Scanner

Deploy

Decision-support for capital allocation. Surfaces current high-confluence signals and applies risk-managed position sizing. Not financial advice.

Phase A backtest result: framework does NOT beat SPY

Tested 6 strategy variants end-to-end on 2022-2025 data. All failed the validation gate. Best variant (trend-following): matched SPY within 1.88 percentage points but did not beat it. See backtest page for full results.

  • If your goal is to beat the market: this system can't do it. Buy SPY/VOO directly.
  • If your goal is to find names worth deeper research: the picks below are legitimate research candidates. Add fundamental thesis, sector context, your own judgement.
  • Per-event signal edge exists at the individual signal level (Bonferroni-passing RSI divergence, t=7.49). But translating that into a portfolio strategy hasn't worked.
  • Position-sizing framework below is risk-managed default — equal-weight, sector caps, cash reserve. Use as scaffolding for your own decisions.

This page is research support. It is not a trading signal. It is not investment advice.

Capital
$10,000
Reference deployment size
Active signals
10
Tickers with at least one BUY or WATCH across the 3 modes
Picks selected
5
Top high-confluence names below capital allocation
Suggested allocation (equal-weight, 30% cash reserve, 20% per-name cap)
Equal weight across top BUY-rated names by mode confluence, capped at 20% per position, with 30% held in cash as reserve. This is a sensible default — not optimized, but hard to fault.
NEM
Newmont
Single-mode BUY + 1 WATCH confirmation · includes Oversold (best in-sample, failed OOS — treat with caution)
$1400
14.0%
INTU
Intuit
3-mode WATCH (no BUY) · includes Oversold (best in-sample, failed OOS — treat with caution)
$1400
14.0%
FICO
Fair Isaac
2-mode WATCH (no BUY) · includes Oversold (best in-sample, failed OOS — treat with caution)
$1400
14.0%
KLAC
KLA Corp
2-mode WATCH (no BUY) · includes Oversold (best in-sample, failed OOS — treat with caution)
$1400
14.0%
ADP
Automatic Data Processing
2-mode WATCH (no BUY) · Quality/Acceleration only
$1400
14.0%
CASH
Reserve
Cash reserve — wait for stronger signals OR drawdown opportunity
$3000
30.0%
Top candidates by mode confluence
Multi-mode agreement is the most empirically interesting signal we can construct from current data — independent modes saying the same thing.
TickerOversoldAccelerationQualityμ Score
NEMWATCH 72BUY 75—73.4
INTUWATCH 68WATCH 62WATCH 5862.8
FICOWATCH 65WATCH 73—69.1
KLACWATCH 70WATCH 64—67.1
ADP—WATCH 55WATCH 6358.8
LLY—WATCH 74—73.8
PAYX——WATCH 6867.7
GEN—WATCH 63—62.8
DXCM—WATCH 61—60.8
NVDA——WATCH 6059.7
Risk framework
Stop loss: -15% per position. If a name drops 15% from entry, sell — the thesis was wrong.
Time horizon: Minimum 3 months, ideally 6 months. The 63d/126d windows are where any (modest) excess return emerged in backtest. Shorter horizons = noise.
Per-name cap:20% of total capital. Even if a name looks "obvious," idiosyncratic risk is large.
Cash reserve: 30% minimum. Lets you add to winners on pullbacks OR redeploy if better signals appear.
Re-evaluate cadence:Check positions weekly. If a stock's mode score drops below 50, consider exit.
Realistic expected outcome:-20% to +20% over 6 months. The system's small validated edge (one signal, 57% hit rate) suggests modestly above breakeven. Anyone selling certainty is selling fiction.
What this page does NOT do
  • Does not execute trades — copy ticker symbols to your broker manually
  • Does not predict returns — the framework has tiny historical excess returns and unproven mode composites
  • Does not account for taxes, transaction costs (~0.1-0.2% per trade), or your personal financial situation
  • Does not constitute personalized financial advice — consult a fiduciary if you need that