MB Trading Futures - Introduction | Page 475 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Introduction and trading basics 27 replies

Introduction to Level 2 4 replies

VTTrader: How to Write Your Own VT Indicator Introduction. 0 replies

Interest Rates: An Introduction 0 replies

An introduction to spread betting 0 replies

Attachments: MB Trading Futures - Introduction
Exit Attachments

MB Trading Futures - Introduction

  • #9,481
  • May 18, 2012 2:42pm May 18, 2012 2:42pm
Quoting cts022
Perhaps we should let them focus on repairing the brokeass iOS platforms before moving on to new projects.

tl;dr: please fix iOS!
I have been crying about this for the longest. I just got a brush off statement from Justin - basically they are not aware of any problems.

I guess they all use android at the office , but people on here have said issues on andriod exists too. LOL

I know the perfect broker doesn't exist but they should at least try to do better. The iOS release is broke right from the word go. That is silly and unacceptable in this day and age.
  • #9,482
  • May 18, 2012 3:05pm May 18, 2012 3:05pm
Quoting mediaboy
I have been crying about this for the longest. I just got a brush off statement from Justin - basically they are not aware of any problems.

I guess they all use android at the office , but people on here have said issues on andriod exists too. LOL

I know the perfect broker doesn't exist but they should at least try to do better. The iOS release is broke right from the word go. That is silly and unacceptable in this day and age.
MBeTa broker!
  • #9,483
  • Edited 5:28pm May 18, 2012 5:16pm | Edited 5:28pm
I will preface everything by saying that i know friday is not the best day to trade especially after 12pm est.

Anyways i had a stop order that was partially filled with -20pips of slippage. I did get some of the order to fill near the stop location.

This was not news related (i don't think) just a sudden short squeeze. I have notice same issue but nothing this bad on order fills in recent time frames.

20 pips of slippage is pretty extreme to me ... Trade was taken in eur/usd . spreads were the usual 1 or < 1 pip just prior to the spike up.

Details:
sold short eur/usd at 1.2730 11:25 am
buy order eur/usd at 1.2757 at 3:04:28 pm (inline with my expectations)
buy order eur/usd at 1.2774 at 3:04:43 pm (this is the one i scratch head on)

The stop for this order was set at 1.2755

What are you folks thinking on this? Is this kind of fill to be expected given the context? Again this not really a news situation, i attribute it to a short covering spike.

I am wondering if i would have faired any better under at a market maker type broker or if this pay for limit vs fee exn stuff is a contributing factor.

P.S sorry about subject line of post... the hit was 20pips not 40. If it was 40 i would be on phone with this broker now.. haha
  • #9,484
  • May 18, 2012 7:36pm May 18, 2012 7:36pm
I'm surprised that MBTrading did not roll out a beta testing period for their mobile platforms, in a similar way that they did with mt4. This way, there is a smaller chance of stability issues if you roll out updates in a phased approach.
Measure trends automatically with zero lag
  • #9,485
  • May 18, 2012 8:03pm May 18, 2012 8:03pm
mediaboy: write a mail to support to make sure this is not a glitch. But stop-orders have been executed at the low/high of a spike in very fast market conditions with MB in the past. JustinLeBlang has explained how this works in the past in this thread. Basically banks refuse to execute the order and MB tries to hit the next bank, gets refused and so on and so forth. Until the market settles down. A weakness in their execution-model if you ask me.
  • #9,486
  • May 19, 2012 7:56am May 19, 2012 7:56am
Quoting mediaboy
I will preface everything by saying that i know friday is not the best day to trade especially after 12pm est.

Anyways i had a stop order that was partially filled with -20pips of slippage. I did get some of the order to fill near the stop location.

This was not news related (i don't think) just a sudden short squeeze. I have notice same issue but nothing this bad on order fills in recent time frames.

20 pips of slippage is pretty extreme to me ... Trade was taken in eur/usd . spreads were the usual 1 or < 1 pip just prior to the spike...
Mediaboy,
Not that it helps with this specific trade, but I will be compiling daily slippage information from my MBT trading and can share it with the forum. It will be a different as it will be "per microlot" normalized, but should help.

For example, Friday, May 18th had a net of -8.6pip slippage on 138 microlots traded. I had +9.4pips on limit orders (76 micros) and -18.0pips on stop markets (60 micros). It was an off day where I was nearly 1:1 limit:markets, and one 8 micro stop market was responsible for -8.8pip total (-1.1p per micro).

Anyways, it is something to track, and one of many, many variables that anyone who truly considers this a business, and life-long commitment, should follow.

Cheers!
BroncoCap

PS: Justin, I too sent you details about a similar trade w/ nasty slippage you said you would investigate...any progress???
  • #9,487
  • May 19, 2012 8:01am May 19, 2012 8:01am
Quoting shrike
mediaboy: write a mail to support to make sure this is not a glitch. But stop-orders have been executed at the low/high of a spike in very fast market conditions with MB in the past. JustinLeBlang has explained how this works in the past in this thread. Basically banks refuse to execute the order and MB tries to hit the next bank, gets refused and so on and so forth. Until the market settles down. A weakness in their execution-model if you ask me.
Actually, it sounds pretty straight-forward to me. A stop order is not guaranteed to be filled. So it would make sense that if one LP cannot fill the order, to try a different LP. edit: I do see your point in that a market order should be filled at the next available price by any of the LPs. I don't think the reaction from the trading community will be favorable if the "slippage" was 20 to 50 pips or more, which is plausable in a very fast market.

I think forex.com does this (guaranteed stops), but then there are minimum restrictions on pending (limit) orders. Like they have to be 5 pips away (or was it 50 pips) from current prices.
Measure trends automatically with zero lag
  • #9,488
  • May 19, 2012 10:52am May 19, 2012 10:52am
ybf: yes, probably not much a broker can do about this, if you set up like that. Exchanges do not have this problem - they match orders themselfs instead of sending them out to LPs and see what comes back.

The only question is how much leeway MB gives to LPs - how many milliseconds the LP has to either fill an order or reject it. If this timespan is too long i guess there are problems like this. If its something like 300ms or more its like free money for the LP.
  • #9,489
  • Edited 6:36pm May 19, 2012 6:21pm | Edited 6:36pm
Quoting BroncoCap
Mediaboy,

Anyways, it is something to track, and one of many, many variables that anyone who truly considers this a business, and life-long commitment, should follow.

Cheers!
BroncoCap

PS: Justin, I too sent you details about a similar trade w/ nasty slippage you said you would investigate...any progress???
BronocoCap,

You make an excellent point. Up until now i have had the attitude that the stop loss would be the risk control tool. Well now my attitude is different. I will pay much more attention to the slippage factor.

Plan to report this trade to MB but I doubt they will do anything about it (like shrike and ybf have said).

I come from futures and equities trade and have never experienced slippage like this. I usually do the same trade in my live MB account and also in demo Oanda. I dint do that in this trade but wish I did.

Also i think the fact that the order book is now split has hurt my fills - not sure MB is going to get much more of my business.

So much for FX being the most liquid market ..... right.
  • #9,490
  • May 19, 2012 7:46pm May 19, 2012 7:46pm
Quoting shrike
ybf: yes, probably not much a broker can do about this, if you set up like that. Exchanges do not have this problem - they match orders themselfs instead of sending them out to LPs and see what comes back.

The only question is how much leeway MB gives to LPs - how many milliseconds the LP has to either fill an order or reject it. If this timespan is too long i guess there are problems like this. If its something like 300ms or more its like free money for the LP.
It shouldn't be more than 1200ms, including the time for MBT server to check all of the LPs, get fill information, and report it back to the client workstation. The majority of that latency is reporting it back to the client workstation, as the latency between MBT <--> LPs should be pretty low, if MBT offsets some of the trades in house, the latency is actually quicker for those trades as there is less parties in the loop. during news annoucements.

An exchange like futures have independent clearinghouses to guarantee the fullfillment of the counterparty. Plus the process is standardized. Every broker is connected to the same Globex (or whatever) order routing system on a FIFO basis. And because of such a streamlined approach (from the top down; the futures broker is more of a customer service rep), it's much easier to scale those types of systems. Forex doesn't always have that consistency throughout all the brokers, because each broker "makes" the market available to them; and the interpretation of the best approach for handling orders is left largely to the broker and LPs. Luckily because of Metatrader 4 (and eventually mt5) being standard retail forex trading platform, combined with watchdog sites like fxintel, and forums like here where ideas can be shared fairly instantaneously, the self-regulation is effective enough to ensure brokers develop. Plus MBT has their prior experience and reputation to help them.

Futures and stocks has had almost a century to refine its process, whereas retail forex has only become popular over the past decade or so, and especially over say 5-6 years, had to mature quickly. Being a platform pioneer with so many variables isn't drag and drop.
Measure trends automatically with zero lag
  • #9,491
  • May 19, 2012 9:07pm May 19, 2012 9:07pm
Quoting ybfjax
It shouldn't be more than 1200ms, including the time for MBT server to check all of the LPs, get fill information, and report it back to the client workstation. The majority of that latency is reporting it back to the client workstation, as the latency between MBT <--> LPs should be pretty low, if MBT offsets some of the trades in house, the latency is actually quicker for those trades as there is less parties in the loop. during news annoucements.

An exchange like futures have independent clearinghouses to guarantee the fullfillment of the...
You are either very naive or you work for MBeta.

Mb, Oanda, FXCM and all the bucket shops out there profit from these kind of "mistakes".

Just to put in perspective a bank has to put billions of dollars in the market in order to move it a few pips. Brokers do it with a intentional platform glitch.

Wake up people! If platform issues was something bad for brokers they would have addressed it long time ago.

Big institutions have to put big money and their arse in the line in order to move the market and make money. Brokers can generate a spike with ZERO risk.

Slippage, platform freezes, spikes, etc are the goose with the golden eggs for bucket shops.
  • #9,492
  • May 20, 2012 9:52am May 20, 2012 9:52am
Quoting shrike
ybf: yes, probably not much a broker can do about this, if you set up like that. Exchanges do not have this problem - they match orders themselfs instead of sending them out to LPs and see what comes back.

The only question is how much leeway MB gives to LPs - how many milliseconds the LP has to either fill an order or reject it. If this timespan is too long i guess there are problems like this. If its something like 300ms or more its like free money for the LP.
Shrike (and others),
Agreed, ideally, any LP that puts a price into the MBT system should guarantee a fill at that price, as if they had an order in the EXN. It is essentially a "last look" provision that solely benefits the LP and hinders the retail trader. If, rather, their liquidity was on the EXN, then they could move, cancel, change orders just as retail, but the extra step of sending an retail order to the LP would be removed. (I would guarantee that spreads would widen if this was forced upon them.) Basically like an exchange...

Mediaboy, yes, I still stand by the fact that either 1) the market participants have drastically changed or 2) the split Free-EXN/PFL competing systems have changed MBT fills. Hence, my new investigation of stop orders as a way to both mitigate risk and enter trades where limits are currently going unfilled.

No matter what, in the end, a profitable trader must be sufficiently skilled to overcome the laws of probability, the negative skew of these kinds of issues, and ultimately place the burden of success on their back, not upon their broker...I do think it is possible, but I also think it takes a tremendous amount of work and dedication that many just don't possess...so then they blame their broker when in reality they are just a breakeven trader...

BroncoCap
  • #9,493
  • May 20, 2012 11:00am May 20, 2012 11:00am
Quoting BroncoCap
Shrike (and others),
If, rather, their liquidity was on the EXN, then they could move, cancel, change orders just as retail, but the extra step of sending an retail order to the LP would be removed. (I would guarantee that spreads would widen if this was forced upon them.) Basically like an exchange...
That is pretty much how the LMAX's MTF (Multilateral Trading Facility) works, or at least that is what they claim... and the spreads are actually tighter than on any of the MBT plans. I have not tested live trading with them yet but it is definitively on my list.
I'm not insane; my mother had me tested.
  • #9,494
  • May 20, 2012 12:07pm May 20, 2012 12:07pm
Quoting BroncoCap
If, rather, their liquidity was on the EXN, then they could move, cancel, change orders just as retail, but the extra step of sending an retail order to the LP would be removed
MB should seriously consider your proposal! Make the LP's quote "tradable", not just "indicative".

Quoting RoBiK
That is pretty much how the LMAX's MTF (Multilateral Trading Facility) works, or at least that is what they claim... and the spreads are actually tighter than on any of the MBT plans. I have not tested live trading with them yet but it is definitively on my list.
I never look into them seriously but you have sparked my interest on them. Thanks.
  • #9,495
  • May 21, 2012 8:41am May 21, 2012 8:41am
Here is today's slippage info.

Traded 82,000 units.

46 micro limit fills with total slippage of +9.0 pips, roughly 0.2pips per micro.

36 micro stop fills with a total of -11.8 pips, roughly -0.3pips per micro.

Net-net, a balanced day, but also a relatively quiet market.

One thing I will watch is short vs long stop slippage. Today I had 10 micros buy stop filled with only -1.4pips whereas 26 micros sell stop filled had -10.4pips.

Anyways, I'll post this for a bit, if it gets useless or clutters the thread, I will stop.

Cheers!
  • #9,496
  • May 21, 2012 8:42am May 21, 2012 8:42am
Quoting RoBiK
That is pretty much how the LMAX's MTF (Multilateral Trading Facility) works, or at least that is what they claim... and the spreads are actually tighter than on any of the MBT plans. I have not tested live trading with them yet but it is definitively on my list.
LMAX always intrigued me, but alas, as a US citizen, I'm fook'd. Maybe one day...
  • #9,497
  • May 21, 2012 8:59am May 21, 2012 8:59am
Quoting vincegata
Are Plus orders specific to MBT or other brokers also have something similar?

Thanks.
Depends on the exact order. Some brokers have something similar to some of them, but most brokers don't have anywhere near the total catalog of orders that MBT offers.
  • #9,498
  • May 21, 2012 9:20am May 21, 2012 9:20am
Quoting BroncoCap
Shrike (and others),
Agreed, ideally, any LP that puts a price into the MBT system should guarantee a fill at that price, as if they had an order in the EXN. It is essentially a "last look" provision that solely benefits the LP and hinders the retail trader. If, rather, their liquidity was on the EXN, then they could move, cancel, change orders just as retail, but the extra step of sending an retail order to the LP would be removed. (I would guarantee that spreads would widen if this was forced upon them.) Basically like an exchange...

BroncoCap...
The problem with the concept here that you are saying is that if an LP put their orders on one firm's ECN/EXN, they would be dependent on getting hit by that firm's customer base. In reality, Forex works the opposite way. Big banks that are LPs maintain their own price feed and accept orders from most of the brokers. The brokers have to maintain a high deposit with the banks in order to have the ability to route to them, and each banks has a couple of different feeds that the broker can negotiate into being able to direct to their customers.

Again, the LP would never put liquidity directly with just one broker as it would limit their ability to get filled (ultimately, this thing will move to an exchange, but that will require more regulation).

What I think is often lost in the understanding by traders is to analyze the incentive of the LPs in the system. If a broker comes to an LP and says "We are only going to use your liqudity and only route orders to you," that would be the most appealing situation for the LP obviously for a couple of reasons. First, they would always see both sides of each trade, which is important since most retail traders lose money, the bank or LP relies on (like deal desks used to) the orders coming back. Second, they would have to execute everything sent to them.

Now, that same broker says, "We're going to add a second LP." That doesn't mean spreads narrow because now, the first LP is a little less interested because they have to compete to get the order flow. In fact, I bet it means that they would now just view that broker as someone less interesting to them because they don't have control. From the trader's perspective, this should be a better system because you aren't at the whim of just one bank that is ready to screw you when they want to, but it also doesn't mean that as a broker adds more and more banks that spreads will narrow and executions will improve necessarily. While the plus is that you have more liquidity pools to access, the negative is that none of them have (from their own business model perspective) much incentive to go out of their way to be the best bid or ask all of the time and tighten the spreads because they don't know if they will see the order come back, and that's the most important piece to the LP.

I think what MBT did with their FREE EXN plan, and Justin has hinted at this, is that they had used let's say 10-20 LPs over the years and had relationships with them and knew which were the biggest and executed the fastest and most reliably. Instead of just picking one, which would essentially make them a front end for a single desk, I believe that they went to a few of their most reliable LPs and said "Come over to this new quote and there will be less banks to compete with, so you will see more order flow, but you need to narrow your spreads and compete." And they got a few of the banks to do this. This is why the spreads are tighter and why the fills on FREE EXN are faster on average (I have done some testing on this, there is no doubt that they are). I think the concept of liquidity dropping on the PFL plan is just a function of the market volumes being down and nothing more.

But the main point is that you can't expect the LPs to put their liquidity on a broker's EXN. They would have to break up their liquidity to various brokers if they did that. It's just easier for them to maintain their liquidity in-house and have brokers come to them.
FXChant Singing the World of Forex
  • #9,499
  • May 21, 2012 9:27am May 21, 2012 9:27am
Quoting fxchant
The problem with the concept here that you are saying is that if an LP put their orders on one firm's ECN/EXN, they would be dependent on getting hit by that firm's customer base. In reality, Forex works the opposite way. Big banks that are LPs maintain their own price feed and accept orders from most of the brokers. The brokers have to maintain a high deposit with the banks in order to have the ability to route to them, and each banks has a couple of different feeds that the broker can negotiate into being able to direct to their customers.

......
fxchant,

No, I agree 100% and you are correct in everything stated here, namely how MBT is currently set up. The idea of having executable LP prices on the EXN is mostly a pipe dream...but an interesting proposition. As I stated, if an LP were to do it, I'd expect spreads to widen further as LP risk increases.

Great post, thanks for taking the time to write.

Cheers!
  • #9,500
  • May 21, 2012 9:46pm May 21, 2012 9:46pm
Quoting fxchant
The problem with the concept here that you are saying is that if an LP put their orders on one firm's ECN/EXN, they would be dependent on getting hit by that firm's customer base. In reality, Forex works the opposite way. Big banks that are LPs maintain their own price feed and accept orders from most of the brokers. The brokers have to maintain a high deposit with the banks in order to have the ability to route to them, and each banks has a couple of different feeds that the broker can negotiate into being able to direct to their customers.

Again,...
You are wrong in your theory.

The only "incentive" a bank has on MBeta is because of their bulk of deposit. They don't give a s**t about MBeta's order flows because it is something too small for them to bother. Unless their LP is a very small bank.

On any MM the orders are offset internally, bulk orders long against bulk orders short. That process is managed by sophisticated algorithms. Only when unbalances happens the LP kicks in which is very rare. The algos will freeze the platform or resort to other gimmicks instead.

LP cotes work as a proxy between the real inter-bank market and the " virtual" market Market Maker make.

MBeta could pick any price for bid and ask does not bother the bank (because as a explained above the bank does not actually participate in the trade) it only tells you how much you are paying the broker.
11 traders viewing now