Key Takeaways
Algorithmic trading refers to using computer algorithms to execute trades based on predefined strategies. Binance offers two types of algo trading: TWAP and POV.
TWAP involves executing trades evenly over a specified period. POV involves executing trades based on a predefined percentage of the market volume.
Algo orders can reduce slippage and market impact, especially for large trades and illiquid assets. Traders handling small volumes of liquid assets may find traditional market orders more efficient.
Trade size, asset liquidity, and algorithm configuration can all affect order performance and execution quality. Proper configuration, such as setting a suitable duration and limit price, can increase the effectiveness of algo orders.
Introduction
Algorithmic trading has revolutionized how traders execute orders by automating the process using computer algorithms. Binance offers two primary types of algorithmic orders: Time Weighted Average Price (TWAP) and Percentage of Volume (POV). These tools can help traders execute orders according to specific strategies designed to minimize market impact and slippage.
This article delves into some case studies to analyze how algo orders perform compared to traditional market orders under different conditions like asset liquidity, trade size, and algorithm configurations.
What Are TWAP and POV?
Time Weighted Average Price (TWAP)
The TWAP strategy executes trades evenly over a specified time period. For example, a TWAP of 1 hour aims to execute the trade quantity distributed across this time period. This strategy helps to minimize the impact of large orders on the market price by spreading them out over time.
Percentage of Volume (POV)
POV executes trades based on a predefined percentage of the total market volume during a particular period. For instance, an algorithm might aim to execute trades that represent 10% of the total market volume during a specific time frame. This strategy adjusts the execution rate based on market activity to minimize market impact.
Case Studies Overview
To better understand how much better algo orders are when compared to traditional market orders, we referenced approximately 25,000 historical anonymised data points of actual Binance algo trades. The studies compared the slippage (the difference between expected and actual execution prices) between algo orders and conventional market orders as benchmarks.
The edge represents the performance difference between the algo and its benchmark. For example, an edge of +0.01% indicates no significant difference, -1% means the algo is underperforming compared to the benchmark, and +1% shows the algo is delivering significant benefits.
Note: The case studies are based on anonymised, historical execution data from Binance over a specified period. They are intended only to illustrate general trading behaviour and do not provide reliable indicators of future execution quality. Future performance is not guaranteed.ย
Case 1: Overall
Algo Type
On average, the overall slippage across all notionals, asset types, and durations (for TWAP strategy) shows that the TWAP benchmark (market orders) encountered a slippage of -0.17%, while its respective algo execution had a slippage of -0.18%. For POV, the benchmark encountered a slippage of -2.37%, while the algo execution had -0.56%.ย
Overall, the data indicates an edge of -0.01% for TWAP and +1.81% for POV. Remember that the edge is simply the difference between the benchmark and the algo performances.
Asset Type
Looking into asset type, defining BTC and ETH as โLiquidโ and the rest of non-stablecoins as โIlliquidโ, the benchmark for Liquid has -0.01% slippage, and Illiquid has -0.25% slippage. The algo execution for these parameters shows an edge of -0.02% for Liquid and +0.04% for Illiquid.
In a later case study, we will explore why TWAP, as well as Liquid assets, may show an overall negative edge relative to the benchmark.ย
Case 2: Multi-Dimension (Size)
Trade Sizeย
Larger trades suffer higher slippage. Trades over 2 million USDT showed slippages up to -7.4% in the benchmark scenario.
Trade Size vs. Asset Type
In the chart below, we look only at the edge of algo orders across two parameters. Algo orders were particularly effective with large trades in illiquid assets, with edges reaching an average of 13%.
Liquid assets with smaller trade sizes showed a negative edge, which suggests that liquid coins could have sufficient liquidity to absorb these trades without reliance on Algorithms. This negative edge may also occur when the Algo Order is not configured properly (see case study 4).
Case 3: Multi-Dimension (Algo Type)
Algo Type vs. Trade Size
When comparing algo types for large-sized trades, POV looks to outperform TWAP, especially on larger quantities. This could likely be attributed to the behaviour of POVโs execution rate, which increases as market volume increases. This allows the algorithm to essentially โmove with the marketโ, which helps reduce slippage.
Algo Type vs. Duration
Despite the larger edge, it is also important to highlight that the time-to-market for POV trades could fluctuate greatly, and can be a deciding factor in picking which algo to use. We can see below that in some cases, to reap the full benefits of POV, trades could extend upwards of 8 hours.
Case 4: Algo Configurations
Glancing back at Case 1, we saw that the TWAP and Liquid assets overall edge could be slightly negative. However, upon further inspection, we can see that those results may be skewed due to the TWAP configurations.
TWAP duration (Duration vs. Asset Type)
In the chart below, negative edges are more prominent when the duration is above 1 hour for Liquid assets. This suggests that the asset could have sufficient liquidity to complete the trade quickly, but was probably configured for an unnecessarily long duration, leading to increased slippage as the market moved.
Limit price
In this last chart, we explore an optional variable, which is having a limit price for the algo order. By simply inputting a limit price, the median edge is significantly higher compared to leaving it blank.
This can be attributed to how the algo order behaves: not executing when the price breaches the limit, and having the capability to automatically resume when the price is within the limits again. This can protect the trade against drastic short-term volatility.
Summary of Our Findings
Slippage tends to increase with lower liquidity and larger trade sizes. Algo orders provide measurable benefits in such contexts.
In general, both POV and TWAP can be effective for large illiquid asset trades, where TWAP may require more careful configurations (e.g., adequate duration) depending on your objectives and timeframe.
Setting a limit price is a recommended best practice to protect trades from short-term price fluctuations.
Users trading small volumes of liquid assets may prefer direct market orders due to sufficient liquidity and to avoid potential risks of suboptimal algo settings.
Algo orders can also offer additional benefits for large-sized trades that are not obvious in the above case studies. For example, algo orders can help conceal such trades by breaking orders into smaller parts, or allow traders to participate while mitigating market impact.
Closing Thoughts
Binanceโs algo orders enable traders to better manage slippage and market impact, especially when dealing with large or illiquid assets. However, performance depends not only on the chosen algorithm but also on thoughtful configuration tailored to each trade.
Further Reading
Disclaimer: This content is presented to you on an โas isโ basis for general information and educational purposes only, without representation or warranty of any kind. It should not be construed as financial, legal or other professional advice, nor is it intended to recommend the purchase of any specific product or service. You should seek your own advice from appropriate professional advisors. Products mentioned in this article may not be available in your region. Where the article is contributed by a third party contributor, please note that those views expressed belong to the third party contributor, and do not necessarily reflect those of Binance Academy. Please read our full disclaimer for further details. Digital asset prices can be volatile. The value of your investment may go down or up and you may not get back the amount invested. You are solely responsible for your investment decisions and Binance Academy is not liable for any losses you may incur. This material should not be construed as financial, legal or other professional advice. For more information, see our Terms of Use and Risk Warning.