Execution books

Execution books are used for Execution type strategies - which are, in turn, used for building or disposing of large positions.

Such algos that may be instructed to, for example, 'buy $250k of AAPL shares by today's close for the best price possible' - and where performance is benchmarked not by P&L nor sharpe - but by how cheaply the algo can fulfil this task.

I.e. whilst most books/strategies aim to achieve a return on a given amount of capital, accumulation books start at zero and aim to build a position of size X by time T, optimizing for cost and/or reliability.

Hence performance analysis for accumulation books primarily revolves around price and reliability. I.e. given an instruction to buy X shares of instrument Z by time Y, then the measures of success are (a) how reliably it can acquire them before time runs out, and (b) how cheaply it could acquire them. The benchmark is, by default, VWAP for the period.

Execution strategies are typically used either as intraday execution algos to fulfil orders given by a higher timeframe (daily/weekly) systematic strategy, or manually to execute against a discretionary strategy.