Institutional Practice — Pre-Launch
Quantitative Admissions Intelligence for Counselors & Schools
We are developing decision software that helps counselors construct and evaluate a student’s entire college portfolio — not simply estimate admission odds one school at a time. It is being built for independent educational consultants (IECs), private schools, and school counseling offices that already do portfolio construction by hand and want quantitative admissions analysis alongside professional judgment.
The current system is a research-based prototype. It is not a validated production model, and institution-specific calibration is not currently deployed.
The Problem
The problem is not predicting one school. It’s constructing the portfolio.
Counselors commonly work from fragmented historical data, static acceptance-rate tools, school-by-school probability estimates, spreadsheets, subjective judgment, and hard-won institutional knowledge that is difficult to formalize. Each of those inputs is reasonable in isolation. Together they are hard to reconcile into a defensible view of a college list.
Much of the existing admissions software ecosystem is organized around individual institutions or individual student-school estimates. But a student does not submit one application. They submit a portfolio of applications driven by a single profile, so the relevant optimization problem is joint rather than isolated: how many reaches can a list support, where the binding Early Decision application does the most good, and whether an additional school changes the distribution of outcomes at all.
We set out the full argument, including a worked illustrative example, in why admissions should be treated as a portfolio problem. The conceptual mechanics of the prototype are shown in our methodology and prototype demonstration.
What We Are Building
An admissions decision engine built around the portfolio.
The system is designed to evaluate a student’s portfolio using Monte Carlo simulation, student-school factor modeling, correlation between outcomes, Early Decision optimization, institutional variables, and — where it can lawfully be provided — historical outcome data.
Monte Carlo simulation is simpler than it sounds. Rather than producing one static probability for each school, simulation can evaluate thousands of possible portfolio outcomes and estimate how different school combinations behave together. Instead of a column of percentages, the output describes a distribution: how often a given list produces at least one admission the student would actually enroll at, and how that changes when a school is added, removed, or moved between rounds.
This is the architecture we are developing; the current system is a research-based prototype rather than a validated production model.
System Architecture
Three core system layers
Layer 01
Portfolio Optimization
School-by-school estimates are not sufficient on their own. A college list is a connected set of decisions: the same student profile drives every application, so outcomes can be correlated rather than independent. Adjusting one school changes the shape of the whole list — the balance of likely admissions, the concentration of risk, and the chance of ending a cycle without an offer the student would accept. The objective is not maximizing the probability of admission to a single institution; it is constructing a portfolio that behaves sensibly as a whole.
Layer 02
Early Decision Optimization
Early Decision is a binding commitment mechanism, and its strategic value varies by institution and by student. The relevant question is not whether early rounds look favorable in aggregate, but where a single binding application creates the most useful incremental leverage within a specific student's portfolio. We are developing models to evaluate how Early Decision could change the expected outcomes of a student's broader portfolio.
Layer 03
Institutional Factors
Headline acceptance rates flatten a great deal of variation. Long-term institutional models could incorporate factors that generic public acceptance-rate tools cannot fully represent: school-specific historical outcomes, academic and geographic context, program or intended-major context, institutional priorities, historical yield behavior, and application-round behavior. The long-term system is designed to incorporate institution-specific historical data where appropriate and where an institutional partner can lawfully provide it.
The reasoning behind Layer 01 is developed at length in why admissions should be treated as a portfolio problem.
Development Roadmap
From generic admissions data to institution-specific intelligence.
A model built only on public data describes the national picture. It cannot represent how a particular school’s students are read in their own context, or how a given practice’s applicant pool differs from the national distribution. The long-term institutional model is intended to work as follows.
- 01Institutional dataAn institution provides appropriate historical outcomes data — student profiles, applications submitted, and results received across prior cycles — under terms it can lawfully support.
- 02Model calibrationNadia Moore develops a model calibrated to that institution's student population, applicant pool, and stated objectives rather than to a national average.
- 03Portfolio evaluationThe system evaluates student portfolios against that institutional context, so estimates reflect the population the tool is actually used on.
- 04Decision supportCounselors receive decision-support outputs: portfolio comparisons, Early Decision placement analysis, and the assumptions behind each result.
- 05RecalibrationAs new cycles produce new outcomes, the model is refit, and prior estimates can be checked against what actually happened.
These institution-specific calibration capabilities are part of the product roadmap and are not currently deployed. Nadia Moore does not currently hold proprietary institutional datasets, and no customer historical data is being collected today.
Who It Is For
Built for practitioners who already do this work by hand
Independent educational consultants
For practitioners managing complex student portfolios who want quantitative decision support without replacing their professional judgment.
Private schools
For counseling offices seeking institution-specific analytical infrastructure across a larger student population.
Sophisticated counseling teams
For organizations that want to combine professional counseling judgment with quantitative portfolio analysis.
Families looking for individual admissions guidance rather than software should visit the four-year family advisory practice, which serves a capped cohort of twelve families per cycle.
Product Status
Where the product stands today
Current Stage
Research-based prototype
The current system is in prototype development. The underlying methodology and interface concepts have been developed, but production-grade software, institution-specific calibration, and validated institutional models remain on the roadmap.
With appropriate development funding and institutional collaboration, our objective is to build Version 1 over the next development cycle.
We make no claims of validated predictive accuracy, institutional customers, or production deployment. The admissions research underlying the model’s priors is published openly in Admissions Intelligence.
Roadmap
How the system is intended to develop
01
Research Prototype
Public admissions data and research-based priors.
Current stage
02
Portfolio Engine
Monte Carlo portfolio simulation and student-school factor modeling.
Development objective
03
Institutional Calibration
Institution-specific historical outcome modeling where appropriate data is available.
Development objective
04
Counselor Platform
Decision-support interface for IECs and school counseling teams.
Development objective
Stages 02 through 04 represent development objectives, not currently deployed functionality.
Institutional Conversation
Discuss the institutional model
We are currently speaking with a small number of counselors and schools as we develop the institutional version of the platform.
If you work with complex student portfolios and are interested in helping shape the decision infrastructure behind the next generation of admissions counseling, we’d like to hear from you. Conversations are conducted directly by the founder and include a walkthrough of the prototype, an honest account of what it does not yet do, and where its assumptions are weakest.
Detailed model mechanics and sandbox access are shared under a mutual Non-Disclosure Agreement. The NDA covers the implementation, not the description — what the product is intended to do and what stage it is at is set out on this page.
Common Questions
Questions counselors and schools ask first
- What is Nadia Moore building?
- A quantitative admissions decision engine designed to evaluate a student's entire college portfolio rather than estimating admission odds one school at a time. The intended architecture combines student-school factor modeling, correlation between outcomes, Monte Carlo portfolio simulation, and Early Decision analysis. The current system is a research-based prototype rather than a validated production model.
- Who is the product designed for?
- Independent educational consultants (IECs), private schools, and school counseling offices — practitioners who already construct college lists by hand and want quantitative decision support alongside, not instead of, professional judgment.
- Is the software currently available?
- No. The product is pre-launch. A research-based prototype exists and the methodology and interface concepts have been developed, but production software, institution-specific calibration, and validated institutional models remain on the roadmap. There is no commercial availability and no pricing.
- How is this different from a college probability calculator?
- A probability calculator answers a per-school question: how likely is this student here? The portfolio question is different: given this student's profile and objectives, how should the whole list be constructed? Because applications share a single student profile, outcomes are not independent, so optimizing each school separately does not necessarily optimize the portfolio as a whole.
- What is Monte Carlo simulation doing in an admissions model?
- Rather than producing one static probability per school, simulation evaluates thousands of possible portfolio outcomes and estimates how different school combinations behave together — for example, how often a given list produces at least one admission the student would actually enroll at. It is a way of describing a distribution of outcomes rather than a single number.
- Will the system use institution-specific data?
- That is the long-term design. The system is intended to incorporate institution-specific historical data where appropriate and where an institutional partner can lawfully provide it. Nadia Moore does not currently hold proprietary institutional datasets, and institution-specific calibration is not currently deployed.
- When will the institutional product be available?
- There is no announced availability date. With appropriate development funding and institutional collaboration, our objective is to build Version 1 over the next development cycle. We are speaking with counselors and schools now to shape that work.
Help shape the institutional model.
We are seeking conversations with experienced counselors, IECs, and schools as we develop the first production version of Nadia Moore’s admissions intelligence platform.