Invest with Probability, Not Opinions

Screen ~980 stocks → simulate 2,000 returns → build optimal portfolios with measurable odds.

Avg Merkapital Confidence Score™ (Top 10): 82
Median P(Up 1Y): 78%+

For diligent investors, advisors, and allocators — not noise-driven traders.

980+ stocks analyzed daily40+ years of research foundationsMerkapital Simulation Engine™: logic you can audit

Live probabilities and today's confidence metric

Why This Matters

Most investors rely on price targets. Merkapital quantifies probability.

“Probability beats predictions. Insights beat opinions.”

Transparent, evidence-backed modelingNo black boxes. No hidden weights.Trackable, auditable decision logic
~980
Stocks Screened
5
Research-Backed Factors
2,000
Simulations per Ticker
14
Academic Citations

“Merkapital's Conviction Scores replaced my gut calls with defensible odds.”

— RIA, $350M AUM

How It Works

The pipeline from data to conviction.

~980 stocks screened
5 research-validated factors
2,000 simulations per stock
probabilistic return ranges
top-10 portfolio construction

A Three-Stage Selection Pipeline

Nearly 1,000 stocks → narrowed to the 10 with highest risk-adjusted probability.

STAGE 01

Universe & Data

~990 stocks

S&P 500, large/mid-cap NYSE and NASDAQ, select ADRs. Ten fundamental data points per stock: price, earnings estimates, profitability, analyst targets.

Micro-cap and OTC excluded.

STAGE 02

Scoring & Simulation

5 factors + 2,000 paths

Scored 0–100 across five factors, then 2,000 GBM paths per stock via Merkapital Simulation Engine™. Full 1-year return distribution.

Bayesian shrinkage on drift. Per-stock volatility from Forward P/E.

STAGE 03

Portfolio Construction

Top 10, equal weight

Ranked by risk-adjusted composite: Sharpe proxy, upside probability, downside protection, quality, data completeness. Top 10 equal-weight, validated by 5,000 portfolio simulations.

Equal weight outperforms optimization (DeMiguel et al., 2009).

Five Factors. Four Decades of Research.

Peer-reviewed factors, weighted by empirical predictive power for 1-year returns.

30%

Earnings Growth

Forward EPS growth + YoY momentum. Strongest 1-year predictor.

O'Shaughnessy (2011); Chan, Jegadeesh & Lakonishok (1996)
20%

Analyst Consensus

Price vs consensus target. Predicts direction ~60% of the time.

Brav & Lehavy (2003); Womack (1996)
20%

Value (PEG Ratio)

Growth cheap or expensive. PEG < 1 historically outperforms.

Lynch (1989); Easton (2004); Fama & French (1992)
20%

Quality (ROE + Margins)

ROE + margins. Competitive moats and earnings durability.

Novy-Marx (2013); Asness, Frazzini & Pedersen (2019)
10%

Market Base Rate

Bayesian prior anchored to S&P ~10%/yr. Prevents excess bearishness.

Dimson, Marsh & Staunton (2002)

Merkapital Simulation Engine™

2,000 GBM price paths per stock over one year. Drift from five-factor score; volatility from Forward P/E.

Full probability distribution — not a point estimate. P(positive return), median outcome, 10th and 90th percentiles.

S(T) = S(0) × exp((μ − σ²/2) × T + σ × √T × Z)
where Z ~ N(0,1), T = 1 year
What You Get Per Stock
Composite Score
0 – 100
Weighted fundamental quality across all five factors
P(Up 1Y)
0% – 100%
Probability the price is higher in one year
Expected Return
Median outcome
50th percentile of 2,000 simulated paths
Downside (10th %ile)
Realistic bad case
Only 10% of simulations end worse than this
Upside (90th %ile)
Realistic good case
Only 10% of simulations end better than this

Optimal Portfolio Construction

Ranked universe → top 10 by risk-adjusted return and downside protection.

Ranking Criteria

Sharpe Proxy (Return / Risk)30%
Probability of Positive Return25%
Downside Protection20%
Fundamental Quality Score15%
Data Completeness10%

Why Equal Weight?

1/N equal-weight outperforms mean-variance optimization in practice. Estimation errors in returns and covariance negate theoretical gains.

“Optimal Versus Naive Diversification: How Inefficient is the 1/N Portfolio Strategy?”
DeMiguel, Garlappi & Uppal (2009), Review of Financial Studies
10
Positions
10%
Each Weight

Portfolio-Level Validation

5,000 portfolio-level simulations. Independent draws per holding; equal-weighted outcome. Captures diversification and aggregate risk.

Expected Return
Mean of 5,000 paths
Portfolio Volatility
Diversification-reduced
Sharpe Ratio
Return / Volatility
P(Positive 1Y)
% of paths above zero

Research Foundations

Decades of asset pricing, factor forecasting, Bayesian estimation. Every design decision traceable to peer-reviewed literature (1952–2019).

Fama & French (1992)
The Cross-Section of Expected Stock Returns
Journal of Finance
Novy-Marx (2013)
The Gross Profitability Premium
Journal of Financial Economics
O'Shaughnessy (2011)
What Works on Wall Street
McGraw-Hill, 4th Edition
Brav & Lehavy (2003)
Analysts' Target Prices
Journal of Finance
DeMiguel et al. (2009)
Optimal vs. Naive Diversification
Review of Financial Studies
Markowitz (1952)
Portfolio Selection
Journal of Finance
Sharpe (1966)
Mutual Fund Performance
Journal of Business
Asness et al. (2019)
Quality Minus Junk
Review of Accounting Studies
Easton (2004)
PE Ratios, PEG Ratios
The Accounting Review
For Retail Investors

Clear signals, not noise.

Highest-probability stocks, downside risk, and factor rationale at a glance. Plain language. No black boxes.

For Institutional Investors

Methodology you can audit.

Factor decomposition, documented weights, Bayesian shrinkage, per-stock volatility, portfolio simulation. Transparent codebase.

The numbers you need — not opinions you don't.

From data to decision in minutes.

Join other thoughtful investors. Start with a free account — see top stocks and probabilities today.