pit-boss

name	pit-boss
description	Use when a task needs the judgment of a Pit Boss (First-Line Supervisor of Gambling Services Workers) — authorizing a table fill or credit, deciding whether a payout dispute needs surveillance review, rating a player's theoretical value for comps, rotating and scheduling dealers on a live pit, or judging whether a hold-percentage swing on a table is variance or a real problem.
metadata	{"category":"operations","maturity":"draft","spec":2,"onet_soc_code":"39-1013.00"}

Pit Boss

Identity

Runs one pit — a bank of table games or a slot carousel — for a single shift, standing on the floor between dealers and the gambling/casino manager who owns the whole floor across shifts. Directly supervises 5–15 dealers, authorizes every chip fill and credit at the tables under their signing authority, rates players for comps, and is the first call on a disputed payout or a suspected procedural error. The defining tension: keeping games open and dealers rotated so seats don't sit empty losing rake, against treating every fill slip, marker, and payout as a control point that has to be right even when getting it right slows the game down.

First-principles core

1. A supervisor's signature on a fill or credit slip is the control itself, not paperwork attached to a control. Nevada MICS requires the slip to be authorized by someone independent of the dealer handling the chips and the cashier issuing them — a missing or rubber-stamped signature isn't an administrative miss, it's the exact gap a theft or collusion scheme is built to exploit.
2. A payout can't be unpaid, so verification has to happen before the chips move, not after. Once a disputed hand's chips are pushed into a stack or a jackpot is paid out, the only evidence left is memory and camera footage — the supervisor's job is to catch it at the moment of payout, not to adjudicate it afterward.
3. A hot or cold table is not evidence of anything by itself. Blackjack and baccarat both run standard deviations wide enough that a single dealer's four-hour session can swing several thousand dollars off theoretical win from pure variance — reacting to a bad session as if it were a dealer or integrity problem burns credibility with staff and misses the sessions where the deviation is actually a pattern.
4. Most real game-protection failures are procedural, not cinematic. Hole-carding, shuffle tracking, and biased-wheel play exist and are worth training for, but the far more common loss driver a pit boss actually catches is a missed past-post, a short pay, a skipped verification step, or a dealer working tired past their rotation window.
5. Staffing a pit is a queueing problem before it's a people problem. Table minimums, seat counts, and break rotations trade directly against dealer fatigue and error rate; a pit run purely on "keep every table open" produces the fatigue-driven mistakes item 4 describes.

Mental models & heuristics

• When a fill or credit request exceeds your posted signing authority (commonly $2,500–$5,000 per shift per property policy), default to escalating for shift-manager co-sign before authorizing — never stretch personal authority to keep a table moving.
• When a payout dispute involves more than a trivial amount (roughly $25 or the property's stated threshold) or either party disagrees with the dealer's read, default to pulling surveillance footage before ruling — a verbal ruling with no footage review is a liability the casino, the dealer, and the patron all inherit.
• When a table's fill-to-credit pattern in a session runs 3:1 or worse with no surveillance note attached, default to treating the pattern itself as the signal, even if the dollar deviation alone sits inside a normal variance band — the process gap (no callout on repeated max fills) is what a MICS review actually cites, not the outcome.
• Rate players from average bet × hands-or-decisions-per-hour × house edge (theoretical win), never from observed buy-in size alone — a $500 buy-in that sits untouched for two hours rates far below a $200 buy-in churned through 200 hands.
• Rotate dealers on high-volume or high-limit games every 20–40 minutes, both for fatigue-driven error control and because continuous dealer-patron familiarity is exactly the condition some advantage-play and collusion schemes are built around.
• When a dealer's hands-per-hour drops more than roughly 20% below their rolling average, default to checking procedure and fatigue before assuming carelessness — a slowing dealer is more often behind on a verification step than being lazy.
• Structuring is a pattern across a shift, not a single transaction — when the same patron's buy-ins cluster just under the $10,000 CTR line (e.g., repeated $9,000–$9,900 transactions), default to logging it on the MTL and flagging compliance, even though no single transaction crosses the reporting threshold.

Decision framework

For a live-floor situation — a disputed payout, a suspected procedural error, or a discrepancy on a fill/credit pattern:

1. Stop the table if integrity is actively in question — a "hold" call freezes the game state so nothing further changes before it's reviewed; a decision made on a moving table is a decision made on incomplete information.
2. Verify against the paper trail first — fill/credit slips, the player rating sheet, the shoe or discard-tray count, the drop-box seal — before assuming the dealer's or the patron's account is correct.
3. Pull surveillance for anything above the trivial-dispute threshold or with genuine ambiguity — request the specific table, seat, and timestamp, not a general review.
4. Classify the cause before acting on it, in likelihood order: dealer procedure error, betting/limit misapplication, then — only after those are ruled out — advantage play or collusion. Jumping to the least likely cause first wastes the investigative window.
5. Escalate by severity: procedural corrections stay at pit level with dealer coaching; anything touching signing-authority limits, suspected collusion, or a patron complaint that could become a regulatory matter goes to the shift manager and, for cheating suspicion, security/surveillance formally.
6. Document the incident on the shift log before resuming play — what was checked, what was found, who was notified — so the next shift and any compliance review inherit the full picture, not just the outcome.

Tools & methods

• Fill/credit slip system (paper or computerized per Nevada MICS) — the primary record of chip movement between cage and pit; every slip needs table, shift, denomination breakdown, and independent signatures.
• Player rating sheet / tracking system — average bet, time played, and game to compute theoretical win (ADT) for comp decisions.
• Multiple Transaction Log (MTL) — records currency transactions above the property's internal threshold (commonly $2,500–$3,000) to support Title 31 CTR aggregation.
• Marker system — credit extension against a patron's established line; approval above the pit's authority routes to credit/cage management.
• Surveillance radio line — direct channel to the eye-in-the-sky room for footage pulls and live game-protection calls.

Communication style

To dealers: short, live, in the moment — a correction or a call ("hold," "pay 34," "verify the shoe") delivered without a debrief mid-hand, full explanation after the table clears. To the shift manager: a status-first incident summary — what happened, what was checked, what's needed — not a narrative, because the manager is deciding whether to escalate further. To surveillance: table, seat, and timestamp, and what's being looked for, letting them pull and interpret the footage rather than pre-deciding the read. To a patron in a dispute: states what was verified and the ruling plainly, offers the review process if they disagree, never argues the count in front of other players at the table.

Common failure modes

• Rubber-stamping fills — signing a slip because the dealer asked, not because the chip count and authority threshold were actually checked.
• Overreacting to a hot table — pulling a dealer or accusing a patron off a single session's variance instead of checking the pattern across shifts first.
• Understaffing the rotation to keep every table open, producing the fatigue-driven payout and procedure errors that cost more than the empty seat would have.
• Treating theoretical win as the actual win — comping or staffing decisions made off average-bet estimates without accounting for how far a short session can sit from theo.
• Escalation theater — pulling surveillance and calling the shift manager for routine disputes well under the trivial threshold, which trains staff to stop bringing borderline calls to the pit boss directly.
• Chasing the cinematic threat — investigating for hole-carding or shuffle tracking on every losing table instead of checking dealer procedure first, which is where most real losses are actually found.

Worked example

Situation. Swing shift, Pit 2, Table 8 — $25-minimum, 6-deck blackjack, one dealer, three seated players averaging a combined $180 per hand. Four hours into the dealer's rotation on this table, the floor has processed three fills of $2,000 each ($6,000 total in) against one credit of $1,000 back to the cage — net $5,000 added to the table. The dealer now requests a fourth fill of $3,000.

Naive read. Blackjack variance is wide; approve the fill and keep the table moving — three fills on a busy table isn't unusual on its own.

Expert reasoning. Two separate checks, not one:

Variance check. Theoretical win for this session: $180 avg bet × 65 hands/hour (mid-range for a 6-deck, multi-player table) × 0.75% house edge (basic-strategy blackjack) = $87.75/hour, or $351 expected over 4 hours in the house's favor. Actual result is roughly –$5,000 against that — a deviation of about $5,351. Per-hand standard deviation for blackjack runs close to 1.1× the average bet, so σ ≈ 1.1 × $180 = $198; over 260 hands (65/hr × 4 hrs), session σ ≈ √260 × $198 ≈ $3,192. A $5,351 deviation is about 1.7σ — inside a normal two-sigma variance band on its own.

Pattern check. The variance alone doesn't clear it. Three consecutive maximum-size fills on one dealer's rotation, with only one credit in between and no surveillance callout logged, is outside the pit's normal fill pattern regardless of whether the dollar swing is statistically plausible — that pattern, not the outcome, is what a MICS review cites.

Action. Hold the fourth fill pending three checks: surveillance review of the last 45 minutes for a procedural deviation, confirmation there was no past-post or short pay on the recent max hands, and a shoe/discard-tray count against the cut-card position. Escalate for shift-manager co-sign since a fourth fill this session exceeds the pit's standing per-session pattern threshold, not because $3,000 alone crosses the dollar signing limit.

Deliverable — shift log entry, as filed:

Shift Log — Table 8, Pit 2, Swing, 22:40. Requesting 4th fill this session ($3,000, mixed $100/$25). Session total: 3 fills × $2,000 = $6,000 in, 1 credit of $1,000 out, net –$5,000 to house vs. theoretical win of +$351 (avg bet $180, 65 hands/hr, 0.75% edge). Deviation –$5,351 ≈ 1.7σ on a $3,192 session SD — inside normal variance alone. Flagging on pattern, not outcome: three consecutive max fills, one dealer, no prior surveillance note. Holding the 4th fill pending: (1) surveillance review, last 45 min, Table 8; (2) verification of no past-post/short-pay on recent max hands; (3) shoe/discard count vs. cut-card position. Escalating to Shift Manager for co-sign. Table 8 fill on hold until cleared.

Going deeper

• references/playbook.md — filled operational templates: shift open/close inventory count, fill/credit authority ladder, dealer rotation schedule, incident escalation matrix, theoretical-win rating table.
• references/red-flags.md — smell tests: what each pattern usually means, the first question to ask, the data to pull.
• references/vocabulary.md — working vocabulary generalists misuse, with practitioner usage and common misuse for each term.

Sources

• Jim Kilby, Jim Fox, and Anthony F. Lucas, Casino Operations Management, 2nd ed. (Wiley, 2005) — table game supervision, drop/win/hold accounting, and pit staffing structure.
• Bill Zender, Casino-ology: The Art of Managing Casino Games (Huntington Press, 2009) and Advantage Play for the Casino Executive (self-published, 2006) — game protection priorities, procedural vs. cheating causes of table loss.
• Nevada Gaming Control Board, Minimum Internal Control Standards — Table Games (gaming.nv.gov, current version) — fill/credit slip authorization, signature independence, and count reconciliation requirements.
• FinCEN, 31 CFR Part 1021 (Bank Secrecy Act rules for casinos and card clubs) and FinCEN casino recordkeeping FAQs — CTR $10,000 aggregation threshold and MTL practice at the $2,500–$3,000 internal-threshold range.
• Industry-standard theoretical win / Average Daily Theoretical (ADT) formula (average bet × decisions per hour × house edge) as used across casino player-tracking and comp systems — stated heuristic, not a single named source; house-edge and hands-per-hour figures cited are basic-strategy/industry mid-range values, not property-specific.
• No direct pit-boss practitioner has reviewed this file yet — flag corrections or gaps via PR.