From c7d0d929be053a6879a979c2ca7a1737ada68d47 Mon Sep 17 00:00:00 2001
From: Valentin Haudiquet <valentin.haudiquet@canonical.com>
Date: Thu, 30 Apr 2026 17:51:14 +0200
Subject: [PATCH] fix/match_collector: change region tagging logic

---
 match_collector/src/item_tree.ts | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/match_collector/src/item_tree.ts b/match_collector/src/item_tree.ts
index 9f3638e..db2c754 100644
--- a/match_collector/src/item_tree.ts
+++ b/match_collector/src/item_tree.ts
@@ -264,12 +264,8 @@ function deriveTags(node: ItemTree, expectedRegionDistribution?: PlatformCounts)
     const totalExpected = REGION_KEYS.reduce((sum, key) => sum + expectedRegionDistribution[key], 0)
 
     if (totalExpected > 0) {
-      // Tag if the item is significantly more popular in a region (>= 1.5x expected rate)
-      // and has a minimum absolute percentage (>= 10%)
-      const SIGNIFICANCE_THRESHOLD = 1.5
-      const MINIMUM_PCT = 0.1
-
-      // Loop through all regions to derive tags
+      // Tag if one region accounts for >= 60% of the normalized distribution
+      // Normalized value = actual percentage / expected percentage ratio
       const regionTags: Array<{ key: keyof PlatformCounts; tag: ItemTag }> = [
         { key: 'euw', tag: 'region_euw' },
         { key: 'eun', tag: 'region_eun' },
@@ -277,12 +273,23 @@ function deriveTags(node: ItemTree, expectedRegionDistribution?: PlatformCounts)
         { key: 'kr', tag: 'region_kr' }
       ]
 
-      for (const { key, tag } of regionTags) {
+      // Calculate normalized values (actual/expected ratio) for each region
+      const normalizedValues = regionTags.map(({ key, tag }) => {
         const expectedPct = expectedRegionDistribution[key] / totalExpected
         const actualPct = node.platformCount[key] / totalRegionCount
+        const normalizedValue = expectedPct > 0 ? actualPct / expectedPct : 0
+        return { tag, value: normalizedValue }
+      })
 
-        if (actualPct >= expectedPct * SIGNIFICANCE_THRESHOLD && actualPct >= MINIMUM_PCT) {
-          tags.push(tag)
+      const totalNormalized = normalizedValues.reduce((sum, { value }) => sum + value, 0)
+
+      // Tag the region if it accounts for >= 60% of the normalized distribution
+      if (totalNormalized > 0) {
+        for (const { tag, value } of normalizedValues) {
+          if (value / totalNormalized >= 0.6) {
+            tags.push(tag)
+            break // Only tag the most dominant region
+          }
         }
       }
     }