Growth is rarely a straight line.
New customer adoption and expansion can follow nonlinear paths. The key question is when growth accelerates, matures, or changes direction.
The research framework
The physics of equities.
Without the crystal ball.
Financial statements show what happened. Our research asks what drives a business to change—and how those mechanisms may inform valuation. This is a conceptual preview, not a published forecasting model.
Why physics belongs in the conversation
Businesses move. Assumptions should too.
A high-growth company is not a static balance sheet. Demand can accelerate, slow or saturate; customers influence future growth; and the effects of a decision may take years to emerge. We study these as dynamic systems while keeping the mathematics behind the scenes.
Yesterday can affect tomorrow.
Customer experience, deployment cycles and switching costs may create feedback or delayed responses. A business can carry the imprint of its past decisions.
One path is not the whole story.
Competition, budgets, macroeconomic shifts and execution risk create alternative outcomes. Any quantitative treatment must be checked against evidence.
Where DCF fits
We are not replacing cash-flow valuation with physics.
Discounted cash flow turns assumptions about future cash generation into a present-value estimate. Our intended contribution is to examine the operating assumptions upstream of DCF: adoption, revenue composition, profitability, and their interactions. A valuation remains conditional on those assumptions.
Four connected layers
This is a high-level research map, not a disclosure of the complete modelling engine or a claim that a forecasting product is available.
Company operating drivers
Adoption and retention, product pricing, customer segments, cost structures, reinvestment, margins, and cash flows.
Dynamic modelling
Adoption patterns, feedback and stochastic disturbances—introduced only where evidence justifies the added complexity.
Financial translation
Translate business assumptions into cash-flow and DCF scenarios, distinguishing intrinsic-value estimates from market prices.
Out-of-sample evaluation
Freeze the information available at a historical forecast date and evaluate predictions against later observed company fundamentals.
Our standard of evidence
What counts as a meaningful historical test?
For each published case study, we intend to specify the forecast origin, information cutoff, target variable, forecasting horizon, baseline methods, observed outcome, and uncertainty assumptions. A historical test is only meaningful if later information did not leak into the original model inputs.
StockPhys will not present reconstructed predictions as if they had been made and archived earlier. We will distinguish true timestamped forecasts from retrospective backtests.
Limits are part of the method.
- Identifiability: Financial data may not uniquely determine customer behavior or latent model parameters.
- Model misspecification: A mathematically consistent model can still omit crucial business mechanisms.
- Mathematics is not validation: Path-integral and other advanced formulations do not establish predictive superiority without comparative testing.
- Market-price distinction: An intrinsic value estimate can differ from the traded price for prolonged periods.
This content describes a research project and potential software functionality. It is not a recommendation concerning any security.
What comes after this preview?
See the staged plan for research case studies and interactive forecasting tools.