BACKTESTED & VERIFIED  |  22 YEARS  |  106 EVENTS

Buy the Rumor, Sell the News.

A backtested playbook for AAPL, NVDA & GOOG.
The money is made before the event. Not during. Not after.

0%
WWDC Win Rate
0%
WWDC Total Return
0%
NVDA GTC Win Rate
Explore the Edge
Signal Detected

The Core Insight

Apple, NVIDIA, and Google follow predictable annual event cycles — product launches, developer conferences, quarterly earnings — that create recurring windows where the stock reliably moves in one direction.

AAPL

Jan 2003 – Feb 2026

5,821

trading days · 106 events

NVDA

Jan 2014 – Feb 2026

39

events (GTC + earnings)

GOOG

Jan 2014 – Feb 2026

48

events (earnings, I/O, Pixel)

Exit before the event. Every time. No exceptions.

Methodology: Returns are simple percentage change (entry to exit). Alpha = strategy return minus SPY over the same window. Win rate = trades with positive return / total. Total return = compounded across all trades. Sharpe = annualized excess return / std dev. Max drawdown = largest peak-to-trough in cumulative equity.

The Proof

Strategy Scorecard

Nine strategies tested. Eight profitable. One anti-pattern to avoid.

# Strategy Ticker Trades Win Rate Avg Return Alpha Sharpe Total Return
Win Rate & Avg Return by Strategy (Dual Axis)
Return vs Alpha — All Trades
Accumulation

Pre-WWDC Accumulation

A · 85% Confidence

WWDC is Apple's annual developer conference. In the weeks before the keynote, developer and media hype builds anticipation. Extensive leaks mean the keynote itself rarely surprises. Between 2006 and 2017, AAPL had a 12-year consecutive losing streak on keynote day. The trade: buy the hype, sell before the event.

T-20 ENTER T-15T-10T-5 T-1 EXIT EVENT
Rules
EntryT-20 trading days before WWDC
ExitT-1 (day before keynote)
Holding~20 trading days (~4 weeks)
Stop loss1.5x ATR trailing
Results (2003–2025)
Total trades23
Win rate78.3%
Avg return+4.75%
Alpha vs SPY+3.74%
Best / Worst+24.18% / -17.00%
Sharpe2.39
Max drawdown-18.66%
Total return+169.28%
Without Stop
Win rate78.3%
Avg return+4.75%
Total return+169.28%
Max DD-18.66%
With 1.5x ATR Trailing
Win rate78.3%
Avg return+5.44%
Total return+221.12%
Max DD-8.94%
Stopped out17%
Equity Curve — Pre-WWDC (Compounded)
Return by Year
Year-by-Year Trade Data
YearWWDC DateEntryExitReturnSPYAlphaResult
Momentum

Pre-iPhone Run-Up

A- · 75% Confidence

The iPhone announcement is the most anticipated consumer electronics event on the calendar. The key finding: in the 2018+ era, exiting a full week before the event (T-7) produces a perfect 100% win rate over 8 trades.

T-30 ENTERT-25T-15T-7 T-1 EXITEVENT
Rules
Entry (default)T-30
Exit (default)T-1
Optimized (2018+)T-25 / T-7 → 100% WR
Stop lossNo stop (or use optimized window)
Results (2007–2025)
Total trades16
Win rate68.8%
Avg return+3.78%
Alpha vs SPY+3.50%
Best / Worst+18.05% / -8.58%
Sharpe1.74
Total return+73.30%
Equity Curve — Pre-iPhone (Compounded)
Return by Year
Year-by-Year Trade Data
YearEvent DateEntryExitReturnSPYAlphaResult
Recovery

Sept Dip to Oct Recovery

A- · 80% Confidence

After the iPhone announcement, AAPL enters a predictable dip as sell-the-news pressure kicks in. The stock recovers through October driven by shipment reports, holiday anticipation, and buyback restoration. Highest average return of any AAPL strategy: +6.62% per trade.

iPHONE (SELL-OFF)+1w+2w+3w+4w ~OCT 31 EXIT
Rules
Entry~Sept 8–18 (post-iPhone dip)
ExitLate October (~Oct 31)
Holding~35–45 trading days
Stop loss2.0x ATR trailing (wider for long hold)
Results (2003–2025)
Total trades23
Win rate73.9%
Avg return+6.62%
Alpha vs SPY+5.88%
Best / Worst+44.15% / -31.87%
Sharpe1.95
Max drawdown-31.87%
Total return+248.31%
Without Stop
Win rate73.9%
Avg return+6.62%
Total return+248.31%
Max DD-31.87%
With 2.0x ATR Trailing
Win rate78.3%
Avg return+7.34%
Total return+341.19%
Max DD-10.17%
Stopped out35%
Equity Curve — Sept-Oct Recovery (Compounded)
Return by Year
Year-by-Year Trade Data
YearEntryExitReturnSPYAlphaResult
The Workhorse

Pre-Earnings Drift

B+ · 70% Confidence

AAPL drifts upward in the 14 trading days before earnings as analysts raise estimates, buyback blackouts end, and institutions position ahead of the report. This fires four times per year — the workhorse strategy providing steady compounding.

T-14 ENTERT-10T-7T-3 T-1 EXITEARNINGS
Rules
EntryT-14 trading days before earnings
ExitT-1 (day before earnings report)
Holding~14 trading days (~3 weeks)
Stop loss1.5x ATR trailing
Results (FY2015–FY2025)
Total trades44
Win rate65.9%
Avg return+1.88%
Alpha vs SPY+0.89%
Best / Worst+13.08% / -9.04%
Sharpe1.36
Max drawdown-15.55%
Total return+115.85%
Without Stop
Win rate65.9%
Avg return+1.88%
Total return+115.85%
Max DD-15.55%
With 1.5x ATR Trailing
Win rate70.5%
Avg return+2.51%
Total return+187.30%
Max DD-8.31%
Stopped out27%
Volume Filter Insight

High-volume entries hurt this strategy. Low-volume drift = healthy accumulation. High-volume = hedging/uncertainty.

42.9% WR
Volume confirmed (above avg) · -0.26% avg
87.0% WR
Low volume (below avg) · +3.83% avg
Equity Curve — Pre-Earnings (Compounded)
Return by Quarter
Year-by-Year Trade Data
YearEarnings DateEntryExitReturnSPYAlphaResult
Volatility

Pre-Earnings IV Capture

A · Live via Schwab

Implied volatility systematically rises 7–14 days before earnings. By entering options positions when IV is cheap and exiting when IV is elevated (T-1), you capture IV expansion without earnings gap risk. This is a volatility trade, not a directional trade.

Signal Logic
IV percentile< 30% → Entry signal
Days to earningsT-7 to T-21 window
VIX regimeContango + RISK_ON
C/P ratio> 1.0 (bullish flow)
ExitT-1 to T-3 or IV > 70th pctl
Vehicle Selection
Bullish (C/P > 1.5)Long ATM/OTM calls
Bearish (C/P < 0.8)Long ATM/OTM puts
NeutralLong ATM straddle
PlatformSchwab Dashboard (port 8000)
Pattern Extends

NVDA & GOOG

The event-driven framework extends beyond Apple. NVIDIA's GTC is the single highest-conviction trade in the entire system.

NVDA — The AI Hype Machine

Pre-GTC Run-Up (T-10 / T+3)
Trades11
Win rate90.9%
Avg return+12.65%
Alpha vs SPY+11.82%
Sharpe3.44
Total return+249.58%
Pre-Earnings Drift (T-14 / T-1)
Trades27
Win rate74.1%
Avg return+6.74%
Alpha vs SPY+4.98%
Sharpe2.64
Total return+389.04%

GOOG — Earnings Yes, Pixel No

Pre-Earnings (T-14 / T-1)
Trades28
Win rate75.0%
Avg return+2.86%
Sharpe2.33
Total return+112.40%
Pre-I/O (T-10 / T+3)
Trades10
Win rate70.0%
Avg return+2.61%
Sharpe1.39
AVOID Pre-Pixel (T-20 / T-1)
Trades10
Win rate30.0%
Avg return-2.52%
Sharpe-1.50
Total return-23.74%
Adaptation

The Pattern Is Evolving

All strategies show some degradation in the 2018+ era, consistent with patterns becoming more widely known and front-run. The pattern hasn't disappeared — it has shifted.

Win Rate: Early Era vs 2018+
Optimized Windows for the Current Regime

Parameter sweeps on the 2018+ regime reveal that shorter, later entry windows now perform better. Front-running has compressed the early portion of the trade.

StrategyDefaultOptimized2018+ WR2018+ Avg Return
Pre-WWDCT-20 / T-1T-10 / T+387.5%+4.17%
Pre-iPhoneT-30 / T-1T-25 / T-7100.0%+6.17%
Pre-EarningsT-14 / T-1T-14 / T+371.9%+2.94%
Protecting the Edge

What Can Go Wrong

Regime Shifts
A prolonged bear market or macro crisis can overwhelm seasonal patterns. The 2022 WWDC trade lost -7.57% during the Fed hiking cycle. Stop losses catch these, but size should already be reduced in risk-off environments.
👥
Pattern Crowding
As more participants trade these windows, the edge compresses. Degradation data shows this is already happening for WWDC and Sept-Oct. Optimized windows partially compensate, but the trade will continue to evolve.
🎯
Single-Stock Concentration
These are all single-name equity trades. A company-specific event can overwhelm any seasonal pattern. Diversifying across AAPL, NVDA, and GOOG partially mitigates this.
⚠️
Liquidity & Execution
Strategies assume fills at the close. In practice, slippage on entry and exit reduces net returns, especially during high-volatility windows.
Half-Kelly Position Sizing

Upper-bound position sizes based on historical win rates and win/loss ratios. Reduce further in risk-off regimes.

StrategyWin RateWin/Loss RatioHalf-Kelly$100K Account
Pre-WWDC78.3%1.2530.4%~$30K
Sept-Oct Recovery73.9%1.3527.3%~$27K
Pre-iPhone68.8%1.5024.0%~$24K
Pre-Earnings65.9%1.3119.9%~$20K
ATR Stop Loss Recommendations
StrategyConfigRationale
Pre-WWDC1.5x ATR trailingShort hold, tight control — cuts max DD in half
Pre-Earnings1.5x ATR trailingCatches macro shocks that overwhelm the gentle drift
Sept-Oct Recovery2.0x ATR trailingLonger hold needs wider stop — still cuts DD from -32% to -10%
Pre-iPhoneNo stop (use T-25/T-7)Stops reduce WR to 50% — the drawdowns recover
Next Cycle Begins

2026 Trading Calendar

All AAPL, NVDA, and GOOG events mapped with entry/exit dates and expected win rates. Automated signals are active.

Trade Windows — 2026
AAPL 2026
EntryEventExitStrategyExpected WR
Apr 10Q2 FY26 Earnings (Apr 30)May 5Pre-Earnings66%
May 25WWDC 2026 (~Jun 8)Jun 11Pre-WWDC (optimized)88%
Jul 10Q3 FY26 Earnings (Jul 30)Aug 4Pre-Earnings66%
Aug 4iPhone 18 (~Sep 8)Aug 28Pre-iPhone (optimized)100%
Sep 10Sept Dip (post-iPhone)Oct 30Sept-Oct Recovery74%
Oct 9Q4 FY26 Earnings (Oct 29)Nov 3Pre-Earnings66%
Jan 8Q1 FY27 Earnings (Jan 28)Feb 2Pre-Earnings66%
NVDA 2026
EntryEventExitStrategyExpected WR
Mar 2GTC 2026 (Mar 16)Mar 19Pre-GTC91%
May 7Q1 FY27 Earnings (May 27)May 26Pre-Earnings74%
Aug 6Q2 FY27 Earnings (Aug 26)Aug 25Pre-Earnings74%
Oct 29Q3 FY27 Earnings (Nov 18)Nov 17Pre-Earnings74%
GOOG 2026
EntryEventExitStrategyExpected WR
Apr 8Q1 Earnings (Apr 28)Apr 27Pre-Earnings75%
Apr 28Google I/O 2026 (~May 12)May 15Pre-I/O70%
Jul 8Q2 Earnings (Jul 28)Jul 27Pre-Earnings75%
Oct 7Q3 Earnings (Oct 27)Oct 26Pre-Earnings75%

Calendar overlap to watch: NVDA GTC (mid-March) and GOOG Q1 Earnings / I/O (late April–mid May) create a dense signal cluster. Size accordingly to avoid over-concentration in mega-cap tech.

Execution

Daily Workflow

1
Check Signals
Run python3 src/live_signals.py --report for current entry/exit signals across all events.
2
Check IV
If within 7–21 days of earnings, run python3 src/iv_capture.py for IV percentile assessment.
3
Size the Position
Use the Half-Kelly table as an upper bound. Reduce in risk-off regimes (VIX > 25, correlation spikes, macro events).
4
Set the Stop
Enter the ATR trailing stop at the recommended multiplier on day one. Adjust only upward (trail), never down.
5
Exit on Schedule
The exit date is predetermined. Do not override it. The entire edge of this strategy is in the discipline of the exit.
Conviction Hierarchy

When you have to choose which trades to take and which to skip:

  1. 1.
    NVDA Pre-GTC — 91% WR, +12.65% avg, 3.44 Sharpe. Take this every year.
  2. 2.
    AAPL Pre-WWDC (optimized) — 88% WR in 2018+. The flagship.
  3. 3.
    AAPL Pre-iPhone (optimized) — 100% WR in 2018+ with T-25/T-7. Perfect record.
  4. 4.
    NVDA Pre-Earnings — 74% WR, +6.74% avg. Strong and frequent.
  5. 5.
    AAPL Sept-Oct Recovery — 74% WR, highest avg return. Needs wider stop.
  6. 6.
    GOOG Pre-Earnings — 75% WR. Consistent compounder.
  7. 7.
    AAPL Pre-Earnings — 66% WR. Workhorse, smallest edge. Volume-filter helps.
  8. 8.
    GOOG Pre-I/O — 70% WR. Decent but small sample.
Never trade: GOOG Pre-Pixel (30% WR, -2.52% avg).