I recently started trading one of my automated NT strategies to two different brokers and discovered an amazing difference. In simple terms, I am getting significantly different fills. Now I always was a skeptic of the importance of the quality of fills as I hold positions an hour or two looking for a good run. I thought it could not matter if I got in at a fraction of a better price. But when I saw the consistency of the difference I realized why it is so important. It is risk free money, something you can't normally get in trading!
I will give a concrete example. If I trade one contract of TF for a year I will make $20,000 to $25,000 in that time. It will also take about 500 trades. With a fill difference of 0.1 I will make (or save) an additional $5000. That is 25% of my total net return!
I was getting good commissions at Interactive Brokers, and I still have good commissions, but now at Optimus I have seen between $5 to $10 better profit per trade. The first few days I kept checking that identical orders had gone out to both brokers at the same time. Even if the order to Interactive Brokers is queued first, every single time it will be a better fill at Optimus. It does vary, as I stated, but it doesn't go first a little better at IB and then better at Optimus, it is always better at Optimus even if the amount is less.
The other thing to keep in mind is that these are limit orders, not market. I expect with market orders you may see more range of fills, where IB is sometimes better simply because the market changed by the time they executed the order. The importance of the limit order is that it shows the actual delay and quality of the execution better than a market order. A limit order is really a market order below/above a set price. So if you are first in line at that price you get a better fill.
At any rate I am transferring more of my accounts over to Optimus.
The following 6 users say Thank You to tradetree for this post:
interesting... here are a few flaws with the argument... position in the queue determines when you get executed for limit orders... that is very much true.. however, IBKR might internalize orders first and then pass on to the exchange.. which is why you might see that "delay"...
but it isnt really a delay on execution, and not sure how you are getting different fills if the orders are limit orders... if you enter a limit a 85.75 at either place, you get filled at 85.75... position on the queue only has to do with if you actually get filled or not.. so if there are 1000 ppl in front of you selling/buying.. but only 999 trade on that price when you are 1000, then you dont get filled..
anyhow.. can you post some actual examples? transaction logs, etc? that will support your argument better IMO and will be of more use to all of us..
The following user says Thank You to sysot1t for this post:
1) In terms of a delay or not, that is more an "explanation" for the symptom rather than a cause. In other words, the fact is the fills are poor at IB, and one way I explain the bad fills is in terms of delays. It appears you agree that IB may delay them by "internalizing" them before passing them on to the exchange. That would be a bad process for the trader.
2) Limit orders are guaranteed to either be filled at or better than the price of your limit. My orders are filled with Optimus / Vision at a consistently better price than my fill price. IB gets me within my limit but not quite as good as Optimus.
3) I can certainly share concrete examples. I have nothing to gain by promoting one path of execution over another. I was just pleasantly surprised! I don't mean any offense, but I really don't have anything to prove. If you don't believe it that is fine with me. I don't plan on sharing my brokerage statements as I feel those are private. But I would welcome scrutiny on what is taking place. I will certainly look more at the delta of limit vs. filled prices and share those as I find time.
Finally, this may clarify the way I look at the different fills. My algorithm is looking for overbought / oversold conditions, as any good system does. These inefficiencies in the market are what give us our "alpha". From the time I detect the condition, the market is in a higher level of volatility; it is ready to turn. I put my limit at a price that is within the average true range of the bar. It is not like I am putting an order out there for 20 minutes waiting for the price to reach it. I adjust limit prices on a minute by minute basis. So timing becomes much more important. I want to catch just the right condition. Delay in this case could explain why I see better fills with one FCM over another. - I don't know if it is because of delay, but that is the best explanation for the symptom. I simply see better round trip profit!
Yes, the fill prices are different. I check the order that went out in the "orders" tab of Ninja. They both went out at a set limit price, the same price. But my gain is different by 0.02 to 0.05 (times 2 for round trip), which is $4 to $10 on the TF instrument. They are both within the limit, but one is better.
I do agree that this deserves more scrutiny. What I need to do is to take an example and look at the brokerage statements the next day to ensure that this is reflected and not a Ninja aberration. My original test was easier to track than it is today. I had two accounts with the single same strategy running. Since then I have added a second contract to one of the two accounts. What I could do before was look after two weeks of trading and see which one had a higher bottom line gain.
I should also mention that this is an interesting event that I came upon and it was evidenced by the "net gain" posted within Ninja for the daily account gain. This is not a double-blind scientific experiment. It was as close a comparison I have ever seen for execution. I also happen to know through experience that IB data is not as high quality. But in this particular case I use the higher quality data feed for both accounts trading (ie not IB's).
Finally, the margin requirement difference is enough to get me to move accounts over to Optimus. IB requires $10,500 per TF, while Optimus requires $8,500 per TF for overnight and $4,250 for daytrading per TF.
one, IB data not high quality... nothing new really.. not to be used for charting, that is more than a well known fact...
two, is higher marging really a bad thing? I guess it depends on ones risk management and the account size... leverage is a two way sword, and I dont blame an FCM for wanting to protect oneself... oh btw, IB does have portfolio margin, so if you are big enough, that wouldnt be a concern...
my 2 cents.. just a different side to your argument, nothing more..
typical, raise an argument with opinions and yet dont want to provide accurate test case data...
first... when you say that VFM(the FCM from optimus, and I assume you are using Rithmic) gives you better fills than IBKR and how you are moving your account fully over to VFM, you are promoting one over the other...
second, yes.. you have nothing to prove.. but then why raise the argument and make the point with an opinion? which is all you have unless you actually provide some comparison of data...
third... I dont recall asking for your statements, so please let's not hide behind the "privacy" issue...
all I said was, show us the transactional differences you are referring to.. provides us with an example.. this could be as simple as take 2 transactions from each statement and state where your price was, how often you moved your limit order, etc.. and document your argument... really, that simple.. no need to make things complicated.
The following user says Thank You to sysot1t for this post:
Your post makes no sense. With very few exceptions, a limit order gets filled at the limit price or not at all. Orders further up the queue are more likely to be filled if there's an imbalance of buy/sell orders at the limit price.
But you claim to be consistently getting positive slippage on your Optimus limit orders. I don't buy it.