# 100 Complementary Day-Trading Ideas

Research ideation (not backtested). Designed as **possible 4th+ sleeves** beside the current book:

| # | Token | Principle |
|---|--------|-----------|
| 1 | `ma_slope_intraday` | Cross-sectional intraday momentum → MOC |
| 2 | `pdl_touch_short` | Fade prior-day-low support (high-vol names) |
| 3 | `atr_high_breakout` | Early ATR-anchored breakout long (mega5) |

**Complementarity rule:** different clock, side, level, universe, or regime vs the three — not thin re-parameterizations of the same trade.

**Shared constraints (default):** 5m RTH, flat MOC, 3 bps/side slippage, liquid names; shorts need locate.

---

## A · Cross-sectional momentum / slope (complements #1)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 01 | Bottom-10 slope **short** (confirm 4b) | Short | Exact inverse of #1; hedge book beta |
| 02 | Sector-neutral top-N (sector caps) | Long | Same clock, lower SPY β |
| 03 | Score-proportional sizing (not equal) | Long | Tilt wins without new signals |
| 04 | Re-rank mid-day (12:30) with ½ size | Long | Second pulse after #1 entry |
| 05 | EMA20/100 dual product (slower clock) | Long | Catches later trends #1 misses |
| 06 | Price ROC 30-bar top-N (no MA) | Long | Independent momentum metric |
| 07 | Relative strength vs SPY (α ranking) | Long | Orthogonal to absolute slope |
| 08 | Slope **acceleration** (Δslope) top-N | Long | Inflection, not level |
| 09 | Skip names already in #1 book | Long | Pure satellite sleeve |
| 10 | Afternoon fade of #1 winners (−ATR) | Short | Mean-revert the momentum sleeve |
| 11 | Orphan continuity: hold overnight if slope+ | Long | Overnight pulse (breaks MOC rule — size tiny) |
| 12 | Dual timeframe: daily slope gate × intraday top-N | Long | Regime filter for #1 universe |
| 13 | Volume-confirmed slope (vol ≥ 1.2× ses avg) | Long | Quality overlay |
| 14 | Low-float / high-ADV subset only | Long | Different tape |
| 15 | Top-N underperformers after gap-up | Short | Fade gap + weak slope |

## B · Prior-session & day-range extremes (complements #2)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 16 | Prior-day-**high** touch **long** (failed resistance → support? or fade) | Long/Short | Mirror of PDL; test both |
| 17 | PDH touch **short** (fade resistance) | Short | Classic R1 fade; opposite PDL geography |
| 18 | Prior-week high/low touch fades | Short/Long | Weekly levels ≠ daily PDL |
| 19 | Opening range high/low fade (OR 30m) | Short/Long | OR ≠ prior day |
| 20 | Prior-day mid (50%) reclaim short | Short | Different anchor |
| 21 | Gap-fill to prior close (fade into fill) | Short/Long | Gap dynamics, morning-only |
| 22 | Overnight high/low from Globex proxy (futures) | Short | Extends #2 to 24h structure |
| 23 | Multi-day swing low (3d) touch short | Short | Slower structure than PDL |
| 24 | Pivot Point R1/S1 fade (classic floor) | Short/Long | Separate level family |
| 25 | Previous session VWAP touch short | Short | Session-liquidity level |
| 26 | IB (initial balance) extension fade | Short | Floor-trader framework |
| 27 | RTH vs prior RTH range mid fade | Short | Range theory |
| 28 | Touch PDL **only if above VWAP** (quality) | Short | Overlay on #2, not new; still useful |
| 29 | PDH break **then** reclaim short (false break) | Short | Failed-break motif on highs |
| 30 | Cluster levels: PDL + weekly low confluence | Short | Higher-conviction subset of #2 |
| 31 | First touch only vs third-touch fade | Short | Sequence dependency |
| 32 | Distant PDL (−2×ATR below open) only | Short | Extreme washout shorts |

## C · Breakouts / expansion (complements #3)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 33 | ATR **below** open/POI breakout **short** | Short | Downside mirror of #3 |
| 34 | Opening-range breakout long (OR×1, early) | Long | OR anchor ≠ POI/ATR stop |
| 35 | N-bar high breakout (20-bar) MOC | Long | Donchian-style vs ATR stop |
| 36 | Squeeze → break (BBW / NR7 prior → expand) | Long | Compression motif (failed on mega5 alone — try SP100) |
| 37 | Relative breakout: stock > open while SPY < VWAP | Long | Idiosyncratic strength |
| 38 | Break yesterday’s high after 10:00 | Long | Classic PDH break (vs ATR) |
| 39 | VWAP +1σ band break long | Long | Volume-weighted structure |
| 40 | Gap-and-go: gap >0.5% + hold OR high | Long | Gap regime #3 ignores |
| 41 | Pullback-to-breakout (break, retest, go) | Long | Delayed entry vs first cross |
| 42 | Mega5 breakout **short** failed-break (close < lvl) | Short | We rejected this exit; try as entry signal |
| 43 | Breadth thrust: >60% names > open → index long | Long | Market-level, not stock |
| 44 | Break of AM high after lunch (late range) | Long | Opposite of “early only” (adversarial test) |
| 45 | Cumulative Δ ATR trail stop (Chandelier day) | Long | Exit upgrade for #3 winners |
| 46 | Risk-on basket breakout (equal QQQ/XLK) | Long | Sector ETF, not single name |

## D · Time-of-day / session structure

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 47 | First 15-min trap: fade open drive after stall | Short/Long | Pure clock trade |
| 48 | Lunch reverse (11:45–12:30) vs AM trend | Reverse | Temporal diversifier |
| 49 | Power-hour trend alignment with SPY (15:00+) | Long | Late liquidity #1/#3 miss |
| 50 | Close auction imbalance proxy (last 10m) | Long/Short | Microstructure |
| 51 | Monday / FOMC / OpEx day ruleset | Conditional | Calendar diversifier |
| 52 | Skip first 2 bars always (9:30–9:40) | Overlay | Noise filter for all sleeves |
| 53 | 10:00–10:30 only for shorts; 10:40+ longs | Scheduling | Orthogonal allocation by clock |

## E · VWAP / liquidity (learned: reclaim **long** failed)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 54 | VWAP **fail** short (stretch up → reject VWAP) | Short | Flip of failed long reclaim |
| 55 | SPY below VWAP → only trade #2 shorts | Overlay | Regime gate, low ρ |
| 56 | Anchored VWAP from prior earnings | Long/Short | Event anchoring |
| 57 | Twap deviation fade (±k×ATR vs TWAP) | Fade | Institutional clock |
| 58 | Volume shelf breakdown short | Short | Horizontal liquidity |
| 59 | Relative volume surge + down close short | Short | Climax continuation |
| 60 | Dark-pool / print cluster proxy (large prints) | Long | Info trade (hard data) |

## F · Gap / overnight / news

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 61 | Fade large overnight gap (>1%) toward prior close | Fade | Classic gap-fill |
| 62 | Gap-and-crap: gap up + first 30m red → short | Short | Failed gap continuation |
| 63 | Earnings long-straddle **day-after** MOC | Neutral | Event vol (options) |
| 64 | Index rebalance day (add/deletes) | Long/Short | Mechanical flow |
| 65 | Pre-CPI / post-CPI range break | Long/Short | Macro calendar |

## G · Relative / pairs / market-neutral

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 66 | Long top slope / short bottom slope (dollar-neutral) | L/S | Neutralizes #1 beta |
| 67 | Stock vs sector ETF residual mean-revert | Fade | Idiosyncratic |
| 68 | Pair: mega-cap relative (AAPL/MSFT z-score) | L/S | Name-relative |
| 69 | Long PDL short winners / short PDH fade losers | L/S | Combines B motifs neutrally |
| 70 | Rank by (ret − β×SPY) mid-morning, long residual+ | Long | Pure α selection |

## H · Volatility / VIX / regime overlays

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 71 | VIX >25 → only shorts (#2 style) | Overlay | Crisis mode |
| 72 | VIX <15 → only breakouts (#3 style) | Overlay | Trend mode |
| 73 | VVIX/VIX term steep → fade extremes | Overlay | Vol-of-vol |
| 74 | Realized > implied (VRP+) → avoid short-vol-like fades | Overlay | Skew to mean-reversion risk |
| 75 | ATR%-rank filter dynamic (rolling percentile) | Overlay | Better than static 25%/2% |
| 76 | Dispersion high → prefer #1; low → prefer #2 | Overlay | Cross-sleeve allocator |

## I · ETF / index / sector day-trades

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 77 | QQQ PDL touch short (ETF-level #2) | Short | One ticker, locatable |
| 78 | Sector ETF rotation: top 2 / bottom 2 MOC | L/S | #1 on 11 names |
| 79 | TLT or HYG mean-revert day trade | Fade | Macro, low equity ρ |
| 80 | UVXY / VXX fade spike (intraday) | Fade | Vol product (careful path) |
| 81 | IWM vs SPY relative breakout | Long | Size factor day |
| 82 | Equal-weight RSP vs SPY spread fade | Fade | Concentration |
| 83 | Single-name only if ETF of sector not broken | Overlay | Hierarchy filter |

## J · Microstructure / order-flow proxies (5m)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 84 | Up/down volume imbalance flip | Long/Short | Tape proxy |
| 85 | Range expansion without volume → fade | Fade | False break filter |
| 86 | Stop-run: pierce OR low + close back inside → long | Long | Liquidity grab |
| 87 | Iceberg / repeated print size proxy | Follow | Hard; keep optional |
| 88 | Spread-widen → halt new entries | Overlay | Cost control |

## K · Options-aware / Greek day overlays (same equity book)

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 89 | Avoid shorting names with put/call vol crush day | Overlay | Skew risk |
| 90 | Pin risk: fade into weekly strike magnet PM | Fade | OPEX microstructure |
| 91 | Hedged: stock short + cheap OTM call (PDL sleeve) | Short+ | Softens #2 left tail |

## L · Explicit book combinations / meta

| ID | Idea | Side | Why complementary |
|----|------|------|-------------------|
| 92 | Only trade #2 when #1 book net flat/weak | Conditional | Anti-correlation allocation |
| 93 | Size #3 by overnight VIX spike | Overlay | Vol-scaled breakout |
| 94 | Breadth: #1 only if AD line rising | Overlay | Market health |
| 95 | Correlation spike days: cut #1/#3, keep #2 | Overlay | Regime switch |
| 96 | Max 1 overlapping name across sleeves | Overlay | Capacity / diversify |
| 97 | Fixed-$ notional reporting per sleeve (research hygiene) | Meta | Honest metrics |
| 98 | “Loser names” banned list (rolling 60d avg trade <0) | Overlay | Adaptive universe |
| 99 | Cash sleeve: 0 risk when SPY 5m RV extreme | Cash | Drawdown brake |
| 100 | Synthetic 4th sleeve: equal mix of 66+17+54 (neutral+PDH+VWAP fail) | Mixed | Instant diversifier basket |

---

## Suggested test order (high complementarity, feasible on current data)

1. **01** Bottom-10 slope short (confirm 4b)  
2. **17** PDH touch short  
3. **33** ATR downside breakout short  
4. **34** Opening-range breakout long  
5. **54** VWAP fail short  
6. **61** Gap-fill fade  
7. **66** Dollar-neutral slope L/S  
8. **75** Rolling ATR%-percentile filter on #2  
9. **77** QQQ PDL short  
10. **86** Liquidity-grab long (OR pierce reclaim)

Reproduce on existing stack: Alpaca 5m RTH panels, mega5 / SP100 / slope universe, same metrics (Sharpe, avg trade, yearly). Prefer **avg trade** over compounded % for high-frequency shorts.

---

## Anti-goals (already tested or likely redundant)

- Intraday CM dip long / VWAP reclaim **long**  
- PDL touch **long**  
- PDL bounce short (wait for reclaim)  
- SP100 ATR upside breakout without selection  
- Thin re-tunes: slightly different MA periods without new economics  

---

## Backtest status (2022-01-03 → 2025-12-31)

Runner: `run_day_trading_100_complement.py`  
Artifacts: `RenTech/data/logs/day_trading_100_complement_{results,meta,metrics}.*`

| Status | Count |
|--------|------:|
| ok | 85 |
| skip (no data / options / calendars) | 15 |
| error | 0 |

**Best novel sleeves (not just PDL/ATR rehashes):** QQQ PDL short (**77**), Donchian20 breakout (**35**), PDH false-break short (**29**), PDL×SPY&lt;VWAP (**55**).  
**Slope XS variants 01–15 / 66 / 70:** all negative on SP100-50 in this window (book #1 still uses ~500-name panel).  
**Trust avg trade + Sharpe**; compounded % on high-churn shorts is inflated.

```bash
cd /Users/robzingale/trading_bot && PYTHONUNBUFFERED=1 .venv/bin/python \
  RenTech/strategy_stack/run_day_trading_100_complement.py \
  --start 2022-01-03 --end 2025-12-31 --max-sp100 60 \
  --out-prefix RenTech/data/logs/day_trading_100_complement
```