ADDENDUM: Actionable Edge Mapping + Prediction Market Ladder

Forex Desk — MERGED FINAL

Base: Claude Opus 4.6 (complete, unmodified) + Panel Additions

Panel: Gemini Pro, GPT-4.1, Grok-4 Fast, DeepSeek V3 (cherry-picked unique additions only)

Rule: Kalshi and Polymarket ladder structures are treated equally throughout. Neither is secondary.

Venue hierarchy: OANDA = primary forex execution. Kalshi/Polymarket = prediction market ladders for economic events. Interbank rates = sharp reference.

This addendum supplements the original Forex Desk spec. It does NOT replace the existing spec — it adds two critical layers: (1) mapping every edge to a specific trade on a specific platform, and (2) a systematic methodology for scanning prediction market ladders for forex-relevant contracts.

TOPIC 1: ACTIONABLE EDGE OUTPUT

1. Edge-to-Trade Mapping Table

2. Actionable Output Format

JSON output sent via Telegram and to other desks:

{
  "edge_id": "FOREX-20260315-EURUSD-ST-001",
  "desk": "forex",
  "timestamp": "2026-03-15T14:00:00Z",
  "edge_description": "EUR/USD 4H SuperTrend flipped bullish at 1.0850. Daily SuperTrend bullish (aligned). Yield spread narrowing supports EUR. OANDA retail 68% short (contrarian bullish).",
  "signal_type": "supertrend_flip_htf_aligned",
  "pair": "EUR/USD",
  "direction": "LONG",
  "confidence_level": 75,
  "confluence_score": 4,
  "confluence_details": {
    "supertrend_4h": "bullish (just flipped)",
    "supertrend_daily": "bullish (aligned)",
    "yield_spread": "narrowing (supports EUR)",
    "retail_positioning": "68% short (contrarian bullish)",
    "rsi_14": 52,
    "atr_14_pips": 68
  },
  "sharp_reference": "CME FedWatch + ECB OIS + interbank forward rates",
  "validation_window": "30m",
  "platforms": {
    "OANDA": {
      "action": "Long EUR/USD",
      "entry": 1.0850,
      "stop_loss": 1.0780,
      "take_profit_1": 1.0950,
      "take_profit_2": 1.1020,
      "risk_pips": 70,
      "reward_pips_tp1": 100,
      "reward_pips_tp2": 170,
      "risk_reward": "1.43:1 (TP1) / 2.43:1 (TP2)",
      "suggested_lot_size": "Calculate based on 1-2% account risk / 70 pip stop",
      "automation": "Phoenix bot can execute if auto-trade is enabled",
      "liquidity_note": "High ($10M+ depth); Phoenix bot auto-executes"
    },
    "Kalshi": {
      "action": "Buy YES on 'ECB holds rate in April' (EUR strength = ECB confidence in economy)",
      "current_price": 62,
      "fair_value": 70,
      "edge_cents": 8,
      "edge_after_fees": 6,
      "liquidity_usd": 400,
      "note": "Indirect — forex view expressed through rate expectation"
    },
    "Polymarket": {
      "action": "Buy YES on equivalent ECB rate hold or EUR/USD threshold contract if available",
      "current_price": null,
      "note": "Check active Polymarket FX Majors series and ECB outcome markets. Same forex thesis applies.",
      "liquidity_usd": null
    }
  },
  "time_sensitivity": "MEDIUM — SuperTrend flip is valid until next flip. Pullback entry may offer better price within 2-8 candles (8-32 hours on 4H).",
  "risk_factors": ["FOMC minutes release Wednesday could override technical signal", "DXY near major support — bounce possible", "4H ATR expanding — wider stop needed"]
}

Telegram summary format:

FOREX EDGE: EUR/USD LONG

4H SuperTrend flipped bullish @ 1.0850
Daily: Aligned bullish | Yield spread: Narrowing
Retail: 68% short (contrarian bullish)
Confluence: 4/5

OANDA: Long @ 1.0850 | Stop 1.0780 | TP1 1.0950 | TP2 1.1020
Risk: 70 pips | R:R 1.4:1 (TP1) / 2.4:1 (TP2)
Phoenix bot: Ready to execute

Kalshi: Buy YES 'ECB holds Apr' @ 62c | Fair 70c | Edge 8c
Polymarket: Check EUR/USD or ECB outcome contracts — same thesis applies
Confidence: 75% | Window: 8-32h for pullback entry

3. Platform Priority Ranking

Summary: OANDA is the dominant execution venue for this desk. Kalshi and Polymarket are parallel secondary venues for expressing the macro views that underlie forex trades — both are scanned, both matter, allocation goes to whichever has the better edge on any given contract. The forex desk's most valuable contribution to other desks is its yield spread, COT, and economic data analysis — which directly feeds both Kalshi and Polymarket trading on the Futures and Stocks desks. Interbank rates are the sharp reference for all forex price probability calculations.

TOPIC 2: PREDICTION MARKET LADDER STRUCTURE

1. Desk Data Mapping to Prediction Market Contracts

The Forex Desk's data maps to prediction markets indirectly rather than through direct forex price contracts. Kalshi and Polymarket have very few pure forex price contracts, but they have extensive economic data and rate decision contracts that ARE the fundamental drivers of forex. Where direct forex price ladders do exist on either platform, they are scanned with equal priority alongside the economic data series.

Important: The Forex Desk does NOT have direct forex price prediction market contracts as its primary target. Instead, the desk's macro analysis — yield spreads, economic data expectations, central bank positioning — feeds directly into both Kalshi's and Polymarket's economic data and rate decision ladders. The forex desk should be considered a primary ANALYST for the Futures Desk's prediction market trading. Direct FX price ladders on both platforms are a secondary scanning target.

2. Sharp Reference Identification

3. Probability Curve Construction Method

For economic data contracts (CPI, NFP, GDP):

• Distribution: Normal distribution, centered on the desk's point estimate (weighted average of nowcast models and consensus), with SD from historical forecast errors.
• For rate hike counts: Binomial or Poisson distribution may be more appropriate than Normal. [PANEL: Gemini Pro]
• Calibration data:
  • CPI forecast error: SD ~0.10-0.12% (m/m), based on last 5 years of CPI vs consensus.
  • NFP forecast error: SD ~65-80K jobs, based on last 5 years.
  • GDP forecast error: SD ~0.4-0.6%, based on last 5 years of advance GDP vs GDPNow at release date.
• Forex desk value-add: The forex desk monitors yield curve and rate futures BETWEEN economic releases. If 2-year yields drop 15bps in the week before CPI (market pricing dovish surprise), the forex desk can detect that rate markets are pricing a softer CPI before the consensus estimate adjusts. This rate-market-implied CPI estimate is an additional input.

For rate decision contracts:

• Distribution: Discrete (cut 25bps, cut 50bps, hold, hike 25bps). Probabilities derived from Fed Funds futures.
• Forex desk value-add: COT positioning in interest rate futures can signal when speculators are positioned for a surprise. If specs are max short Eurodollar futures (pricing aggressive hikes) and the forex desk's yield analysis suggests the market is overpricing hawkishness, the desk can flag rate cut contracts as underpriced.

For direct forex price threshold contracts (EUR/USD, GBP/USD, USD/JPY): [PANEL: Gemini Pro]

• Distribution: Lognormal distribution is more appropriate than Normal for currency price levels. Use the current interbank spot rate as the mean anchor. Use historical 30-day realized volatility as the standard deviation input.
• Calibration: For a contract "EUR/USD > 1.10 by June 30" — anchor at current spot + forward points, build lognormal curve with historical vol, calculate P(X > 1.10) using Black-Scholes / lognormal CDF. This gives a more accurate fair probability than a simple normal distribution centered on the current price.
• Recalculation trigger: Whenever the sharp line (interbank spot) moves more than 0.1% or on any major event.

Recalculation frequency:

• Rate decision contracts: Daily, or immediately after Fed speaker events, major data releases.
• Economic data contracts: Daily until T-3 days, then every 6 hours as the release approaches.
• Direct FX price contracts: Every hour during market hours, or whenever sharp line moves >0.1%. More frequently during event windows (central bank meetings, NFP, CPI).
• When yield spreads or COT data shift meaningfully: recalculate all related contract probabilities.

4. Ladder Scanning Methodology

Step-by-step process for forex desk's prediction market scanning — applies to both Kalshi and Polymarket in parallel:

1. Identify active economic/rate contracts: Pull all Kalshi contracts in the CPI, NFP, GDP, and Fed rate series that expire within 30 days. Pull all Polymarket contracts in equivalent economic outcome and FX Majors series that expire within 30 days. Both platforms are scanned simultaneously.

2. Identify active direct FX price contracts: Pull all Kalshi KXFX contracts and Polymarket FX Majors contracts for the tracked pairs (EUR/USD, GBP/USD, USD/JPY, AUD/USD). These are secondary scanning targets but use the same methodology.

3. For each contract, determine the forex desk's edge source:

   • Rate contracts → compare FedWatch / ECB OIS to Kalshi/Polymarket prices. If FedWatch says 72% cut and Kalshi says 60c and Polymarket says 64c, the Kalshi contract has the larger edge (12c vs 8c) — trade Kalshi in this case.
   • CPI contracts → use Cleveland Fed Nowcast + forex desk's TIPS-implied estimate. Calculate fair probability at each threshold. Compare fair value to both Kalshi and Polymarket prices — trade whichever has the larger net edge after fees.
   • NFP contracts → use ADP + claims + forex desk's rate-market-implied employment view.
   • GDP contracts → use GDPNow.
   • Direct FX price contracts → use lognormal model anchored to interbank spot + historical vol.

4. Calculate raw edge at each threshold.

5. Apply forex desk overlays:

   • If COT shows extreme positioning in the rate-sensitive direction, boost confidence (positioning extremes amplify surprise reactions).
   • If yield curve recently inverted/steepened, factor into rate decision probabilities.
   • If DXY is at a technical extreme (SuperTrend at key level), the subsequent USD move on data release may be amplified.

6. Net edge after fees: Subtract 1.5c Kalshi fees. Subtract Polymarket fees (1-2%). Compare net edges across both platforms for the same underlying event.

7. Filter: Minimum 3c net edge, minimum $100 liquidity. Apply to both platforms independently.

8. Track persistence: Log edges over time to see if they are being arbitraged away. Persistent edges are higher quality. Flag any edge that persists >2 hours or increases — these are the highest-conviction opportunities. [PANEL: Gemini Pro]

9. Dual output: Send edges to (a) Telegram for direct Kalshi/Polymarket trading (specify which platform for each trade), AND (b) the Futures Desk and Stocks Desk as supporting signals for their own prediction market analysis.

5. Edge Examples with Real Numbers

Example 1: Fed Rate Decision — FedWatch vs Kalshi and Polymarket Divergence

• Context: FOMC meeting in 2 weeks. Fed speakers have been hawkish this week.
• CME FedWatch: 78% probability of hold, 22% probability of 25bp cut.
• Kalshi contract: "Fed holds rate in June" at 68c (implied 68%).
• Polymarket contract: "Fed holds rate in June" at 71c (implied 71%) — check active Polymarket FOMC series.
• Forex desk analysis: Yield spread (US 2Y-10Y) has steepened 8bps this week (hawkish signal). OANDA retail is 72% long EUR/USD (betting on cut — contrarian says hold is more likely). COT shows specs reducing short USD positions (less conviction in USD weakness → less conviction in a cut). ECB OIS pricing and ECB Speaker Tracker confirm no ECB surprise expected.
• Fair probability of hold: FedWatch 78% + forex desk hawkish overlay → 80%.
• Raw edge vs Kalshi: 80% - 68% = 12 cents. Net edge: 12 - 1.5 = 10.5 cents. Trade Kalshi.
• Raw edge vs Polymarket: 80% - 71% = 9 cents. Net edge after fees: ~7.5 cents. Also tradeable.
• Trade: Buy YES on "Fed holds rate in June" on Kalshi at 68c (larger edge). Also buy on Polymarket if liquidity allows. Also: long USD/JPY on OANDA (hold = USD strength).

Example 2: NFP Ladder — Tail Threshold

• Context: NFP releasing Friday at 8:30 AM ET.
• Consensus: 195K jobs (Investing.com median).
• Forex desk analysis: ADP came in at 230K (strong). Initial claims 4-week MA at 218K (historically correlates with ~210K NFP). Yield curve flattened 5bps post-ADP (market pricing stronger labor market). OANDA positioning shifted to 60% long USD (market expects beat).
• Forex desk point estimate: Weighted: 0.4 * 230K (ADP) + 0.3 * 210K (claims) + 0.3 * 195K (consensus) = 213K
• SD: 70K (historical NFP forecast error)
• Kalshi rung: "NFP above 275K" at 9 cents (implied 9%)
• Polymarket rung: "NFP above 275K" at 11 cents (implied 11%) — if active
• Fair probability: P(Normal(213, 70) > 275) = P(Z > (275-213)/70) = P(Z > 0.886) = 18.8%
• Kalshi raw edge: 18.8% - 9% = 9.8 cents. Net: 8.3 cents. Trade Kalshi (better price).
• Polymarket raw edge: 18.8% - 11% = 7.8 cents. Net: ~6.3 cents. Secondary if Kalshi liquidity exhausted.
• Trade: Buy YES on "NFP above 275K" on Kalshi at 9c first, Polymarket second. The market is underpricing the right tail because the forex desk's macro analysis (ADP + claims + yield signal) suggests the mean is higher than consensus.
• OANDA trade: Also pre-position long USD/JPY ahead of NFP if the desk believes in a beat.

Example 3: CPI Ladder — Nowcast Divergence from Consensus

• Context: CPI releasing Tuesday at 8:30 AM ET.
• Consensus: 3.2% headline CPI y/y.
• Cleveland Fed Nowcast: 3.4% (running hotter).
• Forex desk analysis: 5Y breakeven inflation (TIPS spread) rose 8bps this week to 2.45% — bond market is pricing hotter inflation than consensus. Gasoline prices up 4% month-over-month. Shelter CPI component is sticky (Zillow rents still elevated).
• Forex desk point estimate: 0.6 * 3.4 (nowcast) + 0.4 * 3.2 (consensus) = 3.32%
• SD: 0.11% (CPI historical)
• Kalshi rung: "CPI above 3.4%" at 22 cents (implied 22%)
• Polymarket rung: "CPI above 3.4%" at 24 cents (implied 24%) — if active
• Fair probability: P(Normal(3.32, 0.11) > 3.4) = P(Z > (3.4-3.32)/0.11) = P(Z > 0.727) = 23.4%
• Kalshi: 23.4% - 22% = 1.4 cents. Not tradeable after fees.
• Polymarket: 23.4% - 24% = negative. No edge; in fact, slight reverse edge.
• Check the next rung on both platforms: Kalshi "CPI above 3.5%" at 10 cents (implied 10%). Polymarket "CPI above 3.5%" at 11 cents (implied 11%).
• Fair probability: P(Z > (3.5-3.32)/0.11) = P(Z > 1.636) = 5.1%
• Kalshi reverse edge: Fair value is 5.1%, market is at 10%. The market is overpricing this tail. Buy NO at 90c. Fair NO = 94.9c. Edge = 4.9c. Net = 3.4c. Tradeable.
• Polymarket reverse edge: Fair value 5.1%, market at 11%. Buy NO. Edge = 5.9c. Net after fees: ~4.4c. Also tradeable — slightly better on this rung.
• Trade: Buy NO on "CPI above 3.5%" on both platforms (whichever has better liquidity at the ask). The market is overpricing the hot tail.
• OANDA trade: If CPI comes in at 3.3-3.4%, short EUR/USD (hotter = USD strength). If CPI comes in at 3.1%, long EUR/USD (cooler = dovish Fed expectations).

Example 4: Direct FX Price Threshold — EUR/USD Lognormal Model [PANEL: Gemini Pro]

• Context: EUR/USD at 1.08 current spot. 30-day realized vol: 12% annualized. Forward points add 0.015 (carry). Baseline for 30-day horizon: 1.095.
• Polymarket contract: "EUR/USD closes above 1.10 by June 30" at 28c.
• Kalshi contract: "EUR/USD closes above 1.10 by June 30" at 26c (if listed).
• Model: Lognormal distribution, mean = 1.095 (spot + carry), σ = 12% annualized scaled to 30 days = ~3.5%.
• Fair probability: P(Lognormal(1.095, 0.035) > 1.10) ≈ 34%
• Polymarket edge: 34c - 28c = 6c raw. Net: ~4.5c. Tradeable.
• Kalshi edge: 34c - 26c = 8c raw. Net: ~6.5c. Better if listed.
• Trade: Buy YES on Kalshi first (better price), Polymarket second.

IMPORTANT NOTES

• The Forex Desk's primary prediction market contribution is INDIRECT — its macro analysis feeds both Kalshi and Polymarket economic data and rate decision contracts. Direct forex price prediction markets are rare and illiquid but are scanned on both platforms with equal priority.
• OANDA remains the primary execution venue for forex-specific trades. OANDA is the primary forex trading venue; Kalshi and Polymarket are parallel, equally-weighted prediction market venues for the macro views that underlie those trades. Interbank rates are the sharp reference for all forex price probability calculations.
• The desk should send its economic data fair-value estimates to the Futures Desk and Stocks Desk as inputs for their own prediction market scanning — covering both Kalshi and Polymarket.
• COT data is weekly with a 3-day lag — it's a positioning context signal, not a timing signal. Never trade COT alone; use it as a confidence overlay.
• Yield spread analysis is the highest-value cross-desk contribution from the forex desk. US-DE, US-JP, US-UK spreads directly inform rate decision contract pricing on both Kalshi and Polymarket.
• Phoenix bot trades on OANDA should be informed by, but not dependent on, prediction market signals. The bot has its own VWAP reversion strategy that operates independently.
• When the same underlying event has contracts on both Kalshi and Polymarket, compare net edges after fees and allocate to the platform with the better price. If both have sufficient liquidity, split allocation to maximize fill.
• Persistent edges (not arbitraged away after >2 hours) are the highest-quality signals — log and prioritize these. [PANEL: Gemini Pro]
• For carry trades, the edge is the interest rate differential captured as daily swap — not a technical signal. Carry trades require risk management for sudden risk-off events that unwind carry positions violently. [PANEL: Gemini Pro]
• ECB OIS pricing and ECB Speaker Tracker calibration should be used alongside CME FedWatch as the sharp reference for European rate decision contracts on both Kalshi and Polymarket. [PANEL: Grok-4 Fast]