Built to survive. Not to impress.
Most signal services show you their best trades. We show you our methodology, our failure rate, and why a 70% win rate still means you'll lose 3 times out of 10. That transparency is the product.
Post-Earnings Drift
Institutional money underreacts to earnings beats.
Win Rate
66.7%
Avg Return
+6.42%
Hold Period
20d
Quality Momentum Dip
Strong uptrends with temporary pullbacks.
Win Rate
59.1%
Avg Return
+3.39%
Hold Period
60d
Deep Value Reversion
FCF-rich stocks below intrinsic value. Regime-gated.
Win Rate
65.2%
Avg Return
+15.44%
Hold Period
90d
Seasonal Momentum Dip
Sep/Oct entry into the year's strongest stocks.
Win Rate
69.2%
Avg Return
+5.11%
Hold Period
63d
Conditional Strategies — Accumulating Data
Earnings Over-Reaction Reversal
Stocks that gap up too far on earnings tend to mean-revert. We short the over-reaction.
Short Position Strategy
Takeout signals require selling short — the position profits when the stock price falls. Short selling requires a margin account, carries different risk characteristics than long positions, and is not appropriate for all investors.
68%
Win Rate
153 trades · 5 years
+2.06%
Avg Return Per Trade
4-day hold
98%
CPCV Out-of-Sample
Cross-validated
Validation Evidence
- ✓p = 0.0046 — statistically significant
- ✓98% positive out-of-sample (CPCV)
- ✓Median OOS Sharpe: 2.203
- ✓Direction confirmed
- ✓Gap ranking confirmed
Current Limitations
- ○43 months data — needs 62+ for full certification
- ○White's Reality Check incomplete (data volume)
- ○Full validation target: Q3 2027
- ○Deploy with appropriate position sizing
Takeout is published with full transparency about its validation status. The statistical evidence supports deployment — p=0.0046 with 98% out-of-sample consistency is meaningful. The limitation is data volume, not strategy logic. Signals available to Edge subscribers who understand the conditional framing.
Short selling involves substantial risk. Past results do not guarantee future returns.
Walk-Forward Folds Passed
Including 2022 — when the S&P fell 20%. Fresh Catch returned +10.76% with an 84.8% win rate in a bear market. The edge doesn't require a bull market.
Validated Signals
Signals generated across 5 full years of out-of-sample market conditions. Each one evaluated with a 28-day purge gap to prevent look-ahead bias.
Systems Killed
Hypotheses tested and rejected before they could reach subscribers. Every active sleeve survived this process. Most didn't.
What a 70% Win Rate Actually Means
In a long enough sequence of Fresh Catch signals, roughly 3 out of every 10 will lose money. That's not a flaw in the system — it's what a 70% win rate means.
The edge isn't in any single signal. It's in the aggregate — a positive expected value across hundreds of trades. One signal disappointing tells you nothing. Thirty signals over six months tells you everything.
A coin that lands heads 70% of the time still lands tails. You have to flip it enough times for the edge to materialize.
Why Zero Signals Is a Feature
Some days the scanner finds no signals that meet our threshold. We publish that result. No signals means no edge detected today — and manufacturing a trade to fill the dashboard would be the wrong call.
This is what separates a systematic research platform from a signal service that feels obligated to publish something every day. Restraint is part of the edge.
These signals work best as a system
AppleJax is designed for portfolio-level deployment — not one-off trades. Our backtests assume diversified exposure across multiple signals and sleeves simultaneously. Concentrating in a single signal amplifies both wins and losses in ways that make the underlying edge statistically irrelevant.
Diversify Across Signals
5–10 simultaneous positions lets the win rate work in your favor. Single signals are noise. The portfolio is signal.
Size Consistently
Equal or risk-adjusted sizing across signals prevents one large winner (or loser) from dominating your results. The edge is in the average, not the outlier.
Follow the Hold Period
Each sleeve has a validated hold period backed by backtesting. Exiting early or holding past the window removes you from the edge distribution entirely.
A Note on How This Works
AppleJax publishes historical signal research. Every signal here reflects a pattern that has repeated across 21 years of market data. What we do not publish is the portfolio strategy behind it. The strongest way to trade these signals has been tested extensively and involves specific position sizing, sleeve weighting, and capital allocation rules that we keep private.
Because that strategy is not shared here, and because we do not know your financial situation, AppleJax cannot and does not give financial advice. What we give you is the research. What you do with it is your call.
We encourage every subscriber to do their own research on any stock that shows up in this system. These are historical statistical observations only. They are not recommendations to buy, sell, or hold any security. Past results do not guarantee future performance. Never risk capital you cannot afford to lose.
A Note on Advice
AppleJax publishes systematic research outputs. We don't know your financial situation, your existing holdings, your tax position, or your risk tolerance. Our backtests show what the system produced historically — they don't predict your personal results.
We're confident in the research. We're not in a position to tell you how much to allocate, how to weight it against your other investments, or whether this fits your goals. That determination is yours to make — ideally with a financial advisor who understands systematic strategies.
What we can tell you: we test everything, we publish what survives, and we show our work.