downtrend-duration-analyzer

---
name: downtrend-duration-analyzer
description: Use when a user wants typical correction lengths by sector or market-cap tier, historical drawdown recovery times, or realistic holding-period estimates for mean-reversion strategies.
category: Market & Macro
version: 1.0.0
tools: []
---

# Downtrend Duration Analyzer

Analyze historical price data to identify downtrend periods (peak-to-trough) and build statistical distributions of correction durations, segmented by sector and market cap, so traders can set realistic expectations for mean-reversion strategies.

## When to Use
- Trader asks about typical correction lengths for a sector or market-cap tier
- User wants historical drawdown recovery times
- Building mean-reversion or pullback strategies that need realistic holding-period estimates
- Comparing correction behavior across market segments
- Setting stop-loss timeouts or position holding-period limits

## Methodology
Using web_search / knowledge_base_search to pull OHLC price history for the relevant universe (optionally filtered by sector, over a lookback of several years):

1. Identify local peaks and troughs using rolling-window analysis (a look-back/look-forward window, e.g. ~20 trading days, is the standard baseline).
2. For each peak-to-trough sequence, calculate the duration in trading days and the depth as a % decline.
3. Segment results by sector and by market-cap tier — Mega (>$200B), Large ($10B-$200B), Mid ($2B-$10B), Small (<$2B).
4. Compute summary statistics: median, mean, and percentiles (25th, 75th, 90th).

Interpretation bands:
- Short corrections (5-15 days): typical pullbacks within an uptrend
- Medium corrections (15-40 days): standard sector rotations
- Extended corrections (40+ days): trend changes or bear markets

## Output
Present a summary table (total downtrends counted, median/mean duration, 25th/75th/90th percentile durations) plus a breakdown by market-cap tier and by sector, and 2-3 key insights (e.g. "larger companies recover faster," or "90% of corrections resolve within N trading days for this sector"). If the user wants a visual, describe the distribution narratively (bucket counts, percentile markers) rather than generating a file.

## Key Principles
1. Statistical rigor — use a robust peak/trough method to avoid noise-induced false signals.
2. Segmentation matters — always break out by sector and market cap; a blended average hides important differences.
3. Realistic expectations — use percentiles, not just the mean, to convey the full range of outcomes.