Mostly… fixed income and cross product eTrading

July 29, 2007

Algos and FIX

Filed under: Algos, FIX — holky @ 3:57 pm

Following on from previous post about algos and FIX in Europe, Finextra shows FPL take another step forward in standardising the protocol for algorithmic order types… So far 15 industry participants, including six of the world’s largest broker-dealers, have particpiated in testing of the new language. Each firm has published samples of their algorithmic trading strategies in the new XML format, ensuring that all their current and near-term algorithmic trading strategies can be expressed in the new format.



  1. Full info on this including xml, xsd, xslt and code examples, plus slides and audio presentation:

    Comment by Rick — July 31, 2007 @ 4:01 am

  2. Vicky Pollard moment. Yeah but, no but…how many INSTALLED buy side OMS systems can use this stuff. Not the next generation version, but actually the ones that are in use day in and day out????

    Not a lot I wager.

    Comment by John Greenan — August 3, 2007 @ 11:34 am

  3. John – you are 100% correct, there are precisely zero major OMS players that have this available for customers today. It’s still in Beta testing.

    I can however assure you that the .xml has already been deployed inside at a few BDs for PRODUCTION USE on their own proprietary trading desks. Plus, with 16 aggressive sell-side firms behind it you can be sure this will move forward as rapidly as possible!

    The technical WG is right now working with Fidessa, Charles River, Latent Zero, Marketcetera and other OMS vendors on dropping these .xml files into OMS systems and not driving the rest of established systems crazy. The goal is to have new algo strategies ready to trade in minutes.

    As we work out the final 10-20% of the details this fall other OMS vendors and in house groups are more than welcome to provide last minute needs, try it out, etc. The final standard is not expected to ship till the end of 2007. We expect that the sell side will be able to have virtually ALL of their algo strategies run up in the new xml format within WEEKS of having that final standard.

    Smaller, hot shot trading firms will most likely have this up and running ahead of the larger OMS players, perhaps even before the final version of the standard ships. At the 8/15 press conference you could see a live software demo of the functionality – you drop in the .xml file, up pops the new order ticket, ready to trade that strategy. The trader enters the order, all the rules are available for that to validate locally, including help text on each parameter. Press send and out the newOrderSinge or newOrderMultiLeg goes in plain old comfortable FIX 4.0-4.2 format, right ready for a standard FIX wire connection. It would not take much to go from that demo to a live system provided it didn’t have to integrate with a bunch of other stuff. As long as you had the capability to receive back FIX fills in standard FIX 4.0-4.2 format something basic could be ready for a trader to use in a live environment in about 90 days, perhaps less.

    The thing that will take the most time during the beta testing is insuring it will work well with standing OMS systems. As long as you don’t mind a “pop up window” that brings up a new algo order screen over your standing OMS, that doesn’t talk to that OMS, doesn’t pre populate the order entry ticket with data from the OMS, doesn’t necessarily “look and feel” exactly the same as the rest of the OMS, doesn’t hold your hand in filling out the entire order, and doesn’t present 100 different algo strategies in precisely the same field order, position, look, etc, then you can get all the functionality from this new standard in a jiffy.

    All the basics have been worked out for months. We are just arguing over what bells and whistles to include. Things like screen widgets that appear and disappear as a trader works through a complex new algo trading type, dynamic screen behavior, etc. Plus, we need to have the fixed income guys give it a spin to be sure we aren’t leaving out 1/32nds, 1/64ths and other stuff like that for them.

    Don’t hold your breath on multi asset strategies just yet (e.g. cash bond basket and a bond future) – the new standard will easily enable them but most Sell Side systems will need some significant rearchitecting to take in a complex multi asset algo strategy, get it properly routed to the various execution “desks”, get it done, then integrate the fill and get back to you in an integrated format. But in time, that will come too. Order tickets on traders desktops will not be the hold up. We can render just about anthing.

    So, you definitely win the wager that it’s not ready to go off the shelf in shrink-wrap software packages today, but it is pretty near. That’s why its news. If you are on the tech side now is the time to start learning this stuff in depth.

    If you are a trader be sure to tell your OMS system vendor or your in-house staff – I want to be able to trade any algo, from any sell-side firm, at any time, just as soon as they are available! Don’t expect me to be in competitive markets trading with strategies that are six months to a year old when the guy up the street has new, up to the minute algos, popped in his system overnight and ready to trade at the open the next day.

    Plus, by knowing just a few months early you can make the comment all traders love to say when something “new comes up — “tell me something I didn’t already know!” LOL!


    Comment by Rick look me up at — August 21, 2007 @ 1:43 am

  4. A counterview (combined with a bit of devils advocate)

    Dynamically rendering a GUI from XML is not news. This is something I have worked on in the distant (three or four years ago) past. Microsoft call this XAML I think and there’s an open source version as well. It’s really not anything new to learn for most developers who work on client side user interfaces in modern technical platforms.

    Getting all of the moving parts lined up to work on a production buy-side OMS is news…

    Let’s say that you work for a buy-side running LatentZero – perhaps v3.1 or 3.51. You are planning an upgrade to version 4.0. This functionality is not (as far as I know) contained in v4.0. So it’ll be in v5.0. That’s two years from now before it goes into production.

    I cannot remember the version numbers for Charles River – I think it’s versions 8, 9 and 10 that are comparable but the time frame is similar.

    So, two years for delivery? Not quite so hot news is it?

    Press releases, demos count for nothing. Show me the money. “It would not take much to go from that demo to a live system provided it didn’t have to integrate with a bunch of other stuff” Welcome to the real world – where trading systems indeed do have to integrate with a whole load of other stuff…

    I’m not knocking your endeavours – I am glad that someone with enthusiasm is driving this from within FPL land (I’ve been involved with FPL since 2001/2002 and it can be slow to move – but that’s due to the institutional structures in place rather than the people).

    Anyhow, good luck with it!

    Comment by John Greenan — August 21, 2007 @ 6:52 am

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: