DraftSense
Champ-select companion for League of Legends

Pick to your strengths, not a tier list.

DraftSense ranks every option for your role by matchup, team fit, and your champion mastery, refreshed each patch.

How it works Privacy policy
Example output, you are BOTTOM (blue side)
Kai'Sa best pick
synergy with your Lulu, scales with the comp
52.6%
Kalista
strong into the enemy bot lane
51.9%
Xayah
safe blind, your 240k mastery
51.1%
Several top picks are often statistical ties. DraftSense says so instead of faking a ranking.

What it does

During champ select, DraftSense ranks the candidates for your role by how well they fit this specific draft and how well you play them.

  • Plays to your masteryRanks champions you actually perform on, from each player's champion-mastery totals, not just global averages.
  • Champion-specific matchupsReal counter and synergy edges measured from a ranked-game corpus, not a hand-maintained tier list.
  • Fresh every patchAn automated pipeline retrains on current-patch ranked games, so the advice tracks the live meta.
  • Honest win chanceA calibrated win percentage backs each pick, and flags when the top options are statistical ties.
Overlay in champ select
DraftSense overlay during champ select: ally and enemy rows with champion portraits, a damage-profile comparison, a counterpick note, the recommended pick Kai'Sa with its calibrated win chance and a matchup, synergy, base, and comp-fit value breakdown, plus ranked alternates.

How it works

A thin, read-only overlay on your machine talks to a hosted backend that does the modelling. One production key serves every user; no key ships to your machine.

  1. Read. The overlay reads champ-select state from the League client's local API (LCU), read-only. No automation, no input simulation.
  2. Look up. The backend resolves the lobby's players and fetches their ranked tier and champion mastery server-side.
  3. Score. The model evaluates each candidate for your open slot, conditioned on the board, ranks, and mastery.
  4. Show. Top picks with a win-probability delta and a short reason. You make every decision.

How it is different

Most tools show population averages. DraftSense reads your specific draft and lobby.

Typical stats sites

  • Population win rates and counter tables
  • The same advice for everyone
  • A black-box "strong pick"
  • Tables updated by hand

DraftSense

  • P(win) for this partial draft, from a sequence model
  • Personalized to your mastery and the lobby's ranks
  • Counterfactual drivers: mastery, rank, matchup, bans
  • Patch-trained model with an automated refresh pipeline

How DraftSense uses the Riot API

Transparent by design. Official Web APIs only, rate-limited, no scraping, no write endpoints.

APIWhyWhen
Account-V1Resolve lobby players (Riot ID and puuid) for rank and mastery lookupslive
League-V4Per-player ranked tier and LP, a model inputlive
Champion-Mastery-V4Per-player champion mastery, personalizationlive
Summoner-V4puuid and summoner id where an endpoint requires itlive
Match-V5Ranked match outcomes to train the modeloffline
LCU (local client)Read champ-select state on your machine, read-onlylocal

No spectator, tournament, or write endpoints. We respect every returned rate-limit header and self-throttle below it.

Responsible and ToS-safe by design

Read-only client

The LCU is used only to read the draft. No auto-pick, auto-ban, auto-accept, auto-dodge, or simulated input. You decide everything.

No unfair advantage

Output is informational and available before you lock, the analogue of reading a stats site. It does not act in-game.

Honest about limits

Draft is one noisy factor in a win. We show calibrated estimates and flag statistical ties. We do not promise wins or LP.

Minimal data

Short-lived puuid-keyed caches for a single draft. No per-player profiles, no selling Riot data. See the privacy policy.

Contact

Questions, data requests, or Riot review: contact@draftsense.app