trade-analyzer

name	trade-analyzer
description	Analyzes trade proposals when the user considers trading players, asks 'should I trade X for Y', evaluates a trade offer, or wants to find trade targets. Compares players' stats, evaluates H2H category impact, and considers roster fit. Triggers on: 'trade', 'swap', 'deal', 'give X for Y', 'trade target', 'should I accept', 'trade offer'.

Trade Analyzer

When the user asks about a trade or is evaluating a trade proposal, follow these steps:

Step 1: Understand the Categories

Call get_scoring_categories to know all 14 H2H categories and their direction. This is essential — a trade that looks bad in total points might be great for category balance.

Step 2: Run the Trade Analysis

Call analyze_trade with the players being given and received (comma-separated names). This gives the baseline stat comparison.

Step 3: Deep Player Evaluation

Call get_player_info for each player involved in the trade to get detailed season stats, injury status, and ownership trends.

Step 4: Assess Roster Fit

Call get_my_roster to understand:

• What positions are covered after the trade?
• Does the trade create a positional gap?
• Are there bench players who can fill in?

Step 5: Check Memory Context (if available)

If memory tools are available:

• Call get_category_trends to see which categories have been consistently strong or weak
• Call get_matchup_history to see if certain categories have been costing you matchups
• A trade that hurts a strong category but helps a weak one is often worth it — even if the "point value" goes down

Step 6: Category Impact Analysis

Build a category-by-category impact table:

Category	Direction	Current Strength	Impact	Net Effect
HR	Higher wins	STRONG	Lose 5 HR, Gain 2 HR	-3 HR (slight decline)
SB	Higher wins	WEAK	Lose 0 SB, Gain 15 SB	+15 SB (big improvement)
ERA	Lower wins	AVERAGE	...	...

For each category:

• Will this trade help, hurt, or be neutral?
• Is the affected category one you're already strong/weak in?
• Could this flip a category from losing to winning in typical matchups?

Critical reminders for reverse categories:

• B_SO (lower wins): A player with fewer strikeouts is BETTER
• ERA (lower wins): Lower ERA is BETTER
• WHIP (lower wins): Lower WHIP is BETTER
• L (lower wins): Fewer losses is BETTER

Step 7: Verdict

Give a clear recommendation:

• Accept / Decline / Counter-offer
• How many categories does this trade help vs hurt?
• What's the net category swing? (e.g., "You gain an edge in 3 categories but lose ground in 1")
• Is the positional fit workable?
• Any injury risks to flag?

Step 8: Save to Memory (if available)

If the user completes a trade:

• Call save_roster_move with action="TRADE_SEND" for each player given away
• Call save_roster_move with action="TRADE_RECEIVE" for each player received
• Update save_category_trend if the trade significantly changes category projections