Skip to main content

Transaction Cost Analysis (TCA)

Execution alpha is tiny per trade and enormous over thousands of them — it is a large part of how banks and prop desks keep their edge. TCA measures how far the price you actually achieved drifted from the price when you decided to trade.

Open cBots → Execution Cost (/quant/tca).

What it measures​

Given the arrival (decision) price, the side, and your fills (price × quantity), it reports:

  • Average fill price (VWAP) — the volume-weighted price you actually got.
  • Slippage (bps) — the drift from arrival to VWAP in basis points, signed so a positive number is a cost (buying above arrival or selling below it) and a negative number is price improvement.
  • Implementation shortfall — that cost expressed in price × quantity terms: the money the drift cost you on this order.
POST /api/quant/tca
{ "arrivalPrice": 1.1000, "side": "Buy",
  "fills": [ { "price": 1.1010, "quantity": 100 }, { "price": 1.1020, "quantity": 100 } ] }

Smart slicing (Almgren-Chriss)​

Beyond measuring cost, cMind can plan a large order to minimise it. cBots → Execution Schedule (/quant/execution) builds an Almgren-Chriss optimal-execution schedule: given the total quantity, a number of slices, your risk aversion, volatility and temporary market impact, it returns the size to trade in each slice. Higher risk aversion front-loads the schedule (cutting timing risk); zero risk aversion flattens to an even TWAP. The slices always sum to the total.

POST /api/quant/execution-schedule
{ "totalQuantity": 100, "slices": 5, "riskAversion": 2, "volatility": 0.02, "temporaryImpact": 0.1 }

Why it is reliable​

Pure, deterministic domain code (Core.Execution) with no infrastructure dependency and no external calls — unit-tested for the buy/sell cost sign, price improvement, zero-slippage, VWAP aggregation, and input guards. This is the measurement half of execution quality; it is the same shortfall metric the copy engine uses to judge (and, with smart slicing, reduce) the cost of mirrored orders.