Rule 2-43 and FIFO - many EA needs to be rewritten

New comment
 
Anyone aware that the new rule 2-43 from NFA concerning the first-in first-out principle of orders processing means that we even have to consider how our same currency pairs orders ought to be closed? for example if you have a series of long orders for eurusd, u can close the first order first, followed by the second earliest order. In other words, it is not just about executing orders in opposing direction only.
 
ronaldosim:
Anyone aware that the new rule 2-43 from NFA concerning the first-in first-out principle of orders processing means that we even have to consider how our same currency pairs orders ought to be closed? for example if you have a series of long orders for eurusd, u can close the first order first, followed by the second earliest order. In other words, it is not just about executing orders in opposing direction only.

See 'Offsetting transactions - NFA rule 2-43(b)'


The position with this becomes more and more confusing. On the one hand, I have it on very good authority, from someone who's seen an advance version of it in action, that MT4 will handle the new rule by simply refusing to place a new short order if you're already long (or vice versa). On the other hand, a couple of brokers have sent out very optimistic emails claiming to have a work-around which will allow things to continue as normal following the rule-change. It's possible that Metaquotes are beavering away producing a more elegant, revised solution to the problem.


Separately, I've had a broker claim that orders can still be closed in any order, and that stops etc are still allowed to apply to individual orders rather than operating on an order-independent, FIFO basis. However, that opinion isn't 100% reliable.

 
jjc:

See 'Offsetting transactions - NFA rule 2-43(b)'


The position with this becomes more and more confusing. On the one hand, I have it on very good authority, from someone who's seen an advance version of it in action, that MT4 will handle the new rule by simply refusing to place a new short order if you're already long (or vice versa). On the other hand, a couple of brokers have sent out very optimistic emails claiming to have a work-around which will allow things to continue as normal following the rule-change. It's possible that Metaquotes are beavering away producing a more elegant, revised solution to the problem.


Separately, I've had a broker claim that orders can still be closed in any order, and that stops etc are still allowed to apply to individual orders rather than operating on an order-independent, FIFO basis. However, that opinion isn't 100% reliable.

jtc/ thks for the insights; looks like given the many uncertainties, not many ppl realised that their EAs may be in danger of not working like before cos it is not about hedging on the same currency pairs now which we can just avoid but it is about the FIFO; and with different brokers giving different answers, it looks like a Y2K for MetaTrader EAs from 15 May 2009, and we will see many ppl with unattended EA having their accounts wiped out!!!

 
jjc:

Separately, I've had a broker claim that orders can still be closed in any order, and that stops etc are still allowed to apply to individual orders rather than operating on an order-independent, FIFO basis. However, that opinion isn't 100% reliable.

I've just had this opinion reversed by the broker who gave it to me. He's now saying that stops etc do operate on an order-independent basis. Let's say that you place two long (or short) orders of identical size, and the second one has a tighter stop-loss than the first one. The first order will be closed at the second order's stop. It's basically imposing an order-based regime rather than a position-based one, comparable to something like trading futures on the CME.


If this is correct, it has the potential to break a huge number of EAs - far, far more than just grid systems. Anything which has more than one order open at a time potentially contains assumptions in its code which will cause it to misbehave, even if the orders are in the same direction. Very, very glad that we've already shut down our NFA-regulated accounts...

 
jjc:

I have it on very good authority, from someone who's seen an advance version of it in action, that MT4 will handle the new rule by simply refusing to place a new short order if you're already long (or vice versa) [...]

As of this morning, my IBFX demo account is indeed now generating "Trade is disabled" errors if you try to place a sell order when already long (or vice versa). Also doesn't allow pending orders in opposite directions.

 
jjc:

As of this morning, my IBFX demo account is indeed now generating "Trade is disabled" errors if you try to place a sell order when already long (or vice versa). Also doesn't allow pending orders in opposite directions.

Yup, I have tried he demo acct at IBFX. But one thing that is not implemented is the FIFO processing orders. If I execute a few buy orders on the same curency pair, I can close any order I like.

 
ronaldosim:

Yup, I have tried he demo acct at IBFX. But one thing that is not implemented is the FIFO processing orders. If I execute a few buy orders on the same curency pair, I can close any order I like.

From what I hear that is coming, but brokers have been given a little more time to implement it. The source for that isn't quite 100% reliable, however.

New comment
Reason: