fmt

analysis/build_pages.py

@@ -18,7 +18,6 @@ from __future__ import annotations
 
 import html
 import json
-import urllib.parse
 from pathlib import Path
 
 ROOT = Path(".")

analysis/build_video_scripts.py

@@ -74,7 +74,7 @@ def twin_kit(f: dict) -> str:
     ]
     captions = [
         f"{pn} vs {wn}",
-        f"Same station. Same schools.",
+        "Same station. Same schools.",
         f"{money} cheaper",
         f"Same {typ}, ~{s['build_year']}",
         f"{gap} less per m²",
@@ -135,7 +135,7 @@ def twin_kit(f: dict) -> str:
         "",
         f"**Page:** {SITE}{f['page_path']}  ·  **Format:** faceless screen-record, ~45–60s long + a 9:16 Short cut",
         "",
-        f"## 🎬 Map URL to record (open this, hit record)",
+        "## 🎬 Map URL to record (open this, hit record)",
         f"`{url}`",
         "*(filters are pre-applied so the value is on screen immediately)*",
         "",

analysis/generate_findings.py

@@ -139,9 +139,6 @@ def curate(tw: pl.DataFrame) -> list[dict]:
 
 
 def national_findings(idx: pl.DataFrame, facts: dict) -> list[dict]:
-    named = idx.with_columns(
-        pl.col("Sector").str.extract(r"^([A-Z]+[0-9][A-Z0-9]?)", 1).alias("outward")
-    )
     best = facts["best_value_sector"]
     dear = facts["dearest_sector"]
     return [

analysis/newbuild_vs_existing_london.ipynb

@@ -315,7 +315,8 @@
     "ax1.plot(yr, piv[\"existing\"].to_numpy(), marker=\"o\", label=\"Existing houses\", color=\"#0f766e\")\n",
     "ax1.plot(yr, piv[\"new\"].to_numpy(), marker=\"s\", label=\"New-build houses\", color=\"#d97706\")\n",
     "ax1.set_title(\"Median £/sqm by sale year\")\n",
-    "ax1.set_xlabel(\"Year\"); ax1.set_ylabel(\"£/sqm\")\n",
+    "ax1.set_xlabel(\"Year\")\n",
+    "ax1.set_ylabel(\"£/sqm\")\n",
     "ax1.yaxis.set_major_formatter(mticker.StrMethodFormatter(\"£{x:,.0f}\"))\n",
     "ax1.legend()\n",
     "\n",
@@ -323,9 +324,11 @@
     "ax2.bar(yr, piv[\"premium_pct\"].to_numpy(), color=colors)\n",
     "ax2.axhline(0, color=\"black\", lw=0.8)\n",
     "ax2.set_title(\"New-build premium (median £/sqm, new vs existing)\")\n",
-    "ax2.set_xlabel(\"Year\"); ax2.set_ylabel(\"Premium %\")\n",
+    "ax2.set_xlabel(\"Year\")\n",
+    "ax2.set_ylabel(\"Premium %\")\n",
     "ax2.yaxis.set_major_formatter(mticker.PercentFormatter(decimals=0))\n",
-    "plt.tight_layout(); plt.show()\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n",
     "\n",
     "piv.select(\"year\", \"existing\", \"new\", \"premium_pct\", \"n_new\", \"n_existing\")"
    ]
@@ -426,7 +429,8 @@
     "for y, v, nnew in zip(range(len(labels)), vals, area_prem[\"n_new\"].to_list()):\n",
     "    ax.text(v + (0.4 if v >= 0 else -0.4), y, f\"n={nnew}\", va=\"center\",\n",
     "            ha=\"left\" if v >= 0 else \"right\", fontsize=8, color=\"#555\")\n",
-    "plt.tight_layout(); plt.show()\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n",
     "area_prem"
    ]
   },
@@ -523,10 +527,12 @@
     "ax.axvline(np.median(old_cagr), color=\"#0f766e\", ls=\"--\", lw=1.2)\n",
     "ax.axvline(np.median(new_cagr), color=\"#d97706\", ls=\"--\", lw=1.2)\n",
     "ax.set_title(\"Distribution of annualised resale growth (CAGR)\")\n",
-    "ax.set_xlabel(\"CAGR % per year\"); ax.set_ylabel(\"density\")\n",
+    "ax.set_xlabel(\"CAGR % per year\")\n",
+    "ax.set_ylabel(\"density\")\n",
     "ax.xaxis.set_major_formatter(mticker.PercentFormatter(decimals=0))\n",
     "ax.legend()\n",
-    "plt.tight_layout(); plt.show()"
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -603,16 +609,19 @@
     "\n",
     "area_cagr = cagr_by(\"area\")\n",
     "labels = area_cagr[\"area\"].to_list()\n",
-    "x = np.arange(len(labels)); w = 0.4\n",
+    "x = np.arange(len(labels))\n",
+    "w = 0.4\n",
     "fig, ax = plt.subplots(figsize=(10, 4.5))\n",
     "ax.bar(x - w/2, area_cagr[\"existing_cagr\"].to_numpy(), w, label=\"Existing\", color=\"#0f766e\")\n",
     "ax.bar(x + w/2, area_cagr[\"newbuild_cagr\"].to_numpy(), w, label=\"New-build\", color=\"#d97706\")\n",
-    "ax.set_xticks(x); ax.set_xticklabels(labels)\n",
+    "ax.set_xticks(x)\n",
+    "ax.set_xticklabels(labels)\n",
     "ax.set_title(\"Median resale CAGR by postal area\")\n",
     "ax.set_ylabel(\"CAGR % per year\")\n",
     "ax.yaxis.set_major_formatter(mticker.PercentFormatter(decimals=0))\n",
     "ax.legend()\n",
-    "plt.tight_layout(); plt.show()\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n",
     "area_cagr"
    ]
   },
@@ -699,22 +708,25 @@
     "    area_cagr.select(\"area\", \"cagr_gap\", \"newbuild_cagr\", \"existing_cagr\"), on=\"area\", how=\"inner\"\n",
     ")\n",
     "fig, ax = plt.subplots(figsize=(8, 6))\n",
-    "xs = combined[\"premium_pct\"].to_numpy(); ys = combined[\"cagr_gap\"].to_numpy()\n",
+    "xs = combined[\"premium_pct\"].to_numpy()\n",
+    "ys = combined[\"cagr_gap\"].to_numpy()\n",
     "ax.scatter(xs, ys, s=70, color=\"#d97706\", zorder=3)\n",
     "for a, xv, yv in zip(combined[\"area\"].to_list(), xs, ys):\n",
     "    ax.annotate(a, (xv, yv), textcoords=\"offset points\", xytext=(6, 4), fontsize=10)\n",
-    "ax.axhline(0, color=\"black\", lw=0.8); ax.axvline(0, color=\"black\", lw=0.8)\n",
+    "ax.axhline(0, color=\"black\", lw=0.8)\n",
+    "ax.axvline(0, color=\"black\", lw=0.8)\n",
     "ax.set_xlabel(\"New-build £/sqm premium at purchase\")\n",
     "ax.set_ylabel(\"New-build CAGR minus existing CAGR (pp/yr)\")\n",
     "ax.xaxis.set_major_formatter(mticker.PercentFormatter(decimals=0))\n",
     "ax.set_title(\"Premium paid vs appreciation gap, by London postal area\")\n",
-    "plt.tight_layout(); plt.show()\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n",
     "\n",
     "# Headline numbers for the writeup.\n",
     "hl = summary.to_dicts()\n",
     "new_row = next(r for r in hl if r[\"is_newbuild_prop\"])\n",
     "old_row = next(r for r in hl if not r[\"is_newbuild_prop\"])\n",
-    "print(f\"London terraced/detached, repeat sales:\")\n",
+    "print(\"London terraced/detached, repeat sales:\")\n",
     "print(f\"  existing  median CAGR: {old_row['median_cagr_pct']}%/yr (n={old_row['n']:,})\")\n",
     "print(f\"  new-build median CAGR: {new_row['median_cagr_pct']}%/yr (n={new_row['n']:,})\")\n",
     "print(f\"  appreciation gap: {round(new_row['median_cagr_pct']-old_row['median_cagr_pct'],2)} pp/yr\")\n",