UCL FDR

Fixture Difficulty Rating tool for the UEFA Champions League league phase — React + Vite, mobile-first, Vercel-ready.

Live: ucl-fdr.vercel.app

Features

• FDR Table — all 36 teams as rows, MD1–MD8 as columns. Each cell shows the opponent, home/away and a difficulty band (colour and a printed digit). The sticky team column carries each team's average over the selected range.
• Best Fixture Runs — every team ranked by average difficulty, with real tiebreaks (home games, then count of very hard fixtures) and a "biggest swing" panel showing whose fixtures improve most inside the selected window.
• Compare Teams — up to five teams in one aligned grid, sorted easiest run first.
• Players — a form-and-fixtures guide per player over the selected matchdays: a 0–100 score (100 = best player for those matchdays) from points per match so far, minutes played and the fixtures in the range. Deliberately not shown as expected points: the model knows form and fixtures, not injuries, team news or expected goals. Filter by position, max price (a slider in €0.5m steps) and ownership; sort by score, score per €m or form. Appears in the menu once a player sheet is connected (see below); #/players?demo=1 previews it with made-up players.
• Team Strength — override any team's 1–5 rating. Searchable, changed teams are flagged with their original value, and everything recalculates live.

How difficulty is calculated

1. Every team carries a 1–5 strength rating, seeded from its UEFA coefficient and league-phase pot.
2. A fixture takes the difficulty of the opponent, not of the team playing it.
3. Home ties are eased by half a step, away ties hardened by half a step (VENUE_DELTA in src/utils/difficulty.js), which moves most fixtures one colour band. The Home/away switch turns this off.
4. Every rating is user-overridable and the overrides ride along in the URL.

Note that over the full MD1–MD8 range the venue adjustment nets out — every team plays exactly four home and four away — so it changes colours and per-fixture reading, not the season-long ranking. Over a shorter window, where the home/away split is uneven, it moves the averages.

URL state

The whole view lives in the hash, so any state is linkable:

#/table?from=1&to=8
#/runs?from=2&to=5
#/compare?from=1&to=8&teams=ARS,INT,BAY
#/table?r=ARS4BVB2          # custom ratings, packed 4 chars per team

Changing tab pushes a history entry (so mobile back walks through views); every other change replaces, so the back button doesn't fill up with rating taps.

Fixture data

src/data/fixtures.js holds the GW1–8 fixtures for the 2026/27 league phase, verified against the official draw: every fixture has a matching reciprocal with the opposite venue, all 36 teams play 8 matches (4 home / 4 away), no team faces a domestic rival, and nobody faces more than two clubs from one association.

Each entry has the shape:

{ gw: 1, opp: 'BAY', venue: 'H' } // matchday, opponent abbreviation, home (H) or away (A)

The source list lives in scripts/raw-fixtures.txt (one block per team, GW<n> <OPP> <H|A> per line). To regenerate src/data/fixtures.js after editing it:

node scripts/build-fixtures.mjs

The script cross-checks every fixture against its reciprocal and refuses to write the file if anything doesn't line up. Bump DATA_UPDATED in the same file when the data changes — the footer reads it.

Matchday dates live in src/data/matchdays.js.

Player sheet

The Players tab and GK Rotation both read this one published Google Sheet (GK Rotation uses its goalkeepers, with their minutes, points, price and ownership).

1. A sheet with this header row (any order, see scripts/players-template.csv): Player, Club, Position, Price, Points, Minutes
   • Club: the club's name as the site knows it (full name, short name or 3-letter code all work).
   • Position: GK, DEF, MID or FWD.
   • Points / Minutes: season totals so far; leave empty for someone who hasn't played.
   • Optional, used when present: Player ID (stable row key), Ownership % (the Differentials filter), Apps, MD1, MD2, … (points per matchday, shown in a player's details), and Goals, Assists, Clean sheets, Saves, Balls recovered, POTM. Any other column is ignored.
2. File → Share → Publish to web → pick that tab → Comma-separated values (.csv) → Publish, and copy the link.
3. Paste the link into PLAYERS_SHEET_URL in src/utils/useLivePlayers.js. The tab then shows up in the menu.

From then on edits to the sheet show up live; every deploy also saves a fallback copy to src/data/players.js (scripts/build-players.mjs, run by npm run build), which the browser only downloads when the Players tab opens. Rows the site can't read (unknown club, bad position or price) are skipped and listed in the browser console and the build log. Update the totals after every matchday: points per match divide by the matchdays played by today, so a sheet that lags behind under-rates everyone (evenly).

Assets

Club badges ship as 96px WebP in src/assets/badges/ (every file under 7 KB, crisp up to a 3× phone). The original vector art stays in src/assets/logos/ as the source of truth but is not imported — those files total ~490 KB, with a single crest reaching 31 KB gzipped for a badge drawn at 20 pixels.

The share card at public/og-image.jpg (1200×630) is what X and other platforms render for a shared link.

Development

npm install
npm run dev       # local dev server
npm run build     # production build (dist/)
npm run preview   # preview the build
npm run lint
npm test          # player sheet parsing + projection checks

Deploy

Standard Vite project — import the repo into Vercel and the build (npm run build, output in dist/) is auto-detected.

Not affiliated with UEFA. Club badges are the property of their respective clubs.