstyle="round", zorder=3)

# 3. Event Markers (placeholders for now)
events = [
    (1984, "Drinking Age 21 law", "bottom", -2.3), # Added 'bottom' for va
    (1991, "Federal alcohol\n tax increase","bottom", -0.9), # Added 'bottom' for va
    (2008, "Great Recession", "top", -0.9),
    (2020, "COVID-19", "bottom", -1.5)
    #(2022, "Minimum Pricing", "bottom", -1.0) # placeholder
]

for year, label, va, offset in events:
    # Skip if year not present
    if year not in set(usa[time_col].values):
        continue
    y_val = float(usa.loc[usa[time_col] == year, value_col].iloc[0])
    ax.axvline(year, color=HIGHLIGHT_COLOR, linestyle="--", linewidth=1, alpha=0.4, zorder=2)
    ax.text(year, y_val + offset, label, color=HIGHLIGHT_COLOR,
            fontweight='bold', fontsize=9, ha='center', va=va)

# 4. Key Data Points
peak_i = usa[value_col].idxmax()
peak_year = int(usa.loc[peak_i, time_col])
peak_val  = float(usa.loc[peak_i, value_col])

ax.scatter([peak_year, latest_year], [peak_val, y_latest],
           color=LINE_COLOR, s=70, edgecolors='white', zorder=5)

# 5. Clean Annotations
#ax.annotate(f"Peak: {peak_val:.1f}L",
#            xy=(peak_year, peak_val), xytext=(peak_year-2, peak_val+0.6),
#            fontweight='bold', fontsize=11)

ax.annotate(f"Peak: {peak_year} ({peak_val:.1f}L)",
            xy=(peak_year, peak_val),
            xytext=(peak_year-15, peak_val-0.15),
            arrowprops=dict(arrowstyle="->", color="gray", lw=1),
            fontweight='bold', fontsize=11, ha='left', va='bottom')
#ax.annotate(f"Back to {match_year} levels ({y_latest:.1f}L)",
#            xy=(latest_year, y_latest), xytext=(latest_year-9, y_latest-1.2),
#            arrowprops=dict(arrowstyle='->', connectionstyle='arc3,rad=.2', color='gray'),
#            fontsize=10)

# Current (latest) level
latest_year = int(usa[time_col].iloc[-1])      # 2022
latest_val  = float(usa[value_col].iloc[-1])   # 9.5

# Exact historical match(es) you found
y_1970 = float(usa.loc[usa[time_col] == 1970, value_col].iloc[0])

# Near match in 1990s
#y_1988 = float(usa.loc[usa[time_col] == 1988, value_col].iloc[0])

# Horizontal comparison line at current level
ax.hlines(y=latest_val, xmin=1970, xmax=latest_year,
          colors="gray", linestyles="--", alpha=0.35, linewidth=1.2, zorder=1)

# Highlight the comparison points y_1988,
ax.scatter([1970, latest_year], [y_1970,  latest_val],
           s=55, color=LINE_COLOR, edgecolors="white", zorder=6)

# Label / callout
ax.annotate("2022 matches 1970 (9.5L)",
            xy=(latest_year, latest_val),
            xytext=(latest_year-16, latest_val-1.1),
            arrowprops=dict(arrowstyle="->", connectionstyle="arc3,rad=.2", color="gray"),
            fontsize=10, ha="left")

# 6. Title and Labels (placeholder text)
plt.title("Americans’ Average Alcohol Consumption",
          fontsize=20, fontweight='bold', pad=25, loc='left', color="#2C3E50")

ax.set_ylabel("Litres per capita per year (age 15+)", fontsize=11, fontweight='bold', alpha=0.7)
ax.grid(True, axis='y', alpha=0.3, color=GRID_COLOR)
ax.spines["top"].set_visible(False)
ax.spines["right"].set_visible(False)

# Footer
plt.figtext(0.9, 0.001, "Data: OECD Graph: @iamreddave", ha="right", fontsize=9, alpha=0.5)

plt.tight_layout()
plt.savefig("usa_alcohol_oecd.png", dpi=300, bbox_inches="tight")
plt.show()

irelandalcohol.py

import pandas as pd

url = "https://sdmx.oecd.org/public/rest/data/OECD.ELS.HD,DSD_HEALTH_LVNG@DF_HEALTH_LVNG_AC,1.0/.A.....?dimensionAtObservation=AllDimensions&format=csvfilewithlabels&startPeriod=1960"

df = pd.read_csv(
    url,
    storage_options={
        "User-Agent": "Mozilla/5.0"
    }
)

# --- Color Palette (Modern "Journalist" Look) ---
LINE_COLOR = "#1B4F72"   # Deep Navy
HIGHLIGHT_COLOR = "#CB4335" # Muted Red for events
GRID_COLOR = "#EAECEE"

match_year = 1987
latest_year = int(ireland[time_col].iloc[-1])
y_match = float(ireland.loc[ireland[time_col] == match_year, value_col].iloc[0])
y_latest = float(ireland.loc[ireland[time_col] == latest_year, value_col].iloc[0])

fig, ax = plt.subplots(figsize=(11, 6), facecolor='white')

# 1. Comparison Line (1987 level)
ax.hlines(y=y_match, xmin=1987, xmax=latest_year, 
          colors='gray', linestyles='--', alpha=0.3, linewidth=1, zorder=1)

# 2. Main Line
ax.plot(ireland[time_col], ireland[value_col], color=LINE_COLOR, 
        linewidth=3.5, solid_capstyle="round", zorder=3)

# 3. Event Markers (Moved Smoking Ban down)
events = [
    (2004, "Smoking Ban", "bottom", -8.8), # Negative offset to move it DOWN
    (2008, "Economic Crash", "top", -6.3),
    (2020, "COVID-19", "bottom", -5.)
]

for year, label, va, offset in events:
    y_val = float(ireland.loc[ireland[time_col] == year, value_col].iloc[0])
    ax.axvline(year, color=HIGHLIGHT_COLOR, linestyle="--", linewidth=1, alpha=0.4, zorder=2)
    ax.text(year, y_val + offset, label, color=HIGHLIGHT_COLOR, 
            fontweight='bold', fontsize=9, ha='center', va=va)

# 4. Key Data Points
peak_i = ireland[value_col].idxmax()
peak_year = int(ireland.loc[peak_i, time_col])
peak_val  = float(ireland.loc[peak_i, value_col])

ax.scatter([peak_year, latest_year], [peak_val, y_latest], 
           color=LINE_COLOR, s=70, edgecolors='white', zorder=5)

# 5. Clean Annotations
ax.annotate(f"Peak: {peak_val:.1f}L", xy=(peak_year, peak_val), xytext=(peak_year-2, peak_val+0.6),
            fontweight='bold', fontsize=11)

ax.annotate(f"Back to {match_year} levels ({y_latest:.1f}L)", 
            xy=(latest_year, y_latest), xytext=(latest_year-9, y_latest-1.2),
            arrowprops=dict(arrowstyle="->", connectionstyle="arc3,rad=.2", color="gray"),
            fontsize=10)

# 6. Title and Labels (Centered title option)
plt.title("Ireland's Alcohol Consumption: A 25-Year Decline", 
          fontsize=20, fontweight='bold', pad=25, loc='left', color="#2C3E50")

ax.set_ylabel("Litres per capita (age 15+)", fontsize=11, fontweight='bold', alpha=0.7)
ax.grid(True, axis='y', alpha=0.3, color=GRID_COLOR)
ax.spines["top"].set_visible(False)
ax.spines["right"].set_visible(False)

# Footer
plt.figtext(0.9, 0.001, "Data: OECD • Visual: @iamreddave", ha="right", fontsize=9, alpha=0.5)

plt.tight_layout()
plt.show()