Value Creation AnalyticsAn independent modeling initiative by Monte Fisher

Forensic modeling · Decision analytics

Scenarios you run, scored on what your choices cost.

One scoring engine, applied to eleven operating problems. Set the rules, make the calls, watch the ledger. Working models, not PowerPoints.

The flagship · five parts

The Spread

A full stocker cattle decision model: price slide, cost of gain, breakeven, and a hedge signal. Then a twelve-turn season with 500 head across five lots, sickness and slow-gain events, and a decision ledger that shows what each call cost you.

Input provenance

Every input tagged · The Spread

Feeder cattle futures CME / Barchart, refreshed at load. Live
Cash market reports USDA AMS regional auction summaries. Live
Corn and feed basis CBOT plus local basis, entered by the operator. Manual
Cost of gain Derived from ration, ADG and days on feed. Estimate
Current weights No live feed. Needs a weigh-in process on the operation. Not linked

The model surfaces and suggests; it does not decide. It shows the economic consequences of the available choices.

Step 1

The model

Breakeven, price slide, cost of gain, and a LOCK/WAIT hedge signal.

Step 2

Set the rules

Write the policy the agent runs on. Your thresholds, your tolerances.

Step 3

Run the season

Twelve turns, 500 head, five lots. Sickness and slow gain arrive uninvited.

Step 4

Command center

Watch the position, the ledger, and what each decision cost.

Tool

Feed planner

Plan the ration, get the date. Feeds cost of gain back into the model.

Audit

Data sources

Every input, every tag, every named source. Nothing hidden.

Open The Spread

The portfolio

Same engine, other domains

Each one puts you inside a situation, takes your decisions turn by turn, and scores what they cost. The domain changes; the method does not.

Simulation · AI projects · 20 turns

The AI Markup

A vendor is overcharging you. The invoices look fine. Find the markup.

Runnable

Simulation · Fraud · 14 turns

The Reconciliation

Workers' comp payroll fraud at a trucking firm. Can you prove it?

Runnable

Simulation · Fraud · 9 schemes

Blind Spot

Nine ways employees quietly take from a small business, and what each costs.

Runnable

Simulation · AI projects

The Navigator

An AI project simulation: scope, vendors, and where the money goes.

Runnable

Brief · 32 hubs

The Airport Brief

Transfers, traps and ground truth. Judgment under time pressure.

Reference

Simulation · Judgment · 17 countries

The Unwritten Brief

Eight high-stakes situations. No trick answers, only trade-offs.

Runnable

Written

The guides

Each model has a written companion — the same argument, made in prose for anyone who wants the reasoning before the simulation.

Why most people lose money buying and selling cattle

Pairs with The Spread

10 signs your AI contractor is overcharging you

Pairs with The Markup

How a Texas trucking family stole $9M in comp premiums

Pairs with The Reconciliation

9 ways employees quietly steal from a small business

Pairs with Blind Spot

Where school bus fleet budgets actually leak

Forensic analysis

What a forensic accountant sees in a $60B deal

Forensic analysis

Method

The honest version

Every input in every model carries a tag: Live pulled from a named source, Manual entered by the operator, Estimate derived from other inputs, Not linked no feed exists and the model says so.

Most modeling hides this. Showing it is the point — a model you can't audit is a slide with arithmetic on it.

These aren't products for sale. They're proofs of capability. The differentiator is forensic judgement — CPA, CFE, Lean Six Sigma — not the software.

The governance and controls side of this work lives at Fisher Governance →