Founding lifetime access: $100, $250, or $500

Then $100/month · first 1,000 members · 642 spots left

Find the strategies that held up before risking capital.

Leroy turns an 800GB-class market archive into ranked strategies, model evidence, and trade-by-trade proof you can actually inspect.

Ask a what-if market question.

What happened when you tradedDonchian Breakout 20onARMthe last 30 sessions

Result

Return

Trades

ModelSPYVTIMay 18 - Jun 30 Eastern

Historical research only. Not investment advice, not a promise of future returns, and not a broker order.

Get Access See Today's Model

How the lab works

A shorter path from idea to evidence.

The product is not one magic chart. It is an evidence ladder: replay the idea, test combinations, add market context, then let the model decide what deserves attention.

Replay the rule

First, answer the clean question: what would this setup have done on this ticker over this exact window?

Add context

Then compare the same signal against market regime, volume, volatility, and benchmark movement.

Let the model filter

Finally, ask whether the adaptive model would have taken the trade or blocked it.

Single strategy replays

1 / 4

Start with one clean trading idea.

Before combining anything, the lab asks the simplest possible question: if this one rule traded this ticker over this period, what happened?

Evidence ladder

Step 01

Question

Should VWAP Reversion matter here?

Idea

One clean rule

Replay

Historical result

Context

Market conditions

Decision

Model ranking

Inputs

VWAP ReversionBollinger BandsDonchian BreakoutOpening RangeRSI Reversion

Checks

Output

Return, drawdown, win rate, trade count

Evidence

Trade-by-trade modeled ledger

Question

Did the rule work on its own?

Answer

Start with one clean trading idea.

Before combining anything, the lab asks the simplest possible question: if this one rule traded this ticker over this period, what happened?

Output

Return, drawdown, win rate, trade count

Evidence

Trade-by-trade modeled ledger

Question

Did the rule work on its own?

Examples under test

VWAP Reversion

Bollinger Bands

Donchian Breakout

Opening Range

RSI Reversion

Keltner Squeeze

First-Hour Momentum

Volatility Breakout

MACD Trend

Relative Strength

What gets proven

OutputReturn, drawdown, win rate, trade count

EvidenceTrade-by-trade modeled ledger

QuestionDid the rule work on its own?

Strategy combinations

2 / 4

Then test whether agreement actually helps.

Two signals firing together can mean stronger evidence. It can also mean duplicate exposure. The lab measures the difference instead of assuming confluence is good.

Evidence ladder

Step 02

Question

Should VWAP + RSI matter here?

Idea

One clean rule

Replay

Historical result

Context

Market conditions

Decision

Model ranking

Inputs

VWAP + RSIBollinger + MACDOpening Range + MomentumDonchian + VolumeKeltner + ATR

Checks

Output

Pair and basket-level results

Evidence

Same-minute signal overlap

Question

Did the blend beat either rule alone?

Answer

Then test whether agreement actually helps.

Two signals firing together can mean stronger evidence. It can also mean duplicate exposure. The lab measures the difference instead of assuming confluence is good.

Output

Pair and basket-level results

Evidence

Same-minute signal overlap

Question

Did the blend beat either rule alone?

Examples under test

VWAP + RSI

Bollinger + MACD

Opening Range + Momentum

Donchian + Volume

Keltner + ATR

Relative Strength + Trend

Reversion + Liquidity

Breakout + Volume Expansion

What gets proven

OutputPair and basket-level results

EvidenceSame-minute signal overlap

QuestionDid the blend beat either rule alone?

Market context layer

3 / 4

Next, layer in the conditions around the signal.

The same setup can mean very different things depending on the market around it. This is where the system starts separating a bounce from a falling knife.

Evidence ladder

Step 03

Question

Should Time Of Day matter here?

Idea

One clean rule

Replay

Historical result

Context

Market conditions

Decision

Model ranking

Inputs

Time Of DaySPY / QQQ TrendVTI BaselineVolume RegimeATR / Volatility

Checks

Output

Context-matched cohorts

Evidence

Similar historical moments

Question

When does this setup deserve trust?

Answer

Next, layer in the conditions around the signal.

The same setup can mean very different things depending on the market around it. This is where the system starts separating a bounce from a falling knife.

Output

Context-matched cohorts

Evidence

Similar historical moments

Question

When does this setup deserve trust?

Examples under test

Time Of Day

SPY / QQQ Trend

VTI Baseline

Volume Regime

ATR / Volatility

VWAP Distance

Sector Behavior

Spread / Liquidity

Earnings Windows

Macro Windows

What gets proven

OutputContext-matched cohorts

EvidenceSimilar historical moments

QuestionWhen does this setup deserve trust?

Adaptive model routing

4 / 4

Now the model decides what deserves attention.

The model is fed the rule, the blend, and the context. It scores the setup, ranks comparable opportunities, and can block trades that look weak against historical evidence.

Evidence ladder

Step 04

Question

Should Signal Score matter here?

Idea

One clean rule

Replay

Historical result

Context

Market conditions

Decision

Model ranking

Inputs

Signal ScoreValidation CohortExpected MarkoutModel ConfidenceRejection Tracking

Checks

Output

Approved, blocked, and shadow results

Evidence

Out-of-sample model checks

Question

Should this setup be traded now?

Answer

Now the model decides what deserves attention.

The model is fed the rule, the blend, and the context. It scores the setup, ranks comparable opportunities, and can block trades that look weak against historical evidence.

Output

Approved, blocked, and shadow results

Evidence

Out-of-sample model checks

Question

Should this setup be traded now?

Examples under test

Signal Score

Validation Cohort

Expected Markout

Model Confidence

Rejection Tracking

Protectiveness Threshold

Shadow Ledger

Paper Routing Priority

What gets proven

OutputApproved, blocked, and shadow results

EvidenceOut-of-sample model checks

QuestionShould this setup be traded now?

Live Paper Model

So what happened when the model made the decision today?

This section moves from historical replay to the live paper model: approved trades, rejected candidates, benchmarks, and the protectiveness slider.

LiveUpdated 11:15:30 AM Eastern

How’s the model doing today?

Approved paper fills are −0.08% today. At this strictness, 105 blocked candidates would have fired and marked +2.68% after 15 minutes.

Model Protectiveness

Lower = more signals. Higher = more protection.

35%

Would Fire

105

Wins / Losses

78/27

15m Mark

+$2,678.97

Approved Model

−0.08%

−$81.67

Rejected Set

+2.68%

105 Would Fire

SPY Today

−0.34%

$742.17

VTI Today

−0.48%

Benchmark Proxy

ModelRejected SetSPYVTI9:30 AM–11:14 AM Eastern

Latest model actions

Recent paper fills behind the model line.

No paper model fills have been recorded in the current visible session yet.

Research Center

The value is not one magic chart. It is the evidence system behind it.

The paid research center is where the archive, simulations, model comparisons, and proof ledgers live. The public page shows the shape; subscribers see the exact names, scores, paths, and ledgers.

Market Archive

800GB-Class

Cleaned, stored, indexed, and turned into repeatable simulations.

Minute Data

Since 2005

Historical one-minute stock bars powering repeatable simulations.

Strategy Types

Rules To Models

Single signals, blended signals, context filters, and adaptive models compared side by side.

Trade Ledgers

Line By Line

Every headline result can be checked against the modeled buys and sells behind it.

Lifetime Simulation View

Here's how a $100k research account is doing.

Ranked results, benchmark context, and modeled account value stay tied to the trade ledgers behind each simulation.

Class	Public Model	Dates	Observations	Starting Capital	Modeled P/L	Return	Ending Value	Evidence
Latest Model	Latest Model	Today						Live Paper Snapshot

These are research simulations and paper snapshots, not live broker returns, investment advice, or future-return promises.

Research Access

Start with the leaderboard. Audit with the ledger.

Claim founding lifetime access starting at $100, or subscribe later for full access at $100/month.

Subscribe Sign In