Guru's Verification engine ensures consistency, confidence, and trust in the knowledge your organization shares. Learn more.

LP Measures

What is important in an LP can vary between time of day, or different market states e.g. around unscheduled news.

It's important to assess the feeds LPs are providing you and also in turn the feeds you are providing your clients.

The below measures are available in real-time Compass via the Trading Dashboard and also in Pricing Simulations

Traditional Measures

We publish some of what we like to call traditional measures that are useful to assess LPs by. They are still very relevant and useful measures.

  • Spread
    • what is the $/M charged to execute via them. Does a single venue have the best spreads across all instruments? How do I check and incorporate it across all instruments? Can I volume weight which LP is giving me my best pricing according to my client trading volumes (i.e. the ones that matter to me)?
  • Spread at Depth
    • The top of book spread may be nice, but sometimes the quantities displayed in these numbers mean that almost any meaningfully sized order has to consume liquidity from lowed down in the stack.
  • Fill Ratio
    • there are countless times when spread ends up being ephemeral. Any attempts to hit the quote are met with a cancel and an additional cost of replacing the order when the price has moved.
  • Depth
    • Particularly in B2B scenarios clients need to feel they can get a large order done with you. Who is showing you depth is important, particularly changes to that period on period and through differing market states.
  • Slippage
    • The difference between the price published and the price an LP fills at. Further exploration here
  • Availability
    • For how long during the week does an LP not give you prices? Does this mean shorter trading weeks? Trouble with the market open, or even around news? How can price discovery be done if all LPs withdraw their rates?


Predictive Measures

We're very into predictivity, it's a great indicator for who's important to listen to, particularly in liquid timezones.

  • 1 Second predictivity
    • who's the best LP for predicting where the price will be in 1 second time (A good candidate for No Arbitrage Pricing Node)
  • 30/60/180 second predictivity
    • The predictivity of LPs can span further ahead than some people think. Sources of this can be passive hedging large orders / inventory, machine learning models sending skew to market etc.
  • 1 Second laggardness (new)


Alternative Measures

Whilst predictivity is great in liquid timezones. Sometimes businesses and price feeds may prioritise other considerations depending on who the feed is intended for and time of day. If for instance a large spread is being made during an illiquid time of day (providing plenty of protection), it may be better to prioritise price stability during such time and minimise anything that unnecessarily trips stop losses, take profits or margin stop outs.

  • Opinion
    • how far in $/M terms is an LP from its peers. Are some LPs more opinionated than others. Opinions are good, where they are right, its good to listen to, where they are wrong, they can be good sources for risk clearance.
  • Variation/Flappiness
    • what is the average change in price in $/M terms there is from an LP. Are they flappy? Are they unnecessarily moving the price (increasing the risk of hitting client SL/TP/SO limits)
  • Chattiness
  • Arbitrage Counts
    • it may not always be executable. But how often and in which instrument is an LP arbitrageable. If someone is always arbitrageable and they are incorporated in aggregation, you may be slowing your rate formation down. The other side of the coin is that this may indicate opportunity for prop strategies and provide another resolution of visualising during the trading week, on when they may appear.

You must have Author or Collection Owner permission to create Guru Cards. Contact your team's Guru admins to use this template.