Glossary & data model

1 The two entities

A session has many events. Live metrics ($/hr, stack-in-BB, ICM equity) are derived (§14) — never stored. Two satellite concepts hang off both: Media (§11) and Connections (§4).

2 Field kinds & the three states

Kinds: text select tags datetime money chips count number level hand media rating toggle ref meta.

3 Tag system new

One extensible type:value tag system, used on sessions, events, and media. Adding a capability or category never needs a schema change — just a new namespace or value. This replaces the old confusing formatTags + free tags split.

4 Connections & entity memory new

Every session can connect to canonical pkr2go entities. All nullable — but even when null, we still remember and group so details auto-populate next time.

Why it matters — mass data & auto-populate

• Group similar events. Same venue / operator / recurring event → cluster sessions so we can summarise, compare, and spot patterns across many.
• Auto-populate from what's known. A recurring "Tuesday $80 at The Tote" already has currency, typical structure, buy-in, payouts — recall them so a new session is near-zero input.
• Don't reinvent the wheel. Reuse the existing pkr2go venue/operator/event directory rather than a parallel one.

5 Session fields

✱ = auto-generated (editable + pins). "Shows" = main card, behind More, or gated by a format: tag (§10). Money fields use the §6 convention.

6 Money model — one naming convention reworked

No more guessing "cost vs fee". For every entry action X ∈ {buyin, reentry, rebuy, addon}, the same four parts:

Worked examples

• Standard $100+$20: buyin 100 · buyinFee 20 · buyinBounty 0 · buyinCost 120.
• PKO "$150 + $15" 50/50: buyin 150 · buyinFee 15 · buyinBounty 75 (so 75 prize-pool + 75 bounty) · buyinCost 165.

Same rule for re-entry/rebuy/add-on. So: buyin + buyinFee = buyinCost, and bounty is a clearly-labelled split, never a mystery extra. Open to your naming — if you prefer buyinRake over buyinFee, say so and it's a one-line change here.

Cost basis — how the entry was paid new

A seat won in a satellite costs what you paid for the ticket, not the seat's face value — get this wrong and ROI is fiction. Every entry / re-entry carries:

Worked: a $10,000 seat won from a $1,100 satellite → ticketFace 10,000 · cashPaid 1,100 · invested 1,100. A $1,975 cash means ROI +79.5% on the real outlay, not −80% on the face.

Currency & FX — one home-currency bankroll new

Sessions can be in any currency; the bankroll rolls them into your home currency at a rate captured per session, so the consolidated figure is auditable.

Worked: WSOP net +US$21,500 at fxRate 1.52 → netHome +A$32,680. The ledger shows the rate so the AUD roll reconciles.

7 Event fields

8 Event catalog — what each one shows

The event types below, grouped into 16 card templates (some share a card — e.g. cashed / busted / payout). Key fields = the primary zone; the universal tail is implicit. Trigger: + the log sheet · footer a session action button · auto system. Lifecycle events can carry a place — there's no special "menu only" rule any more.

9 Mirrors — events that write back to the session

Results auto-apply; config is only offered (never silent). [GR6]

10 Gates — format: tags unlock fields & event types

A format: tag (§3) is what makes the relevant fields/events appear. [GR7]

11 Media library new

Photos go two ways: attached to an event and into a tagged, classified library we can batch-analyse later (OCR chips/levels, read receipts, auto-populate fields).

12 Mental-game pulse new

Decision: mood is captured as typed tags, not 1–5 numbers. A controlled vocabulary in the mind: / variance: namespaces — one tap, honest, and filterable like every other tag. Grounded in what coaches track (which "game" you played; tilt & its triggers) but easier to log and to query than sliders.

• Logged from a mood event (one tap, multi-select) and surfaced in the review's Mind glance as colored cards. Defaults private; deeper reflection goes in privateNote.
• The coaching signal is the query: "results on mind:disciplined nights vs not", "every mind:out-of-plan bubble" — because mood is in the same type:value system as everything else (§3).

Sources: Tendler — The Mental Game of Poker · 888poker mental-game guide · vip-grinders session management.

13 The clock & target stack new

The two questions that actually drive late-bag play: "how long until I can stop?" and "do I even need to push?" track answers both — a countdown to the next break / end of day, and a target stack you set so the dashboard tells you whether to step on the gas or coast and rest. You enter the goal once; we do the math every hand.

The countdown

Derived from the blind structure (§14 derive) + the break schedule — no manual entry. Surfaced on the live dashboard as a single line.

Target stack — survival vs accumulation

You set a goal for the next milestone (end of day, the bubble, a pay jump). track compares it to your live stack and gives one verdict — the "be lazy when you don't have to push" signal. A deep stack near bag is worth protecting; ICM says survival often beats marginal accumulation.

• COAST — at/over target with little clock left. You don't have to play a hand to bag the goal; fold up, rest, protect the stack. The "be lazy" green light.
• HOLD — on pace. Keep playing your game; no need to force spots.
• PUSH — under target with the clock running out. You need chips before bag — open up, take the +EV gambles now.

Theory: late-day survival vs accumulation is an ICM call — bagging a healthy stack often outweighs a marginal double near the end of a day. Sources: ICMIZER — ICM & bag-time decisions · PokerNews — bagging up in multi-day events.

14 Derived metrics — never stored, always computed

15 What's still open — candidate fields decide

Surfaced by the review. Each is real for a serious bankroll tracker; none are in v1 unless you say so — flagged so we don't quietly balloon scope.

16 Plain-English term glossary