brackt/database/schema.ts

907 lines
36 KiB
TypeScript
Raw Normal View History

fix: harden draft timer system with race-condition safety and DRY refactor (#34) - Fix settings action silently resetting timer values when draft speed select is disabled (add null guard before overwriting draftInitialTime/ draftIncrementTime) - Make timer decrement and pick increment atomic using SQL expressions to prevent read-modify-write races between the timer loop and HTTP handlers - Add UNIQUE INDEX on (season_id, pick_number) in draft_picks to prevent duplicate picks from concurrent requests (TOCTOU guard) - Add socket join-draft team ownership validation via DB query - Add iteration cap to autodraft chain while loop (max = totalTeams) - Add draftPaused re-check before firing autodraft chain in make-pick and force-manual-pick - Consolidate all snake draft calculations into calculatePickInfo (DRY); remove duplicated logic from timer.ts, make-pick, force-manual-pick, executeAutoPick, and checkAndTriggerNextAutodraft - Fix calculatePickInfo to return snake-adjusted pickInRound matching draftOrder values instead of raw pre-snake value - Fix timer-update socket events to emit nextPickNumber after a pick instead of the already-completed currentPickNumber - Fix force-manual-pick to validate submitted teamId matches the team whose turn it actually is at the given pick number - Replace inline timer init in draft.start with deleteSeasonTimers + initializeDraftTimers model functions (dead code fix) - Fix draft loader to not crash when getTeamQueue fails (returns []) - Fix misleading 403 message for commissioner picks - Remove dead hidden inputs from league settings form - Document connectedTeams single-instance limitation in socket.ts Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-23 23:23:24 -08:00
import { integer, pgTable, varchar, uuid, timestamp, pgEnum, boolean, text, date, decimal, uniqueIndex } from "drizzle-orm/pg-core";
import { relations } from "drizzle-orm";
// Users table - synced from Clerk
export const users = pgTable("users", {
id: uuid("id").primaryKey().defaultRandom(),
clerkId: varchar("clerk_id", { length: 255 }).notNull().unique(),
email: varchar("email", { length: 255 }).notNull(),
username: varchar("username", { length: 255 }),
displayName: varchar("display_name", { length: 255 }),
firstName: varchar("first_name", { length: 255 }),
lastName: varchar("last_name", { length: 255 }),
imageUrl: varchar("image_url", { length: 512 }),
isAdmin: boolean("is_admin").notNull().default(false),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Fantasy League Tables
export const seasonStatusEnum = pgEnum("season_status", [
"pre_draft",
"draft",
"active",
"completed",
]);
// Sports Tables - Enums
export const sportTypeEnum = pgEnum("sport_type", ["team", "individual"]);
export const sportsSeasonStatusEnum = pgEnum("sports_season_status", [
"upcoming",
"active",
"completed",
]);
export const scoringTypeEnum = pgEnum("scoring_type", [
"playoffs",
"regular_season",
"majors",
]);
export const scoringPatternEnum = pgEnum("scoring_pattern", [
"playoff_bracket",
"season_standings",
"qualifying_points",
]);
export const eventTypeEnum = pgEnum("event_type", [
"playoff_game",
"major_tournament",
"final_standings",
"schedule_event",
]);
export const pickedByTypeEnum = pgEnum("picked_by_type", [
"owner",
"commissioner",
"auto",
]);
export const autodraftModeEnum = pgEnum("autodraft_mode", [
"next_pick",
"while_on",
]);
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
export const probabilitySourceEnum = pgEnum("probability_source", [
"manual",
"futures_odds",
"elo_simulation",
"performance_model",
]);
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
export const simulationStatusEnum = pgEnum("simulation_status", [
"idle",
"running",
"failed",
]);
export const simulatorTypeEnum = pgEnum("simulator_type", [
"f1_standings",
"indycar_standings",
"golf_qualifying_points",
"playoff_bracket",
]);
export const leagues = pgTable("leagues", {
id: uuid("id").primaryKey().defaultRandom(),
name: varchar("name", { length: 255 }).notNull(),
createdBy: varchar("created_by", { length: 255 }).notNull(), // Clerk user ID
currentSeasonId: uuid("current_season_id"), // References the active season
isPublicDraftBoard: boolean("is_public_draft_board").notNull().default(false),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const seasonTemplates = pgTable("season_templates", {
id: uuid("id").primaryKey().defaultRandom(),
name: varchar("name", { length: 255 }).notNull(),
description: text("description"),
year: integer("year").notNull(),
isActive: boolean("is_active").notNull().default(true),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const seasons = pgTable("seasons", {
id: uuid("id").primaryKey().defaultRandom(),
leagueId: uuid("league_id")
.notNull()
.references(() => leagues.id, { onDelete: "cascade" }),
year: integer("year").notNull(),
status: seasonStatusEnum("status").notNull().default("pre_draft"),
templateId: uuid("template_id")
.references(() => seasonTemplates.id, { onDelete: "set null" }),
draftRounds: integer("draft_rounds").notNull().default(20),
flexSpots: integer("flex_spots").notNull().default(0),
draftDateTime: timestamp("draft_date_time"),
draftInitialTime: integer("draft_initial_time").notNull().default(120), // seconds
draftIncrementTime: integer("draft_increment_time").notNull().default(30), // seconds
currentPickNumber: integer("current_pick_number").default(1),
draftStartedAt: timestamp("draft_started_at"),
draftPaused: boolean("draft_paused").notNull().default(false),
inviteCode: varchar("invite_code", { length: 20 }).notNull().unique(),
// Scoring configuration (8 values for 1st through 8th)
pointsFor1st: integer("points_for_1st").notNull().default(100),
pointsFor2nd: integer("points_for_2nd").notNull().default(70),
pointsFor3rd: integer("points_for_3rd").notNull().default(50),
pointsFor4th: integer("points_for_4th").notNull().default(40),
pointsFor5th: integer("points_for_5th").notNull().default(25),
pointsFor6th: integer("points_for_6th").notNull().default(25),
pointsFor7th: integer("points_for_7th").notNull().default(15),
pointsFor8th: integer("points_for_8th").notNull().default(15),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const teams = pgTable("teams", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
name: varchar("name", { length: 255 }).notNull(),
logoUrl: varchar("logo_url", { length: 512 }),
ownerId: varchar("owner_id", { length: 255 }), // Clerk user ID, nullable
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const draftSlots = pgTable("draft_slots", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
draftOrder: integer("draft_order").notNull(),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const draftPicks = pgTable("draft_picks", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
pickNumber: integer("pick_number").notNull(),
round: integer("round").notNull(),
pickInRound: integer("pick_in_round").notNull(),
pickedByUserId: varchar("picked_by_user_id", { length: 255 }).notNull(), // Clerk user ID
pickedByType: pickedByTypeEnum("picked_by_type").notNull(),
fix: harden draft timer system with race-condition safety and DRY refactor (#34) - Fix settings action silently resetting timer values when draft speed select is disabled (add null guard before overwriting draftInitialTime/ draftIncrementTime) - Make timer decrement and pick increment atomic using SQL expressions to prevent read-modify-write races between the timer loop and HTTP handlers - Add UNIQUE INDEX on (season_id, pick_number) in draft_picks to prevent duplicate picks from concurrent requests (TOCTOU guard) - Add socket join-draft team ownership validation via DB query - Add iteration cap to autodraft chain while loop (max = totalTeams) - Add draftPaused re-check before firing autodraft chain in make-pick and force-manual-pick - Consolidate all snake draft calculations into calculatePickInfo (DRY); remove duplicated logic from timer.ts, make-pick, force-manual-pick, executeAutoPick, and checkAndTriggerNextAutodraft - Fix calculatePickInfo to return snake-adjusted pickInRound matching draftOrder values instead of raw pre-snake value - Fix timer-update socket events to emit nextPickNumber after a pick instead of the already-completed currentPickNumber - Fix force-manual-pick to validate submitted teamId matches the team whose turn it actually is at the given pick number - Replace inline timer init in draft.start with deleteSeasonTimers + initializeDraftTimers model functions (dead code fix) - Fix draft loader to not crash when getTeamQueue fails (returns []) - Fix misleading 403 message for commissioner picks - Remove dead hidden inputs from league settings form - Document connectedTeams single-instance limitation in socket.ts Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-23 23:23:24 -08:00
timeUsed: integer("time_used").notNull().default(0), // team's time bank (seconds) at the moment the pick was made
createdAt: timestamp("created_at").defaultNow().notNull(),
fix: harden draft timer system with race-condition safety and DRY refactor (#34) - Fix settings action silently resetting timer values when draft speed select is disabled (add null guard before overwriting draftInitialTime/ draftIncrementTime) - Make timer decrement and pick increment atomic using SQL expressions to prevent read-modify-write races between the timer loop and HTTP handlers - Add UNIQUE INDEX on (season_id, pick_number) in draft_picks to prevent duplicate picks from concurrent requests (TOCTOU guard) - Add socket join-draft team ownership validation via DB query - Add iteration cap to autodraft chain while loop (max = totalTeams) - Add draftPaused re-check before firing autodraft chain in make-pick and force-manual-pick - Consolidate all snake draft calculations into calculatePickInfo (DRY); remove duplicated logic from timer.ts, make-pick, force-manual-pick, executeAutoPick, and checkAndTriggerNextAutodraft - Fix calculatePickInfo to return snake-adjusted pickInRound matching draftOrder values instead of raw pre-snake value - Fix timer-update socket events to emit nextPickNumber after a pick instead of the already-completed currentPickNumber - Fix force-manual-pick to validate submitted teamId matches the team whose turn it actually is at the given pick number - Replace inline timer init in draft.start with deleteSeasonTimers + initializeDraftTimers model functions (dead code fix) - Fix draft loader to not crash when getTeamQueue fails (returns []) - Fix misleading 403 message for commissioner picks - Remove dead hidden inputs from league settings form - Document connectedTeams single-instance limitation in socket.ts Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-23 23:23:24 -08:00
}, (t) => ({
// Prevents two concurrent picks from landing in the same slot
uniquePickSlot: uniqueIndex("draft_picks_season_pick_unique").on(t.seasonId, t.pickNumber),
}));
export const draftQueue = pgTable("draft_queue", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
queuePosition: integer("queue_position").notNull(),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const draftTimers = pgTable("draft_timers", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
timeRemaining: integer("time_remaining").notNull(), // seconds
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const autodraftSettings = pgTable("autodraft_settings", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
isEnabled: boolean("is_enabled").notNull().default(false),
mode: autodraftModeEnum("mode").notNull().default("next_pick"),
Claude/redesign autodraft queue c4 kp r (#40) * Redesign autodraft queue system with three-state control and queue-only constraint Core Logic & Database: - Add `queue_only` boolean column to `autodraft_settings` (migration 0031) - Rename autodraft UI states: Off / Next Pick / All Picks (while_on mode maps to All Picks) - `autoPickForTeam`: respects new `queueOnly` param — skips EV fallback when enabled - `executeAutoPick`: auto-disables autodraft + emits socket event when queue empties with queueOnly ON (AC3) - `autodraft-updated` socket event now includes `queueOnly` field Mobile UI Overhaul: - Rename "Lobby" tab → "Available" (AC6) - Add new "Queue" tab to mobile bottom nav with drag-reorder, per-item Draft buttons, and autodraft controls (AC5) - Controls tab retains commissioner tools, notifications, exit; queue controls moved to Queue tab - Turn indicator appears on both Available and Queue tabs Components: - `AutodraftSettings`: replaces toggle+radio with three-state button group (Off | Next Pick | All Picks) + "Only autodraft from queue" switch (AC1, AC2) - `QueueSection`: adds `canPick` prop + per-item Draft buttons for instant drafting when on the clock Desktop (AC4): - Sidebar QueueSection unchanged in position; gains same three-state controls and Draft buttons Tests (AC7): - `autodraft.test.ts`: updated for queueOnly field and socket event shape - `timer-autodraft.test.ts`: new tests for queue-only constraint, auto-shutoff transitions, and all three autodraft states https://claude.ai/code/session_01PYhJicAStoJ2u6q6dV1naB * fix: remove erroneous ?? fallbacks in autodraft socket emissions and add autoPickForTeam tests - Remove `?? true` default on queueOnly in queue-empty auto-disable socket emit (line 488) — was dead code since the column is NOT NULL, but semantically wrong and would have caused client-side UI desync if the type ever relaxed - Remove `?? false` default on the next_pick auto-disable path for consistency - Add app/models/__tests__/auto-pick.test.ts with 6 tests covering the queueOnly constraint: empty queue, all items drafted, partial queue skip, and EV fallback Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: add missing queueOnly prop to AutodraftSettings test fixtures Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * test: rewrite AutodraftSettings tests for three-state button group UI The component was redesigned from a switch + radio buttons to Off/Next Pick/All Picks buttons with a separate queue-only Switch toggle. Updated 17 stale tests and added 5 new tests covering the queue-only toggle and the All Picks/Off button interactions. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude <noreply@anthropic.com>
2026-02-27 22:16:26 -08:00
queueOnly: boolean("queue_only").notNull().default(false),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const commissioners = pgTable("commissioners", {
id: uuid("id").primaryKey().defaultRandom(),
leagueId: uuid("league_id")
.notNull()
.references(() => leagues.id, { onDelete: "cascade" }),
userId: varchar("user_id", { length: 255 }).notNull(), // Clerk user ID
createdAt: timestamp("created_at").defaultNow().notNull(),
});
// Sports Tables
export const sports = pgTable("sports", {
id: uuid("id").primaryKey().defaultRandom(),
name: varchar("name", { length: 255 }).notNull(),
type: sportTypeEnum("type").notNull(),
slug: varchar("slug", { length: 255 }).notNull().unique(),
description: text("description"),
iconUrl: varchar("icon_url", { length: 255 }), // Filename of icon in /public/sports-icons/
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
simulatorType: simulatorTypeEnum("simulator_type"),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const sportsSeasons = pgTable("sports_seasons", {
id: uuid("id").primaryKey().defaultRandom(),
sportId: uuid("sport_id")
.notNull()
.references(() => sports.id, { onDelete: "cascade" }),
name: varchar("name", { length: 255 }).notNull(),
year: integer("year").notNull(),
startDate: date("start_date"),
endDate: date("end_date"),
status: sportsSeasonStatusEnum("status").notNull().default("upcoming"),
scoringType: scoringTypeEnum("scoring_type").notNull(),
// New scoring pattern field (replaces/augments scoringType)
scoringPattern: scoringPatternEnum("scoring_pattern"),
// For qualifying points sports
totalMajors: integer("total_majors"),
majorsCompleted: integer("majors_completed").notNull().default(0),
qualifyingPointsFinalized: boolean("qualifying_points_finalized").notNull().default(false),
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
// EV Calibration parameters (for Elo-based probability generation)
eloCalibrationExponent: decimal("elo_calibration_exponent", { precision: 3, scale: 2 }), // e.g., 0.33
eloMinRating: integer("elo_min_rating").default(1250),
eloMaxRating: integer("elo_max_rating").default(1750),
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
// Simulation status (prevents concurrent simulation runs)
simulationStatus: simulationStatusEnum("simulation_status").notNull().default("idle"),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const participants = pgTable("participants", {
id: uuid("id").primaryKey().defaultRandom(),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
name: varchar("name", { length: 255 }).notNull(),
shortName: varchar("short_name", { length: 100 }),
externalId: varchar("external_id", { length: 255 }),
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
expectedValue: decimal("expected_value", { precision: 10, scale: 4 }).notNull().default("0"),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
export const seasonTemplateSports = pgTable("season_template_sports", {
id: uuid("id").primaryKey().defaultRandom(),
templateId: uuid("template_id")
.notNull()
.references(() => seasonTemplates.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
createdAt: timestamp("created_at").defaultNow().notNull(),
});
export const seasonSports = pgTable("season_sports", {
id: uuid("id").primaryKey().defaultRandom(),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
createdAt: timestamp("created_at").defaultNow().notNull(),
});
export const participantResults = pgTable("participant_results", {
id: uuid("id").primaryKey().defaultRandom(),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
finalPosition: integer("final_position"),
qualifyingPoints: decimal("qualifying_points", { precision: 10, scale: 2 }),
notes: text("notes"),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Scoring System Tables
// Individual scoring events (games, tournaments, races)
export const scoringEvents = pgTable("scoring_events", {
id: uuid("id").primaryKey().defaultRandom(),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
name: varchar("name", { length: 255 }).notNull(), // "2024 Masters", "Super Bowl LIX", etc.
eventDate: date("event_date"),
eventType: eventTypeEnum("event_type").notNull(),
// For playoff events
playoffRound: varchar("playoff_round", { length: 50 }), // "Quarterfinals", "Semifinals", "Finals"
// For qualifying events
isQualifyingEvent: boolean("is_qualifying_event").notNull().default(false),
// Template system (Phase 2.6)
bracketTemplateId: varchar("bracket_template_id", { length: 50 }), // "ncaa_68", "nfl_14", "simple_16", etc.
scoringStartsAtRound: varchar("scoring_starts_at_round", { length: 50 }), // "Elite Eight", "Quarterfinals", etc.
isComplete: boolean("is_complete").notNull().default(false),
completedAt: timestamp("completed_at"),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Results for participants in specific events
export const eventResults = pgTable("event_results", {
id: uuid("id").primaryKey().defaultRandom(),
scoringEventId: uuid("scoring_event_id")
.notNull()
.references(() => scoringEvents.id, { onDelete: "cascade" }),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
// Placement in this specific event
placement: integer("placement"),
// For qualifying events: QP awarded in this event
qualifyingPointsAwarded: decimal("qualifying_points_awarded", { precision: 10, scale: 2 }),
// For playoff events: eliminated or advanced
eliminated: boolean("eliminated"),
// Raw sport-specific data (optional)
rawScore: decimal("raw_score", { precision: 10, scale: 2 }), // strokes, points, time, etc.
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Playoff bracket matches (for display and tracking)
export const playoffMatches = pgTable("playoff_matches", {
id: uuid("id").primaryKey().defaultRandom(),
scoringEventId: uuid("scoring_event_id")
.notNull()
.references(() => scoringEvents.id, { onDelete: "cascade" }),
round: varchar("round", { length: 50 }).notNull(), // "Quarterfinals", "Semifinals", "Finals"
matchNumber: integer("match_number").notNull(), // 1, 2, 3, 4 (for QF); 1, 2 (for SF); 1 (for F)
participant1Id: uuid("participant1_id").references(() => participants.id, { onDelete: "set null" }),
participant2Id: uuid("participant2_id").references(() => participants.id, { onDelete: "set null" }),
winnerId: uuid("winner_id").references(() => participants.id, { onDelete: "set null" }),
loserId: uuid("loser_id").references(() => participants.id, { onDelete: "set null" }),
isComplete: boolean("is_complete").notNull().default(false),
// Optional detailed data
participant1Score: decimal("participant1_score", { precision: 10, scale: 2 }),
participant2Score: decimal("participant2_score", { precision: 10, scale: 2 }),
// Template system (Phase 2.6)
isScoring: boolean("is_scoring").notNull().default(true), // Does this match affect fantasy points?
templateRound: varchar("template_round", { length: 50 }), // "First Four", "Round of 64", "Elite Eight", etc.
seedInfo: varchar("seed_info", { length: 50 }), // "1 vs 16", "11a vs 11b", etc. (for display)
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Aggregated participant qualifying points (for qualifying_points sports)
export const participantQualifyingTotals = pgTable("participant_qualifying_totals", {
id: uuid("id").primaryKey().defaultRandom(),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
totalQualifyingPoints: decimal("total_qualifying_points", { precision: 10, scale: 2 }).notNull().default("0"),
eventsScored: integer("events_scored").notNull().default(0),
// After finalization
finalRanking: integer("final_ranking"), // 1-8 based on QP totals
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Configurable qualifying point values per sports season
export const qualifyingPointConfig = pgTable("qualifying_point_config", {
id: uuid("id").primaryKey().defaultRandom(),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
placement: integer("placement").notNull(), // 1-16
points: decimal("points", { precision: 10, scale: 2 }).notNull(),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Team scores aggregated by sport season
export const teamSportScores = pgTable("team_sport_scores", {
id: uuid("id").primaryKey().defaultRandom(),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
totalPoints: decimal("total_points", { precision: 10, scale: 2 }).notNull().default("0"),
participantsCompleted: integer("participants_completed").notNull().default(0), // How many picks have finished
participantsTotal: integer("participants_total").notNull().default(0), // Total picks for this sport
calculatedAt: timestamp("calculated_at").defaultNow().notNull(),
});
// Overall team standings (snapshot-based for movement tracking)
export const teamStandings = pgTable("team_standings", {
id: uuid("id").primaryKey().defaultRandom(),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
totalPoints: decimal("total_points", { precision: 10, scale: 2 }).notNull().default("0"),
currentRank: integer("current_rank").notNull(),
previousRank: integer("previous_rank"), // For movement indicators
// Tiebreaker data
firstPlaceCount: integer("first_place_count").notNull().default(0),
secondPlaceCount: integer("second_place_count").notNull().default(0),
thirdPlaceCount: integer("third_place_count").notNull().default(0),
fourthPlaceCount: integer("fourth_place_count").notNull().default(0),
fifthPlaceCount: integer("fifth_place_count").notNull().default(0),
sixthPlaceCount: integer("sixth_place_count").notNull().default(0),
seventhPlaceCount: integer("seventh_place_count").notNull().default(0),
eighthPlaceCount: integer("eighth_place_count").notNull().default(0),
participantsRemaining: integer("participants_remaining").notNull().default(0), // Not yet finished
// Expected value tracking (Phase 5.4)
actualPoints: decimal("actual_points", { precision: 10, scale: 2 }), // Points from finished participants
projectedPoints: decimal("projected_points", { precision: 10, scale: 2 }), // actualPoints + EVs of unfinished
participantsFinished: integer("participants_finished"), // Count of finished participants
calculatedAt: timestamp("calculated_at").defaultNow().notNull(),
});
// Daily standings snapshots for historical tracking
export const teamStandingsSnapshots = pgTable("team_standings_snapshots", {
id: uuid("id").primaryKey().defaultRandom(),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
snapshotDate: date("snapshot_date").notNull(),
totalPoints: decimal("total_points", { precision: 10, scale: 2 }).notNull().default("0"),
rank: integer("rank").notNull(),
// Tiebreaker data
firstPlaceCount: integer("first_place_count").notNull().default(0),
secondPlaceCount: integer("second_place_count").notNull().default(0),
thirdPlaceCount: integer("third_place_count").notNull().default(0),
fourthPlaceCount: integer("fourth_place_count").notNull().default(0),
fifthPlaceCount: integer("fifth_place_count").notNull().default(0),
sixthPlaceCount: integer("sixth_place_count").notNull().default(0),
seventhPlaceCount: integer("seventh_place_count").notNull().default(0),
eighthPlaceCount: integer("eighth_place_count").notNull().default(0),
participantsRemaining: integer("participants_remaining").notNull().default(0),
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
// Expected value tracking
actualPoints: decimal("actual_points", { precision: 10, scale: 2 }), // Points from finished participants
projectedPoints: decimal("projected_points", { precision: 10, scale: 2 }), // actualPoints + EVs of unfinished
participantsFinished: integer("participants_finished"), // Count of finished participants
createdAt: timestamp("created_at").defaultNow().notNull(),
});
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
// Expected value tracking (sports-season-specific)
export const participantExpectedValues = pgTable("participant_expected_values", {
id: uuid("id").primaryKey().defaultRandom(),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
sportsSeasonId: uuid("sports_season_id")
.notNull()
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
// Probability distribution (stored as fractions, e.g. 0.1550 = 15.5%)
probFirst: decimal("prob_first", { precision: 6, scale: 4 }).notNull().default("0"),
probSecond: decimal("prob_second", { precision: 6, scale: 4 }).notNull().default("0"),
probThird: decimal("prob_third", { precision: 6, scale: 4 }).notNull().default("0"),
probFourth: decimal("prob_fourth", { precision: 6, scale: 4 }).notNull().default("0"),
probFifth: decimal("prob_fifth", { precision: 6, scale: 4 }).notNull().default("0"),
probSixth: decimal("prob_sixth", { precision: 6, scale: 4 }).notNull().default("0"),
probSeventh: decimal("prob_seventh", { precision: 6, scale: 4 }).notNull().default("0"),
probEighth: decimal("prob_eighth", { precision: 6, scale: 4 }).notNull().default("0"),
// Calculated EV
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
expectedValue: decimal("expected_value", { precision: 10, scale: 4 }).notNull().default("0"),
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
// Metadata
source: probabilitySourceEnum("source").default("manual"), // How probabilities were generated
sourceOdds: integer("source_odds"), // Original odds if source is futures_odds (American odds format)
calculatedAt: timestamp("calculated_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Tournament group stage tables (for FIFA World Cup style tournaments)
export const tournamentGroups = pgTable("tournament_groups", {
id: uuid("id").primaryKey().defaultRandom(),
scoringEventId: uuid("scoring_event_id")
.notNull()
.references(() => scoringEvents.id, { onDelete: "cascade" }),
groupName: varchar("group_name", { length: 10 }).notNull(), // "A" through "L"
createdAt: timestamp("created_at").defaultNow().notNull(),
});
export const tournamentGroupMembers = pgTable("tournament_group_members", {
id: uuid("id").primaryKey().defaultRandom(),
tournamentGroupId: uuid("tournament_group_id")
.notNull()
.references(() => tournamentGroups.id, { onDelete: "cascade" }),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
eliminated: boolean("eliminated").notNull().default(false),
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
// Participant season results (for F1 current points tracking during season)
export const participantSeasonResults = pgTable("participant_season_results", {
id: uuid("id").primaryKey().defaultRandom(),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
currentPoints: decimal("current_points", { precision: 10, scale: 2 }).notNull().default("0"), // Current F1 championship points, etc.
currentPosition: integer("current_position"), // Current standings position
updatedAt: timestamp("updated_at").defaultNow().notNull(),
});
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
// EV snapshots — historical record of participant EV over time (one per participant per sport season per day)
export const participantEvSnapshots = pgTable("participant_ev_snapshots", {
id: uuid("id").primaryKey().defaultRandom(),
participantId: uuid("participant_id")
.notNull()
.references(() => participants.id, { onDelete: "cascade" }),
sportsSeasonId: uuid("sports_season_id")
.notNull()
.references(() => sportsSeasons.id, { onDelete: "cascade" }),
snapshotDate: date("snapshot_date").notNull(),
// Probability distribution (stored as fractions, e.g. 0.1550 = 15.5%)
probFirst: decimal("prob_first", { precision: 5, scale: 2 }).notNull().default("0"),
probSecond: decimal("prob_second", { precision: 5, scale: 2 }).notNull().default("0"),
probThird: decimal("prob_third", { precision: 5, scale: 2 }).notNull().default("0"),
probFourth: decimal("prob_fourth", { precision: 5, scale: 2 }).notNull().default("0"),
probFifth: decimal("prob_fifth", { precision: 5, scale: 2 }).notNull().default("0"),
probSixth: decimal("prob_sixth", { precision: 5, scale: 2 }).notNull().default("0"),
probSeventh: decimal("prob_seventh", { precision: 5, scale: 2 }).notNull().default("0"),
probEighth: decimal("prob_eighth", { precision: 5, scale: 2 }).notNull().default("0"),
calculatedEV: decimal("calculated_ev", { precision: 10, scale: 4 }).notNull().default("0"),
source: varchar("source", { length: 100 }).notNull(), // e.g. 'bracket_monte_carlo', 'f1_standings_model'
createdAt: timestamp("created_at").defaultNow().notNull(),
updatedAt: timestamp("updated_at").defaultNow().notNull(),
}, (t) => ({
uniqueParticipantSeasonDate: uniqueIndex("participant_ev_snapshots_unique").on(
t.participantId, t.sportsSeasonId, t.snapshotDate
),
}));
// Team EV snapshots — historical projected points per team per fantasy season per day
export const teamEvSnapshots = pgTable("team_ev_snapshots", {
id: uuid("id").primaryKey().defaultRandom(),
teamId: uuid("team_id")
.notNull()
.references(() => teams.id, { onDelete: "cascade" }),
seasonId: uuid("season_id")
.notNull()
.references(() => seasons.id, { onDelete: "cascade" }),
snapshotDate: date("snapshot_date").notNull(),
projectedPoints: decimal("projected_points", { precision: 10, scale: 2 }).notNull().default("0"),
actualPoints: decimal("actual_points", { precision: 10, scale: 2 }).notNull().default("0"),
createdAt: timestamp("created_at").defaultNow().notNull(),
}, (t) => ({
uniqueTeamSeasonDate: uniqueIndex("team_ev_snapshots_unique").on(
t.teamId, t.seasonId, t.snapshotDate
),
}));
// Relations
export const sportsRelations = relations(sports, ({ many }) => ({
sportsSeasons: many(sportsSeasons),
}));
export const sportsSeasonsRelations = relations(sportsSeasons, ({ one, many }) => ({
sport: one(sports, {
fields: [sportsSeasons.sportId],
references: [sports.id],
}),
participants: many(participants),
seasonTemplateSports: many(seasonTemplateSports),
seasonSports: many(seasonSports),
participantResults: many(participantResults),
}));
export const participantsRelations = relations(participants, ({ one, many }) => ({
sportsSeason: one(sportsSeasons, {
fields: [participants.sportsSeasonId],
references: [sportsSeasons.id],
}),
results: many(participantResults),
}));
export const seasonTemplatesRelations = relations(seasonTemplates, ({ many }) => ({
seasonTemplateSports: many(seasonTemplateSports),
seasons: many(seasons),
}));
export const seasonTemplateSportsRelations = relations(seasonTemplateSports, ({ one }) => ({
template: one(seasonTemplates, {
fields: [seasonTemplateSports.templateId],
references: [seasonTemplates.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [seasonTemplateSports.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const seasonsRelations = relations(seasons, ({ one, many }) => ({
league: one(leagues, {
fields: [seasons.leagueId],
references: [leagues.id],
}),
template: one(seasonTemplates, {
fields: [seasons.templateId],
references: [seasonTemplates.id],
}),
teams: many(teams),
seasonSports: many(seasonSports),
}));
export const seasonSportsRelations = relations(seasonSports, ({ one }) => ({
season: one(seasons, {
fields: [seasonSports.seasonId],
references: [seasons.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [seasonSports.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const participantResultsRelations = relations(participantResults, ({ one }) => ({
participant: one(participants, {
fields: [participantResults.participantId],
references: [participants.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [participantResults.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const teamsRelations = relations(teams, ({ one }) => ({
season: one(seasons, {
fields: [teams.seasonId],
references: [seasons.id],
}),
draftSlot: one(draftSlots, {
fields: [teams.id],
references: [draftSlots.teamId],
}),
}));
export const draftSlotsRelations = relations(draftSlots, ({ one }) => ({
season: one(seasons, {
fields: [draftSlots.seasonId],
references: [seasons.id],
}),
team: one(teams, {
fields: [draftSlots.teamId],
references: [teams.id],
}),
}));
export const draftPicksRelations = relations(draftPicks, ({ one }) => ({
season: one(seasons, {
fields: [draftPicks.seasonId],
references: [seasons.id],
}),
team: one(teams, {
fields: [draftPicks.teamId],
references: [teams.id],
}),
participant: one(participants, {
fields: [draftPicks.participantId],
references: [participants.id],
}),
}));
export const draftQueueRelations = relations(draftQueue, ({ one }) => ({
season: one(seasons, {
fields: [draftQueue.seasonId],
references: [seasons.id],
}),
team: one(teams, {
fields: [draftQueue.teamId],
references: [teams.id],
}),
participant: one(participants, {
fields: [draftQueue.participantId],
references: [participants.id],
}),
}));
export const autodraftSettingsRelations = relations(autodraftSettings, ({ one }) => ({
season: one(seasons, {
fields: [autodraftSettings.seasonId],
references: [seasons.id],
}),
team: one(teams, {
fields: [autodraftSettings.teamId],
references: [teams.id],
}),
}));
// Scoring System Relations
export const scoringEventsRelations = relations(scoringEvents, ({ one, many }) => ({
sportsSeason: one(sportsSeasons, {
fields: [scoringEvents.sportsSeasonId],
references: [sportsSeasons.id],
}),
eventResults: many(eventResults),
playoffMatches: many(playoffMatches),
tournamentGroups: many(tournamentGroups),
}));
export const eventResultsRelations = relations(eventResults, ({ one }) => ({
scoringEvent: one(scoringEvents, {
fields: [eventResults.scoringEventId],
references: [scoringEvents.id],
}),
participant: one(participants, {
fields: [eventResults.participantId],
references: [participants.id],
}),
}));
export const playoffMatchesRelations = relations(playoffMatches, ({ one }) => ({
scoringEvent: one(scoringEvents, {
fields: [playoffMatches.scoringEventId],
references: [scoringEvents.id],
}),
participant1: one(participants, {
fields: [playoffMatches.participant1Id],
references: [participants.id],
}),
participant2: one(participants, {
fields: [playoffMatches.participant2Id],
references: [participants.id],
}),
winner: one(participants, {
fields: [playoffMatches.winnerId],
references: [participants.id],
}),
loser: one(participants, {
fields: [playoffMatches.loserId],
references: [participants.id],
}),
}));
export const participantQualifyingTotalsRelations = relations(participantQualifyingTotals, ({ one }) => ({
participant: one(participants, {
fields: [participantQualifyingTotals.participantId],
references: [participants.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [participantQualifyingTotals.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const qualifyingPointConfigRelations = relations(qualifyingPointConfig, ({ one }) => ({
sportsSeason: one(sportsSeasons, {
fields: [qualifyingPointConfig.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const teamSportScoresRelations = relations(teamSportScores, ({ one }) => ({
team: one(teams, {
fields: [teamSportScores.teamId],
references: [teams.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [teamSportScores.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const teamStandingsRelations = relations(teamStandings, ({ one }) => ({
team: one(teams, {
fields: [teamStandings.teamId],
references: [teams.id],
}),
season: one(seasons, {
fields: [teamStandings.seasonId],
references: [seasons.id],
}),
}));
export const teamStandingsSnapshotsRelations = relations(teamStandingsSnapshots, ({ one }) => ({
team: one(teams, {
fields: [teamStandingsSnapshots.teamId],
references: [teams.id],
}),
season: one(seasons, {
fields: [teamStandingsSnapshots.seasonId],
references: [seasons.id],
}),
}));
export const participantExpectedValuesRelations = relations(participantExpectedValues, ({ one }) => ({
participant: one(participants, {
fields: [participantExpectedValues.participantId],
references: [participants.id],
}),
feat: implement Expected Value System with ICM probability calculator Implements Phase 5.2 of the EV system with Harville-Malmuth Independent Chip Model for calculating participant placement probabilities from futures odds. ## Key Features ### ICM Probability Calculator - Implements Harville-Malmuth method for distributing probabilities - Converts American odds to championship probabilities - Generates P(1st) through P(8th) for all participants - Column-normalized: each placement sums to 100% across all teams - Works with any number of participants (not limited to 8) ### Admin UI - Futures Odds Entry - Enter American odds (e.g., +550, -200) for championship futures - Live preview of ICM-calculated probability distributions - Displays all 8 placement probabilities - Persists odds for editing on subsequent visits - Automatic probability normalization (removes bookmaker vig) ### Database Schema Updates - Renamed participant_expected_values.season_id → sports_season_id - Updated foreign key to reference sports_seasons instead of seasons - Added source_odds field to store original futures odds - Migration 0025: Column rename and FK update - Migration 0026: Add source_odds field ### Model Layer - participant-expected-value: CRUD operations for probability distributions - Supports multiple probability sources (manual, futures_odds, elo_simulation) - Automatic EV calculation based on league scoring rules - Probability validation and normalization ### Service Layer - icm-calculator: Harville-Malmuth probability distribution - probability-engine: Odds conversion and Elo utilities (for future use) - bracket-simulator: Monte Carlo simulation (for future hybrid approach) - ev-calculator: Expected value computation from probabilities ## Technical Details - Uses exponential decay favoring top positions for strong teams - Preserves championship probability ordering in final distributions - Row sums vary (strong teams ~100%, weak teams lower) - All probabilities between 0-1, mathematically valid - Comprehensive test suite: 97 tests passing ## Future Enhancements - Hybrid approach: ICM pre-playoffs, bracket simulation during playoffs - Integration with league-specific scoring rules - Historical probability tracking for accuracy analysis 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 22:19:46 -08:00
sportsSeason: one(sportsSeasons, {
fields: [participantExpectedValues.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const participantSeasonResultsRelations = relations(participantSeasonResults, ({ one }) => ({
participant: one(participants, {
fields: [participantSeasonResults.participantId],
references: [participants.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [participantSeasonResults.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
// Tournament Group Relations
export const tournamentGroupsRelations = relations(tournamentGroups, ({ one, many }) => ({
scoringEvent: one(scoringEvents, {
fields: [tournamentGroups.scoringEventId],
references: [scoringEvents.id],
}),
members: many(tournamentGroupMembers),
}));
export const tournamentGroupMembersRelations = relations(tournamentGroupMembers, ({ one }) => ({
group: one(tournamentGroups, {
fields: [tournamentGroupMembers.tournamentGroupId],
references: [tournamentGroups.id],
}),
participant: one(participants, {
fields: [tournamentGroupMembers.participantId],
references: [participants.id],
}),
}));
User/chris/ev f1 framework (#93) * feat: EV simulation framework with F1 Monte Carlo simulator - Add EV snapshot tables (participant_ev_snapshots, team_ev_snapshots) and simulation_status column on sports seasons - Add ev-snapshot model with upsert and history query functions - Add simulator framework: types, bracket/F1/golf simulators, registry - F1 simulator: vig-removed ICM weighted draw (pre-season) + race-by-race Monte Carlo from current standings (in-season); per-position column normalization to prevent floating-point EV drift - Add admin simulate route and Run Simulation button on sports season page - Rework futures-odds admin page to save odds then run simulation in one action - Remove recalculate-probabilities route (superseded by simulate route) - Remove EV trend chart panel and associated DB queries Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat: map simulators to sports via simulatorType field Adds a `simulator_type` enum column to the `sports` table so each sport can be assigned a specific simulation algorithm rather than deriving it from the sports season's scoring pattern. - Add `simulatorTypeEnum` (f1_standings, indycar_standings, golf_qualifying_points, playoff_bracket) + `simulatorType` nullable column on `sports` table; migration 0037 - Rewrite simulator registry to key off `SimulatorType` instead of `ScoringPattern`; indycar_standings shares F1Simulator for now - `findSportsSeasonById` now returns `SportsSeasonWithSport` so callers have typed access to `sport.simulatorType` - Simulate and futures-odds actions read `sport.simulatorType`; guard fires before setting `simulationStatus: running` - Admin sport edit page gains a Simulator Type dropdown Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-09 15:34:31 -07:00
export const participantEvSnapshotsRelations = relations(participantEvSnapshots, ({ one }) => ({
participant: one(participants, {
fields: [participantEvSnapshots.participantId],
references: [participants.id],
}),
sportsSeason: one(sportsSeasons, {
fields: [participantEvSnapshots.sportsSeasonId],
references: [sportsSeasons.id],
}),
}));
export const teamEvSnapshotsRelations = relations(teamEvSnapshots, ({ one }) => ({
team: one(teams, {
fields: [teamEvSnapshots.teamId],
references: [teams.id],
}),
season: one(seasons, {
fields: [teamEvSnapshots.seasonId],
references: [seasons.id],
}),
}));