mrfluffy/vibe-invest
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

it be better

Browse Source
This commit is contained in:
zastian-dev
2026-02-13 13:12:22 +00:00
parent 80a8e7c346
commit 1ef03999b7
6 changed files with 250 additions and 223 deletions
Download Patch File Download Diff File Expand all files Collapse all files

323
.claude/agent-memory/consistency-auditor/MEMORY.md
View File

@@ -1,33 +1,49 @@
# Consistency Auditor Memory
## Last Audit: 2026-02-12 (PDT Protection)
## Last Audit: 2026-02-13 (Post Config Update)
### AUDIT RESULT: ⚠️ 1 CRITICAL BUG FOUND
### AUDIT RESULT: ⚠️ 1 CRITICAL BUG + 1 CRITICAL BEHAVIORAL DIVERGENCE
**PDT (Pattern Day Trading) protection data type mismatch:**
- bot.rs uses `Vec<String>` for day_trades (line 56)
- backtester.rs uses `Vec<NaiveDate>` for day_trades (line 42)
- **Impact**: String parsing on every PDT check, silent failures on malformed dates, performance degradation
- **Fix required**: Change bot.rs to use `Vec<NaiveDate>` internally (see detailed fix below)
**1. CRITICAL BUG: Drawdown Peak Reset Inconsistency**
- **backtester.rs** resets `peak_portfolio_value` to current value on halt expiry (line 132)
- **bot.rs** does NOT reset peak on halt expiry (lines 365-377)
- **Impact**: Bot will re-trigger drawdown halt more frequently than backtest suggests, spending more time in cash. After a 15% drawdown triggers a 10-bar halt, a partial recovery followed by a minor dip will immediately re-trigger the halt in live trading but not in backtest.
- **Fix**: Add `self.peak_portfolio_value = portfolio_value;` to bot.rs after line 374 (inside the halt expiry block)
- **Code location**: `/home/work/Documents/rust/invest-bot/src/bot.rs:365-377`
**2. CRITICAL DIVERGENCE: PDT Enforcement Differs by Timeframe**
- **bot.rs** enforces PDT blocking on non-stop-loss sells (lines 619-628), with $25K exemption (lines 281-285)
- **backtester.rs** has PDT DISABLED entirely for backtest (lines 245-248: "informational only, not blocking")
- **Impact by timeframe**:
- **Daily mode**: No impact (buys in Phase 2, sells in Phase 1 on different bars → day trades impossible by design)
- **Hourly mode**: Potential divergence IF portfolio < $25K. Bot may skip exits to avoid PDT violation, holding positions overnight that backtest would have exited same-day. Backtester relies on late-day entry prevention (line 158-166) instead of exit blocking.
- **Portfolios >= $25K**: No divergence (PDT rule doesn't apply)
- **Verdict**: Acceptable for daily mode. Document for hourly mode. If hourly is primary deployment, verify backtest PDT day-trade count matches late-day entry prevention expectations.
---
## Previous Audit: 2026-02-12 (Regime-Adaptive Dual Strategy Update)
## Config Changes Since Last Audit (2026-02-13)
The refactor to extract shared logic into `strategy.rs` has **eliminated all previous consistency issues**. Bot and backtester now share identical implementations for all critical trading logic.
User reported these config changes:
- Drawdown halt: 12% → 15% (`MAX_DRAWDOWN_HALT`)
- Drawdown cooldown: 20 bars → 10 bars (`DRAWDOWN_HALT_BARS`)
- Momentum pool: 10 stocks → 20 stocks (`TOP_MOMENTUM_COUNT`)
- Buy threshold: 4.5 → 4.0 (in signal generation)
**Verified**: All constants consistent between bot.rs and backtester.rs ✅
---
## VERIFIED CONSISTENT (2026-02-12)
## VERIFIED CONSISTENT (2026-02-13 Audit) ✅
### Core Trading Logic ✅
- **Signal generation**: Both use shared `indicators::generate_signal()` (indicators.rs:442-650)
- **Position sizing**: Both use shared `Strategy::calculate_position_size()` (strategy.rs:29-55)
- **Signal generation**: Both use shared `indicators::generate_signal()` (bot:739; bt:583,630)
- **Position sizing**: Both use shared `Strategy::calculate_position_size()` (bot:463-468; bt:199-201)
- Volatility-adjusted via ATR
- Confidence scaling: 0.7 + 0.3 * confidence
- Max position size cap: 25%
- Cash reserve: 5%
- **Stop-loss/trailing/time exit**: Both use shared `Strategy::check_stop_loss_take_profit()` (strategy.rs:57-128)
- **Stop-loss/trailing/time exit**: Both use shared `Strategy::check_stop_loss_take_profit()` (bot:473-486; bt:373-380)
- Hard max loss cap: 5%
- ATR-based stop: 3.0x ATR below entry
- Fixed fallback stop: 2.5%
@@ -35,19 +51,37 @@ The refactor to extract shared logic into `strategy.rs` has **eliminated all pre
- Time exit: 40 bars if below trailing activation threshold
### Portfolio Controls ✅
- **Cooldown timers**: Both implement 5-bar cooldown after stop-loss (bot:395-406,521-533; bt:133-138,242-247)
- **Ramp-up period**: Both limit to 1 new position per bar for first 15 bars (bot:433-441; bt:158-161)
- **Drawdown circuit breaker**: Both halt for 20 bars at 12% drawdown (bot:244-268; bt:83-118)
- **Sector limits**: Both enforce max 2 per sector (bot:423-430; bt:149-156)
- **Max concurrent positions**: Both enforce max 7 (bot:414-421; bt:145-147)
- **Momentum ranking**: Both filter to top 10 momentum stocks (bot:669-690; bt:438-449)
- **bars_held increment**: Both increment at START of trading cycle/bar (bot:614-617; bt:433-436)
- **Cooldown timers**: Both implement 5-bar cooldown after stop-loss (bot:507-517,659-670; bt:169-173,294-299)
- **Ramp-up period**: Both limit to 1 new position per cycle/bar for first 15 bars (bot:543-552; bt:194-196)
- **Drawdown circuit breaker**: Both trigger at 15% with 10-bar cooldown (bot:353-362; bt:104-113)
- **BUT**: bot.rs missing peak reset on expiry (see Critical Bug #1)
- **Sector limits**: Both enforce max 2 per sector (bot:534-541; bt:184-191)
- **Max concurrent positions**: Both enforce max 7 (bot:525-532; bt:180-182)
- **Momentum ranking**: Both filter to top 20 momentum stocks (bot:818-838; bt:543-554)
- **bars_held increment**: Both increment at START of trading cycle/bar (bot:763-765; bt:539-541)
### Config Constants — ALL CONSISTENT ✅
Both files import and use identical values from config.rs:
- `ATR_STOP_MULTIPLIER`: 3.0x
- `ATR_TRAIL_MULTIPLIER`: 2.0x
- `ATR_TRAIL_ACTIVATION_MULTIPLIER`: 2.0x
- `MAX_POSITION_SIZE`: 25%
- `MAX_CONCURRENT_POSITIONS`: 7
- `MAX_SECTOR_POSITIONS`: 2
- `MAX_DRAWDOWN_HALT`: 15% (updated from 12%)
- `DRAWDOWN_HALT_BARS`: 10 (updated from 20)
- `REENTRY_COOLDOWN_BARS`: 5
- `RAMPUP_PERIOD_BARS`: 15
- `TOP_MOMENTUM_COUNT`: 20 (updated from 10)
- `TIME_EXIT_BARS`: 40
- `MIN_CASH_RESERVE`: 5%
- `MAX_LOSS_PCT`: 5%
### Warmup Requirements ✅
**Daily mode**: `max(35 MACD, 15 RSI, 50 EMA, 28 ADX, 20 BB, 63 momentum) + 5 = 68 bars`
**Hourly mode**: `max(35 MACD, 15 RSI, 200 EMA, 28 ADX, 20 BB, 63 momentum) + 5 = 205 bars`
Calculation in `config.rs:169-183` (`IndicatorParams::min_bars()`)
Calculation in `config.rs:193-206` (`IndicatorParams::min_bars()`)
- RSI-2/3 warmup covered by RSI-14 requirement (15 > 3)
- MACD needs slow + signal periods (26 + 9 = 35)
- ADX needs 2x period for smoothing (14 * 2 = 28)
@@ -55,13 +89,20 @@ Calculation in `config.rs:169-183` (`IndicatorParams::min_bars()`)
Both bot.rs and backtester.rs fetch sufficient historical data and validate bar count before trading.
### Entry/Exit Flow ✅
**Both follow identical two-phase execution**:
- Phase 1: Process all sells (stop-loss, trailing, time exit, signals)
- Phase 2: Process buys for top momentum stocks only
- bot.rs lines 800-849
- backtester.rs lines 556-642
---
## INTENTIONAL DIFFERENCES (Not Bugs) ✅
### 1. Slippage Modeling
- **Backtester**: Applies 10 bps on both entry and exit (backtester.rs:63-71)
- **Live bot**: Uses actual fill prices from Alpaca API (bot.rs:456-460)
- **Backtester**: Applies 10 bps on both entry and exit (backtester.rs:71-78)
- **Live bot**: Uses actual fill prices from Alpaca API (bot.rs:567-571)
- **Verdict**: Expected difference. Backtester simulates realistic costs; live bot gets market fills.
### 2. RSI Short Period Scaling
@@ -74,34 +115,37 @@ Both bot.rs and backtester.rs fetch sufficient historical data and validate bar
- **Hourly mode**: `ema_trend: 200` (200-hour ≈ 28.5-day trend filter)
- **Verdict**: Hourly uses 4x scaling (not 7x like other indicators) for longer-term trend context. Appears intentional.
### 4. Hourly Late-Day Entry Prevention
- **Backtester**: Blocks entries after 19:00 UTC in hourly mode (backtester.rs:158-166) to prevent same-day stop-loss exits
- **Bot**: Relies on PDT exit blocking instead (bot.rs:619-628)
- **Verdict**: Two different approaches to PDT prevention. See Critical Divergence #2.
---
## STRATEGY ARCHITECTURE (2026-02-12)
## STRATEGY ARCHITECTURE (2026-02-12, Still Current)
### Regime-Adaptive Dual Signal
The new strategy uses **ADX for regime detection** and switches between two modes:
The strategy uses **ADX for regime detection** and switches between two modes:
#### RANGE-BOUND (ADX < 20): Mean Reversion
- **Entry**: Connors RSI-2 extreme oversold (RSI-2 < 10) + price above 200 EMA
- **Entry**: Connors RSI-2 extreme oversold (RSI-2 < 10) + price above EMA trend
- **Exit**: RSI-2 extreme overbought (RSI-2 > 90) or standard exits
- **Conviction boosters**: Bollinger Band extremes, volume confirmation
- **Logic**: indicators.rs:490-526
#### TRENDING (ADX > 25): Momentum Pullback
- **Entry**: Pullbacks in strong trends (RSI-14 dips 25-40, price near EMA support, MACD confirming)
- **Exit**: Trend break (EMA crossover down) or standard exits
- **Conviction boosters**: Strong trend (ADX > 40), DI+/DI- alignment
- **Logic**: indicators.rs:531-557
#### UNIVERSAL SIGNALS (Both Regimes)
- RSI-14 extremes in trending context (indicators.rs:564-570)
- MACD crossovers (indicators.rs:573-583)
- EMA crossovers (indicators.rs:599-608)
- Volume gate (reduces scores 50% if volume < 80% of 20-period MA) (indicators.rs:611-614)
- RSI-14 extremes in trending context
- MACD crossovers
- EMA crossovers
- Volume gate (reduces scores 50% if volume < 80% of 20-period MA)
### Signal Thresholds
### Signal Thresholds (Updated 2026-02-13)
- **StrongBuy**: total_score >= 7.0
- **Buy**: total_score >= 4.5
- **Buy**: total_score >= 4.0 (was 4.5)
- **StrongSell**: total_score <= -7.0
- **Sell**: total_score <= -4.0
- **Hold**: everything else
@@ -110,66 +154,26 @@ Confidence: `(total_score.abs() / 12.0).min(1.0)`
---
## CONFIG PARAMETERS (2026-02-12)
### Indicator Periods
- RSI: 14 (standard), RSI-2 (daily) / RSI-3 (hourly) for mean reversion
- MACD: 12/26/9 (standard)
- Momentum: 63 bars
- EMA: 9/21/50 (daily), 9/21/200 (hourly)
- ADX: 14, thresholds: 20 (range), 25 (trend), 40 (strong)
- Bollinger Bands: 20-period, 2 std dev
- ATR: 14-period
- Volume MA: 20-period, threshold: 0.8x
### Risk Management
- **Position sizing**: 1.2% risk per trade (RISK_PER_TRADE)
- **ATR stop**: 3.0x ATR below entry (was 2.5x)
- **ATR trailing stop**: 2.0x ATR distance, activates after 2.0x ATR gain (was 1.5x/1.5x)
- **Max position size**: 25% (was 22%)
- **Max loss cap**: 5% (was 4%)
- **Stop loss fallback**: 2.5% (when ATR unavailable)
- **Time exit**: 40 bars (was 30)
- **Cash reserve**: 5%
### Portfolio Limits
- **Max concurrent positions**: 7 (was 5)
- **Max per sector**: 2 (unchanged)
- **Momentum ranking**: Top 10 stocks (was 4)
- **Drawdown halt**: 12% triggers 20-bar cooldown (was 35 bars)
- **Reentry cooldown**: 5 bars after stop-loss (was 7)
- **Ramp-up period**: 15 bars, 1 new position per bar (was 30 bars)
- **PDT protection**: Max 3 day trades in rolling 5-business-day window (bot:34-36; bt:279-280)
### Backtester
- **Slippage**: 10 bps per trade
- **Risk-free rate**: 5% annually for Sharpe/Sortino
---
## KEY LESSONS
### 1. Shared Logic Eliminates Drift
Extracting common logic into `strategy.rs` ensures bot and backtester CANNOT diverge. Previously, duplicate implementations led to subtle differences (partial exits, bars_held increment timing, cooldown logic).
Extracting common logic into `strategy.rs` ensures bot and backtester CANNOT diverge. All core trading logic (signal generation, position sizing, stop-loss/trailing/time exit) is now in shared modules.
### 2. Warmup Must Account for Longest Chain
For hourly mode, EMA-200 dominates warmup (205 bars). ADX also needs 2x period (28 bars) for proper smoothing. The `+ 5` safety margin is critical.
### 2. Drawdown Circuit Breaker Needs Peak Reset on Resume
Without resetting the peak when halt expires, any minor dip after partial recovery will immediately re-trigger the halt. This creates cascading halts that keep the bot in cash for extended periods. Backtester had this right; bot.rs needs the fix.
### 3. NaN Handling is Critical
Indicators can produce NaN during warmup or with insufficient data. The signal generator uses safe defaults (e.g., `if adx.is_nan() { 22.0 }`) to prevent scoring errors.
### 3. PDT Protection Strategy Differs by Timeframe
- **Daily mode**: Phase separation (sells Phase 1, buys Phase 2) prevents day trades by construction. PDT enforcement not needed.
- **Hourly mode**: Late-day entry prevention (backtester) vs exit blocking (bot) are two valid approaches, but they're not identical. For portfolios < $25K, bot will hold positions overnight more often than backtest suggests.
### 4. ATR Fallbacks Prevent Edge Cases
When ATR is zero/unavailable (e.g., low volatility or warmup), code falls back to fixed percentage stops. Without this, position sizing could explode or stops could fail.
### 4. Config Constants Must Be Audited After Every Change
Recent changes to drawdown thresholds, momentum pool size, and buy thresholds were all consistent between bot and backtester, but manual audit was required to verify. Future changes should trigger automated consistency checks.
### 5. Slippage Modeling is Non-Negotiable
The backtester applies 10 bps slippage on both sides (20 bps round-trip) to simulate realistic fills. This prevents overfitting to unrealistic backtest performance.
### 5. Warmup Must Account for Longest Indicator Chain
For hourly mode, EMA-200 dominates warmup (205 bars). The `+ 5` safety margin in `min_bars()` is critical.
### 6. Data Type Consistency Matters for PDT Protection
**CRITICAL**: bot.rs and backtester.rs must use the SAME data type for day_trades tracking. Using `Vec<String>` in bot.rs vs `Vec<NaiveDate>` in backtester.rs creates:
- String parsing overhead on every PDT check
- Silent failures if malformed dates enter the system (parse errors return false)
- Inconsistent error handling between live and backtest
- **Fix**: Change bot.rs to store `Vec<NaiveDate>` internally, parse once on load, serialize to JSON as strings
### 6. Data Type Consistency for PDT Tracking (FIXED)
bot.rs now uses `Vec<NaiveDate>` for day_trades (line 56), matching backtester.rs (line 42). Previous audit found this as a critical bug; it's now fixed. ✅
---
@@ -182,7 +186,7 @@ When new changes are made, verify:
3. **Risk management**: Still using shared `Strategy::check_stop_loss_take_profit()`?
4. **Cooldown timers**: Identical logic in both files?
5. **Ramp-up period**: Identical logic in both files?
6. **Drawdown halt**: Identical trigger and resume logic?
6. **Drawdown halt**: Identical trigger logic? Peak reset on expiry?
7. **Sector limits**: Same `MAX_SECTOR_POSITIONS` constant?
8. **Max positions**: Same `MAX_CONCURRENT_POSITIONS` constant?
9. **Momentum ranking**: Same `TOP_MOMENTUM_COUNT` constant?
@@ -191,104 +195,73 @@ When new changes are made, verify:
12. **Config propagation**: Are new constants used consistently?
13. **NaN handling**: Safe defaults for all indicator checks?
14. **ATR guards**: Checks for `> 0.0` before division?
15. **PDT protection**: Same constants, logic, and DATA TYPES in both files?
15. **PDT protection**: Same constants, logic, and data types? Document timeframe-specific behavior.
---
## FILES AUDITED (2026-02-12 PDT Audit)
- `/home/work/Documents/rust/invest-bot/src/bot.rs` (921 lines)
- `/home/work/Documents/rust/invest-bot/src/backtester.rs` (907 lines)
## FILES AUDITED (2026-02-13)
- `/home/work/Documents/rust/invest-bot/src/bot.rs` (933 lines)
- `/home/work/Documents/rust/invest-bot/src/backtester.rs` (1002 lines)
- `/home/work/Documents/rust/invest-bot/src/config.rs` (222 lines)
- `/home/work/Documents/rust/invest-bot/src/strategy.rs` (141 lines)
- `/home/work/Documents/rust/invest-bot/src/types.rs` (234 lines)
- `/home/work/Documents/rust/invest-bot/src/paths.rs` (68 lines)
**Total**: 2,130 lines audited
**Issues found**: 1 critical (data type mismatch), 0 medium, 0 low
**Status**: ⚠️ FIX REQUIRED BEFORE PRODUCTION
**Total**: 2,532 lines audited
**Issues found**: 1 critical bug (drawdown peak reset), 1 critical behavioral divergence (PDT enforcement)
**Status**: ⚠️ FIX REQUIRED BEFORE PRODUCTION (drawdown peak reset)
---
## CRITICAL FIX REQUIRED: PDT Data Type Mismatch
## REQUIRED FIX: Drawdown Peak Reset on Halt Expiry
**Problem**: bot.rs stores day_trades as `Vec<String>`, backtester.rs stores as `Vec<NaiveDate>`
**File**: `/home/work/Documents/rust/invest-bot/src/bot.rs`
**Location**: Lines 365-377 (inside drawdown halt expiry check)
**Required Changes to bot.rs:**
**Current code**:
```rust
// Auto-resume after time-based cooldown
if self.drawdown_halt {
if let Some(halt_start) = self.drawdown_halt_start {
if self.trading_cycle_count >= halt_start + DRAWDOWN_HALT_BARS {
tracing::info!(
"Drawdown halt expired after {} cycles. Resuming trading at {:.2}% drawdown.",
DRAWDOWN_HALT_BARS,
drawdown_pct * 100.0
);
self.drawdown_halt = false;
self.drawdown_halt_start = None;
// MISSING: self.peak_portfolio_value = portfolio_value;
}
}
}
```
1. Line 56: Change field type
```rust
day_trades: Vec<NaiveDate>, // was Vec<String>
```
**Required change**: Add after line 374:
```rust
self.peak_portfolio_value = portfolio_value;
```
2. Lines 197-218: Load with parse-once strategy
```rust
fn load_day_trades(&mut self) {
if LIVE_DAY_TRADES_FILE.exists() {
match std::fs::read_to_string(&*LIVE_DAY_TRADES_FILE) {
Ok(content) if !content.is_empty() => {
match serde_json::from_str::<Vec<String>>(&content) {
Ok(date_strings) => {
self.day_trades = date_strings
.iter()
.filter_map(|s| NaiveDate::parse_from_str(s, "%Y-%m-%d").ok())
.collect();
self.prune_old_day_trades();
if !self.day_trades.is_empty() {
tracing::info!("Loaded {} day trades in rolling window.", self.day_trades.len());
}
}
Err(e) => tracing::error!("Error parsing day trades file: {}", e),
}
}
_ => {}
}
}
}
```
**Complete fixed block**:
```rust
// Auto-resume after time-based cooldown
if self.drawdown_halt {
if let Some(halt_start) = self.drawdown_halt_start {
if self.trading_cycle_count >= halt_start + DRAWDOWN_HALT_BARS {
tracing::info!(
"Drawdown halt expired after {} cycles. Resuming trading. \
Peak reset from ${:.2} to ${:.2} (was {:.2}% drawdown).",
DRAWDOWN_HALT_BARS,
self.peak_portfolio_value,
portfolio_value,
drawdown_pct * 100.0
);
self.drawdown_halt = false;
self.drawdown_halt_start = None;
// Reset peak to current value to prevent cascading re-triggers.
self.peak_portfolio_value = portfolio_value;
}
}
}
```
3. Lines 220-229: Serialize to JSON as strings
```rust
fn save_day_trades(&self) {
let date_strings: Vec<String> = self.day_trades
.iter()
.map(|d| d.format("%Y-%m-%d").to_string())
.collect();
match serde_json::to_string_pretty(&date_strings) {
Ok(json) => {
if let Err(e) = std::fs::write(&*LIVE_DAY_TRADES_FILE, json) {
tracing::error!("Error saving day trades file: {}", e);
}
}
Err(e) => tracing::error!("Error serializing day trades: {}", e),
}
}
```
4. Lines 231-239: Use native date comparison
```rust
fn prune_old_day_trades(&mut self) {
let cutoff = self.business_days_ago(PDT_ROLLING_BUSINESS_DAYS);
self.day_trades.retain(|&d| d >= cutoff);
}
```
5. Lines 256-267: Use native date comparison
```rust
fn day_trades_in_window(&self) -> usize {
let cutoff = self.business_days_ago(PDT_ROLLING_BUSINESS_DAYS);
self.day_trades.iter().filter(|&&d| d >= cutoff).count()
}
```
6. Lines 284-289: Push NaiveDate
```rust
fn record_day_trade(&mut self) {
let today = Utc::now().date_naive();
self.day_trades.push(today);
self.save_day_trades();
}
```
**Benefits of fix:**
- Type safety: malformed dates cannot enter the vec
- Performance: native date comparison vs string parsing on every check
- Consistency: matches backtester implementation exactly
- Reliability: no silent failures from parse errors
**Why this matters**: Without this reset, after a 15% drawdown triggers a 10-bar halt, the bot measures future drawdown from the OLD peak. If the portfolio recovers to 12% drawdown and then dips to 13%, it immediately re-triggers the halt. This creates cascading halts. The backtester correctly resets the peak, so backtest results show fewer/shorter halts than live trading would experience.

37
.claude/agent-memory/quant-rust-strategist/MEMORY.md
View File

@@ -4,26 +4,36 @@
- ~100-symbol universe across 14 sectors (expanded from original 50)
- Hybrid momentum + mean-reversion via regime-adaptive dual signal in `generate_signal()`
- strategy.rs: shared logic between bot.rs and backtester.rs
- Backtester restricts buys to top 10 momentum stocks (TOP_MOMENTUM_COUNT=10)
- Signal thresholds: StrongBuy>=7.0, Buy>=4.5, Sell<=-4.0, StrongSell<=-7.0
- Backtester restricts buys to top momentum stocks (TOP_MOMENTUM_COUNT)
- Signal thresholds: StrongBuy>=7.0, Buy>=4.0, Sell<=-4.0, StrongSell<=-7.0
## Bugs Fixed (2026-02-13)
### 1. calculate_results used self.cash instead of equity curve final value
- backtester.rs line ~686: `let final_value = self.cash` missed open positions
- Fixed: use `self.equity_history.last().portfolio_value`
### 2. Drawdown circuit breaker cascading re-triggers
- peak_portfolio_value was never reset after halt, causing immediate re-trigger
- 7+ triggers in 3yr = ~140 bars (19% of backtest) sitting in cash
- Fixed: reset peak to current value on halt resume
### 3. PDT blocking sells in backtester (disabled)
- PDT sell-blocking removed from backtester; it measures strategy alpha not compliance
- Late-day entry prevention in execute_buy remains for hourly PDT defense
- would_be_day_trade was called AFTER position removal = always false (logic bug)
## PDT Implementation (2026-02-12)
- Tracks day trades in rolling 5-business-day window, max 3 allowed
- CRITICAL: Stop-loss exits must NEVER be blocked by PDT (risk mgmt > compliance)
- Late-day entry prevention: On hourly, block buys after 19:00 UTC (~last 2 hours)
- Prevents entries needing same-day stop-loss exits
- Reduced hourly trades 100->86, improved PF 1.24->1.59
- "PDT performance degradation" was mostly IEX data stochasticity, not actual PDT blocking
- PDT blocking DISABLED in backtester (kept in bot.rs for live trading)
## Backtest Results (3-month, 2026-02-12, post-PDT-fix)
### Hourly: +12.00%, Sharpe 0.12, PF 1.59, 52% WR, 86 trades, MaxDD -9.36%
### Daily: +11.68%, Sharpe 2.65, PF 3.07, 61% WR, 18 trades, MaxDD -5.36%
## Current Parameters (config.rs)
## Current Parameters (config.rs, updated 2026-02-13)
- ATR Stop: 3.0x | Trail: 2.0x distance, 2.0x activation
- Risk: 1.2%/trade, max 25% position, 5% cash reserve, 5% max loss
- Max 7 positions, 2/sector | Drawdown halt: 12% (20 bars) | Time exit: 40
- Max 7 positions, 2/sector | Drawdown halt: 15% (10 bars) | Time exit: 40
- Cooldown: 5 bars | Ramp-up: 15 bars | Slippage: 10bps
- Buy threshold: 4.0 (lowered from 4.5) | Momentum pool: top 20 (widened from 10)
- Daily: momentum=63, ema_trend=50 | Hourly: momentum=63, ema_trend=200
- ADX: range<20, trend>25, strong>40
@@ -33,10 +43,11 @@
## Failed Experiments (avoid repeating)
1. Tighter ATR stop (<3.0x): too many stop-outs on hourly
2. Lower buy threshold (3.5): too many weak entries. Keep 4.5
2. Lower buy threshold (3.5): too many weak entries (but 4.0 is fine)
3. Blocking stop-loss exits for PDT: traps capital in losers, dangerous
4. Lower volume threshold (0.7): bad trades on IEX. Keep 0.8
5. Shorter hourly lookbacks: catastrophic losses
6. Drawdown halt 12% with non-resetting peak: cascading re-triggers in multi-year tests
## IEX Data Stochasticity
- Backtests have significant run-to-run variation from IEX data timing
@@ -44,5 +55,5 @@
- Always run 2-3 times and compare ranges before concluding a change helped/hurt
## Build Notes
- `cargo build --release` compiles clean (only dead_code warnings)
- `cargo build --release` compiles clean (only dead_code warnings for types.rs fields)
- No tests exist