Strategy pattern · Microstructure
Test an order book imbalance strategy with real latency, not zero.
Super Algo Kit grades the statistical relation behind an order book imbalance strategy B. The strategy reads buying and selling pressure at the top of the book to anticipate the next small price move, over milliseconds to seconds. The effect is real but tiny and fades within moments, so it favours the fastest firms. For retail use after costs, the kit rates it no better than C.
strategy facts
Order-book imbalance as a short-horizon signal
- Family
- Microstructure: pressure in the order book and the order flow
- Markets
- Futures, crypto derivatives, crypto spot and equities
- Timeframes
- Tick data, with horizons of milliseconds to seconds
- Evidence grade
- B for the statistical relation; no better than C for retail use after costs
- Status
- Literature: published, untested by the kit
- Build set
- Low-latency engines: simulator first, then shadow mode on live data
- Data needed
- Depth or top-of-book data with exchange timestamps and sequence numbers, trades and measured latency
From the kit's own card. Not investment advice.
Paper first
Test it before you trade it.
Super Algo Kit runs inside the coding agent you already use. It interviews you, writes the spec with pass marks set first, then builds, backtests and paper-tests. Real money starts only when you type the authorize command.
The plan, step by step
Tick them off as you go.
How does order book imbalance predict price?
Over very short horizons, price changes track order-flow imbalance at the top of the book. Net pressure from new orders, cancellations and trades on one side moves the mid-price. Queue imbalance, the bid size against the ask size, also predicts the direction of the next mid-price change.
The effect is small and decays within moments, so it pays the fastest participants. For slower traders it is more useful for timing a planned order, or skewing quotes, than as a standalone strategy.
Order book imbalance rules in plain words
This is a summary of the card's outline. All values are declared before a test.
- Measure order-flow imbalance and queue imbalance at the top of the book.
- Predict the direction of the next mid-price move over a fixed short horizon.
- Act only when the predicted move beats spread, fees and expected slippage.
- Otherwise use the reading to time an order you already planned.
- Cancel on disconnect, and cap message rate and inventory.
What the evidence says, and where it breaks
Published work links order-book events to price impact, and order imbalance to stock price moves, so the kit grades the relation B. The catch is size. The predicted move is usually smaller than the spread plus fees.
Latency decides the rest. By the time a retail system reacts, faster firms have already moved the price. Bar-based or zero-latency backtests make the idea look far better than it is.
How Super Algo Kit tests an order book imbalance bot
This pattern belongs to the low-latency build set. The kit starts with a simulator, a conservative queue model and measured latency, then runs shadow mode on live data with no orders sent. Pass marks are declared first, and every trial is counted.
Reports show latency percentiles and fill rates, not just a results line. The Backtest Honesty Card requires the queue and fee model to be stated. Out-of-sample checks and the deflated Sharpe ratio come before any paper orders.
Before you rely on it
Where it fails.
- Latency: faster firms move the price before a retail system reacts.
- Flickering and spoofed quotes create false imbalance.
- The predicted move is usually smaller than the spread plus fees.
- Aggregated or throttled feeds hide the events the reading needs.
- Bar-based or zero-latency backtests make the idea look far better than it is.
FAQ
Order-book imbalance: questions.
What is order book imbalance?
Order book imbalance compares buying and selling pressure at the top of the book, such as the bid size against the ask size. Research found it predicts the direction of the next small mid-price move. Super Algo Kit grades that relation B.
Can a retail trader trade order book imbalance?
It is hard, because the move is tiny and fades within moments. Faster firms usually act first, and the move is often smaller than the spread plus fees. Super Algo Kit rates retail use no better than C and suggests using order book imbalance to time planned orders.
Can AI do high frequency trading?
An AI coding agent can write a low-latency system with Super Algo Kit. By default, though, no model sits in the live order path (hard rule H8). Low-latency builds go simulator first, then shadow mode. The kit's paper minimum for HFT-style systems is 20 sessions and 1,000 orders.
What data does an order book imbalance strategy need?
An order book imbalance strategy needs depth or top-of-book data with exchange timestamps and sequence numbers, plus trades. Super Algo Kit also needs measured feed, order-entry and acknowledgement latency, as percentiles. Fee tiers, tick sizes and matching rules come from the market pack and venue card.
Why do order book imbalance backtests look so good?
Bar-based or zero-latency backtests let the strategy act on information it could not have used in time. Super Algo Kit uses a conservative queue model and measured latency instead. It reports latency percentiles and fill rates alongside the results.
Keep reading
More guides.
Crypto · Exchange APIBinanceSuper Algo Kit helps you build a Binance trading bot inside your own AI coding agent.
United States · Data feedDatabentoSuper Algo Kit lets you use Databento data for backtesting a bot you build in your own AI coding agent.Sources the kit's card cites
- Cont, Kukanov and Stoikov (2014). The Price Impact of Order Book Events. Journal of Financial Econometrics.
- Chordia and Subrahmanyam (2004). Order imbalance and individual stock returns: Theory and evidence. Journal of Financial Economics.
Build it with Super Algo Kit.
15 specialist roles and 35 skills inside your coding agent. Paper first, live only when you say.
Super Algo Kit is development software. It is not investment advice, a signal service or a promise of performance. Trading involves substantial risk, including the loss of all capital you trade with.