Smart execution

Prediction-market books are thin — a large marketable order clips the print and attracts adverse selection. Neither venue offers iceberg, TWAP, or pegged orders. Mithril’s execution engine does: preview the impact of an order, then let Mithril work it under a slippage cap and hand you a transaction-cost report at the end.

Preview impact before you trade

POST /v1/complex_orders/preview estimates fillable size and average price against live depth, and recommends a strategy:

curl -X POST https://api.trymithril.com/v1/complex_orders/preview \
  -H "Authorization: Bearer $MITHRIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "market_id": "mkt_...", "token_id": "...", "action": "buy", "quantity": "500" }'

preview = s.post(f"{BASE}/v1/complex_orders/preview", json={
    "market_id": market["id"], "token_id": token_id,
    "action": "buy", "quantity": "500",
}).json()
print(preview["impact"]["fillable_qty"], preview["impact"]["avg_price"])
print(preview["recommended_plan"]["strategy"], "—", preview["recommended_plan"]["reason"])

The response makes the walk-the-book impact and the fee trade-off explicit — crossing pays the taker fee, resting earns the maker rebate:

{
  "impact": {
    "action": "buy",
    "requested_qty": "500",
    "fillable_qty": "500",
    "avg_price": "0.436",
    "worst_price": "0.45",
    "mid": "0.42",
    "notional_cost": "218.00",
    "slippage_bps": 380,
    "levels_consumed": 4,
    "complete": true
  },
  "recommended_plan": { "strategy": "iceberg", "clip_qty": "125", "slices": 4,
                        "reason": "size exceeds top-of-book; work it in clips as liquidity refills" },
  "fee_rate": 0.02,
  "taker": { "taker_fee": "4.36", "maker_rebate": "0", "all_in_avg_price": "0.445", "all_in_cost": "222.36", "fee_bps": 200 },
  "maker": { "taker_fee": "0", "maker_rebate": "2.18", "all_in_avg_price": "0.432", "all_in_cost": "215.82", "fee_bps": 0 }
}

fillable_qty, avg_price, and slippage_bps live under the impact object; recommended_plan is top-level.

Strategies

Strategy	Use it when
`smart_taker`	Take liquidity up to a worst price, sweeping only what’s cheap
`iceberg`	Post size in clips so you don’t show the whole hand
`peg`	Rest passively at/inside the spread, repegging within a band
`adaptive`	”Just get me best execution” — Mithril balances urgency vs impact

Work an order

Submit a complex order the same way as a plain one — with an idempotency-key — choosing a type and its params:

curl -X POST https://api.trymithril.com/v1/complex_orders \
  -H "Authorization: Bearer $MITHRIL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "idempotency-key: $(uuidgen)" \
  -d '{
    "subaccount_id": "sub_...",
    "market_id": "mkt_...",
    "token_id": "...",
    "type": "adaptive",
    "action": "buy",
    "total_quantity": "500",
    "adaptive": { "limit_price": "0.62", "urgency": "low", "max_duration_s": 300 }
  }'

import uuid

co = s.post(f"{BASE}/v1/complex_orders",
    headers={"idempotency-key": str(uuid.uuid4())},
    json={
        "subaccount_id": sub_id, "market_id": market["id"], "token_id": token_id,
        "type": "adaptive", "action": "buy", "total_quantity": "500",
        "adaptive": {"limit_price": "0.62", "urgency": "low", "max_duration_s": 300},
    }).json()
cid = co["id"]

Watch it work

GET /v1/complex_orders/{id}/children returns the plan’s child orders with live queue positions:

{
  "children": [
    { "order_id": "ord_...", "status": "filled", "quantity": "125", "price": "0.43",
      "traded_qty": "125", "average_price": "0.429", "created_at": "2026-07-16T14:30:05Z" },
    { "order_id": "ord_...", "status": "open", "quantity": "125", "price": "0.43",
      "traded_qty": "0", "average_price": "0", "created_at": "2026-07-16T14:31:10Z",
      "queue_ahead": "800" }
  ]
}

Read the execution report (TCA)

GET /v1/complex_orders/{id}/execution_report is your receipt: what you paid versus arrival, versus estimate, versus a naive sweep — measured on your own fills, not claimed.

{
  "action": "buy",
  "total_quantity": "500",
  "filled_qty": "500",
  "realized_avg_price": "0.431",
  "arrival_mid_price": "0.42",
  "estimated_avg_price": "0.436",
  "naive_sweep_avg_price": "0.448",
  "implementation_shortfall_bps": 262,
  "estimate_error_bps": 50,
  "realized_fees_paid": "2.15",
  "realized_all_in_avg_price": "0.435",
  "naive_sweep_all_in_price": "0.457",
  "savings_vs_naive": "11.00",
  "savings_vs_naive_bps": 220,
  "benchmark_available": true,
  "children": [ /* ...ChildFillView... */ ]
}

savings_vs_naive

string

Quote-currency saved versus sweeping the book naively — your number, measured per fill. benchmark_available: false means there wasn’t enough book data to compute it.

Cancel the remainder any time:

curl -X POST https://api.trymithril.com/v1/complex_orders/co_.../cancel \
  -H "Authorization: Bearer $MITHRIL_API_KEY"

Conditional orders (take-profit / stop-loss / stop) are on the roadmap and are rejected at submission today, so you never get a plan that nothing will act on.

Next: Risk & P&L

See the exposure and P&L those fills produced.

​Preview impact before you trade

​Strategies

​Work an order

​Watch it work

​Read the execution report (TCA)

Next: Risk & P&L

Preview impact before you trade

Strategies

Work an order

Watch it work

Read the execution report (TCA)