Forums  > General  > CME's Market by Order rollout  
     
Page 1 of 1
Display using:  

HockeyPlayer


Total Posts: 125
Joined: Nov 2005
 
Posted: 2017-07-26 23:49
Over the last 6 months, CME rolled out Market by Order, giving you the size & priority of each order at a price level.

That appears to have increased the total quantity and number of orders at the inside. For example, in crude the total quantity went from ~40 to ~90 and the number of orders went from 20 to 40.

Why?

algos.arts


Total Posts: 5
Joined: Jun 2017
 
Posted: 2017-07-27 04:37
Assuming you are saying only inside has increased.

It is still ~2 per order based on your statements.
So, more players?

More Infos == More Knowledge => More Strats/Wagers at the moment ???


EspressoLover


Total Posts: 254
Joined: Jan 2015
 
Posted: 2017-07-28 07:37
Think of it from the perspective of a market maker. With an order based feed you know your position with certainty. (Just tag your order with it's MDOrderPriority tag at arrival on the feed). In a level based feed you have to estimate your queue position. That's difficult, messy and involves a fair degree of uncertainty.

If you're trying to manage the adverse selection of your quotes, knowing queue position is really important. Especially in thick-book instruments (like most of the liquid CME futs). Less uncertainty means less adversity. Less adversity means that liquidity providers can quote larger, quote more often and be trigger-shy about cancelling.

In practice, most level-based market makers usually assume something close to worst-case queue position. Being overly optimistic is a quick recipe for being picked off. Most cancels occur in the back of the queue anyway. So worst-case is pretty much the only tenable approach.

But the thing is, even if worst-case is the modal case, some non-negligible proportion of the time you actually do better. So a lot of orders end up cancelled, that otherwise wouldn't be given transparent information. Giving market participants knowledge of their queue positions results in less cancels. Which means more orders alive at any given time. As a second order effect, less spurious cancels increases the chance that an entered order gets executed at an advantageous position. That increases the expected profitability of the marginal order. That should push the break-even back-of-queue size out further.
Previous Thread :: Next Thread 
Page 1 of 1