Portfolio tracker · Preview

Exposure, correlation, and the path

A preview of the surface that would track positions: what is open, how much of it is really the same bet, what has been realised, how much closing line value was captured, and how deep the drawdown went. The real public book behind this site has no open positions at all.

PreviewThis describes a feature that is designed but not built. The figures shown are sample values used to illustrate the layout.

Real open positions: 0
Real graded entries: 33
Real paper units: +2,417.5u
Real money at risk: None

The real book

What is actually open: nothing

This section is the only one on the page with real numbers in it. Everything below it is a preview of a feature that is not built.

Open positions: 0
Current exposure: 0u
Realised paper P&L: +2,417.5u
Pre-game CLV captured: +4.93%

Where the realised result came from

Real ledger

A single P&L figure hides its own composition. On this record the top three winners account for 64.6% of every winning unit, and the single largest winner is worth +1,100u on its own.

Winning units: +4,217.5u
Losing units: -1,800u
Top three winners: +2,725u
Largest single winner: +1,100u
Win rate: 27.3%

A record that loses most of its bets and finishes ahead on a handful of them is not necessarily a bad record — but it is a record with a wide outcome distribution and very little evidence in it. Every row is published →

Where the entries came from

Real ledger

The published record splits into two very different populations, and a tracker that merged them would report a number that means nothing.

Real ledger split by entry phase. CLV is only reported where it is meaningful.
Phase	Entries	Wins	Units	CLV
Pre-game entries	6	0	-600u	Reported
In-play entries	27	9	+3,017.5u	n/a*

* 27 in-play entries are excluded from every CLV figure on this site. Their entry-to-close gap measures innings elapsed rather than price discovery. The 6 pre-game entries are the entire CLV sample.

The preview

What the tracker would show

Sample values for an imagined book, laid out the way the surface would lay them out. None of these numbers describes anything that happened.

PreviewPortfolio — sample metricsSample data

Open positions: 2
Total exposure: 1.5u
Largest correlated cluster: 1.0u
Realised P&L: +18.4u
CLV captured: +2.1 pts
Max drawdown: -9.2u

Sample values for a hypothetical book. The real public paper book has no open positions and none of these figures is a measurement.

PreviewPortfolio — sample open positionsSample data

Sample open positions with stake, entry price, current fair estimate, unrealised CLV, and correlation group. Preview data only.
Market	State	Stake	Entry	Current fair	Unrealised CLV	Correlation group
BOS -1.5	BET	1.0u	+138	+128	+1.84 pts	BOS / TB
SD F5	WAIT	0.5u	-134	-145	+1.92 pts	SD / COL

Invented positions on invented prices. No wager was placed, and no price shown here was available anywhere.

Snapshot 23 Aug 2026, 18:05 UTC · Sample snapshot timestamp — fixed value, not a feed reading. The sample positions do not change, settle, or expire.

Open and pending

Preview

Open means accepted and unsettled. Pending means submitted and not yet confirmed, which is a real state at a real book and a common way exposure gets counted twice or not at all.

Sample exposure across the two positions above is 1.5u. The real book holds 0.

Unrealised CLV

Preview

An open position can already be measured against where the market has moved since entry. It is not a profit and it is not settled, but it is the first evidence available about whether the read was early.

It becomes a real reading only at the close. Until then it is a running estimate that can reverse completely.

Drawdown

Preview

Peak-to-trough decline is what decides whether a staking plan survives long enough to be tested. A plan with a positive expectation and a drawdown deeper than its owner can hold produces the same outcome as a plan with no edge.

Correlation

Two positions on one game are one position

This is the failure that makes tracking worth building. Everything else on this page is arithmetic; this part is the reason a spreadsheet is not enough.

Why the cluster figure sits next to exposure

Two entries on the same game are not two independent positions. A tracker that adds them as if they were will understate risk, which is why the cluster figure sits next to total exposure.

In the sample above, both rows sit in different game groups, so the largest correlated cluster is smaller than total exposure. Put both on the same game and the two numbers converge — which is exactly the case where a naive exposure figure is most misleading.

Where correlation hides

Same game, different markets. A moneyline and a run line on the same team are close to the same bet with different variance.
Same driver, different games. Totals across a slate move together with weather, umpire assignment, and pace.
Same read, different days. A model that is wrong about one team is wrong about it repeatedly, which turns a sequence of entries into one large bet on one assumption.
Same book. Concentration at a single operator is correlation with the operator, not with the market.

Requirements

Six questions a tracker exists to answer

Each carries its own status. Current means the published record already answers it. Preview means designed and not built. Roadmap means it depends on something that does not exist.

How much is at risk right now?

Preview

Total open stake, before any adjustment. The simplest number on the page and the only one most spreadsheets track, which is why most spreadsheets understate risk.

How much of that is the same bet twice?

Preview

Two entries on one game are one position wearing two labels. Exposure that ignores correlation reports a diversified book where none exists.

Where did the realised result come from?

Current

A total tells you nothing about repeatability. The split between a few large winners and a long tail of losers is the part that says whether a record is a process or a streak.

Was the read ahead of the market?

Current

Closing line value, measured per entry and aggregated. It is the fastest-accumulating evidence available, and it is the number this project is actually graded on.

What did the path look like?

Preview

Peak-to-trough drawdown, not just the endpoint. Bankrolls end during the drawdown, not at the average, and a plan that cannot survive the path never reaches the expected value.

What is available to stake?

Roadmap

Balances across books, capital tied up in open positions, and what is free. Requires book integrations, which do not exist here in any form.

Not built

What this preview cannot do

Everything requiring an account, a book, or a feed is absent. Listing it is more honest than shipping a demo that implies otherwise.

Capabilities a working tracker would need, and their real status.
Capability	Status	Detail
Book balances and transfers	Roadmap	Requires sportsbook account integration. VigPulse has none, holds no affiliate relationship, and connects to no operator.
Automatic position capture	Roadmap	Positions would have to be read from a book or entered by hand. There are no accounts on this site, so there is nowhere to put them.
Live unrealised value	Roadmap	Marking an open position to the current market needs a market feed. The site is a static export with no feed connected.
Correlation model	Preview	Grouping by game is the crude version and it is what the sample table shows. Correlated exposure across games, divisions, and totals is a modelling problem, not a grouping key.
Tax and accounting export	Roadmap	Nothing here produces a financial record, and no figure on this site should be used for one.

Book balancesComing soonPosition importComing soonLive mark to marketComing soonCorrelation modelComing soon

No accounts, no money, no product

There is no login on this site, nothing is for sale, and no real money has been staked in the published record. A tracker with no positions to track is not a product — which is why this page is labelled a preview on every screen rather than presented as a working tool.