Volume Splitter

[BlowFish 10]

These are not shown (though in some markets a market maker or specialist can see these)

 

Please,can you say which Market(s) are you talking about?

 

The first that springs to mind is the NYSE which is a specialist market. Richard Ney talks a lot about specialists, he was on a bit of a mission to expose the advantages that they have though.

[cunparis 10]

T

These indicators are extremely inaccurate even if you have the best data feed in the world based on the fact that its logic is flawed. First see example above how the same trade by two traders can give you opposite results and I give you another example: 200 bid by a smaller trader, 800 offered by a larger trader. Smaller trader buys 600 at the offer, then the larger trader enters a limit order for 1000 to sell on the bid, thereby taking out the bid, so 800 remain on the offer. Another 200 are bid below that by the smaller trader. Now assume the same thing happens again. So now according to the delta indicator 1200 were bought and only 400 sold even though the smaller trader bought 1200 and is willing to buy only another 200 while the larger trader trader sold 1200 and is willing to sell another 800. If you really want to know what's going on then you should watch the order book.

 

I've been wanting to study this example for a while now and finally got to it. It's a great example and I thank you for sharing it

 

I wrote it out on paper so it's not easy to share but at the end I have:

 

DOM
--- 200
--- 200
--- 800
200 ---

Ladder
--- 600
200 600
200 ---

 

So the delta for the ladder is 400 sold and 1200 bought, which is in agreement with what you said. I'm happy to have gotten that right since I'm still new to the delta concepts.

 

You distinguish between a large & small trader. Anyone can put any size on the book so I think the only thing we can know for a fact is the trades that executed. In your example the small trader was more aggressive when buying 600 at the offer which is why the delta shows a net of +800. The delta is showing that the buyers are more aggressive and that seems to be what happened. I think the logic is ok here.

 

If I understand you correctly, you said the logic is flawed here but I think that depends on how one uses the delta. A positive delta doesn't necessarily mean price is going up.

 

There is the fact that your larger trader is prepared to sell more and you imply that this is bearish and in contradiction to the delta which is relatively bullish. So I agree with you that in your particular example there are more aggressive buyers yet price isn't likely to go up. This is something I've observed in my trading. If I see buyers coming in on the footprint and price is going up that means those buyers are buying into supply and if the price is going lower that means the supply is cutting off the buyers. This is indeed bearish and it's a setup I've been practicing. I then look for aggressive shorts to hit the bid and that's when I enter, knowing the supply is behind me protecting my stop.

 

Is the logic flawed? I think that depends on how one is using the delta. In this example the delta is showing an accurate depiction of what happened. But one has to interpret this delta in the context of the current price movement to know if this is bullish or bearish.

 

I'd love to hear more about your suggestion to study the book. The challenge I have here is there are fake orders on the book that get pulled. I understand that is designed to trick people into leaning on these orders and then the orders are pulled while the trader (sometimes called "flipper") goes short. The unsuspecting trader leaned on orders that weren't really there and is taken out.

 

I've been observing this lately but it's not easy for me to know when there are fake orders and when the orders are real. To be honest so far I find the DOM more of a distraction but I'm determined to learn more about it and how to see through the tricks. If you have any suggestions or resources for doing this I'd be very grateful.

 

Thanks for your example, it was fun to think it through.

[cunparis 10]

The first that springs to mind is the NYSE which is a specialist market. Richard Ney talks a lot about specialists, he was on a bit of a mission to expose the advantages that they have though.

 

I read a resume of Ney's ideas on the web, it was really interesting. I've known the specialists had a huge advantage but didn't really understand to what extent. I took a mechanical stock trading system that I trade and I ran a backtest for Nasdaq & NYSE. Surprise, the NYSE results were not as good as the Nasdaq results. The nasdaq stocks had a higher profit factor and much less drawdown.

 

Not only that but in my real trading I've been observing terrible fills on the NYSE stocks. For example a stock gaps up and I have a limit order to short at yesterday's close. The price will come down, fill my order (closing the gap) and then go back up. I know the specialist isn't really after my tiny order but I'm sure others have orders like mine and he's filling them at the worst price for us.

 

I seriously consider just trading nasdaq stocks. I also think if these specialists have such an edge we should be able to detect their activity and trade with them, profiting from their manipulation.

[BlowFish 10]

I should re-iterate that perhaps the biggest problem is with how tradestation and multicharts implement certain functions (asynchronously). This is before you even consider how you are going to utilise the tools.

[AgeKay 10]

If I understand you correctly, you said the logic is flawed here but I think that depends on how one uses the delta. A positive delta doesn't necessarily mean price is going up.

 

Is the logic flawed? I think that depends on how one is using the delta. In this example the delta is showing an accurate depiction of what happened. But one has to interpret this delta in the context of the current price movement to know if this is bullish or bearish.

 

You're obviously smart and you've put a lot of thought into this so yes, the logic is not flawed if you fully understand what's really going on in the market, which you do. I just know there aren't many people like you. They think "buy if delta positive, sell if delta negative". But of course, it's never as simple as that.

 

 

I'd love to hear more about your suggestion to study the book. The challenge I have here is there are fake orders on the book that get pulled. I understand that is designed to trick people into leaning on these orders and then the orders are pulled while the trader (sometimes called "flipper") goes short. The unsuspecting trader leaned on orders that weren't really there and is taken out.

 

I've been observing this lately but it's not easy for me to know when there are fake orders and when the orders are real. To be honest so far I find the DOM more of a distraction but I'm determined to learn more about it and how to see through the tricks. If you have any suggestions or resources for doing this I'd be very grateful.

 

I base all my trading decision based on what I see in the DOM and visualizing the individual depth changes had definitely helped and I can see whether a large order is real or just there to push the market. But I don't think they really use them to screw over traders leaning on price, the risk/reward is not worth it. What follows is what I understand from watching the Bund, so other markets might be completely different. Many fake orders are entered using an algorithm that quickly enter a random number of contracts until the desired number of contracts appear. For example, it goes +34, +76, +22, +89, etc. which of which the sum then (surprisingly or not) ends up to be exactly 500, 1000 or 2000. This then usually get pulled very quickly without even trading or upon the first trade. I think they enter it using smaller contracts to avoid the risk of getting completely filled if they just flash 1000. Sometimes I've seen it happen, where some guy flashes 1000 a couple of times to push the market one way and another trader who then just waited for that took at the entire offer as soon as he flashed his 1000 again. He must have used a special software though because it got taken out before I could even see the offer with my eyes. I could only see in my software that it did happen. If you have two large traders like that, it will be the one with the most money who wins. Just last week, I saw the market going down to an important support, as soon as that price went offer, all offers were swooped up. This kept going until 12k contracts traded at that price alone without the inside market even changing which is a lot in the Bund for just one inside market. So this looked very bullish and I went long. It traded down the next 5 ticks for another 12k contracts (I got out after it went 1 tick against me). Then it reversed and went straight up like 30 ticks. You obviously never know these guys risk tolerance or profit targets. The long guy might have been unlucky that momentum was against him and he might have cut his losses which might have the reason you saw another 12k contracts trade on the next 5 ticks. Or, he might have been a massive trader and he might have just bought another 12k making sure that it wouldn't drop much further against him and thinking that the shorts would take profits if they see so much more buying and loss of momentum. If he did that, then he obviously cleaned up that day.

 

Sometimes it's just large spreaders that put of huge bids or offers hoping to get filled while the opportunity is there for them. If it disappears they pull them because it's not favourable for them to get filled any more. I pay a lot of attention to the Bobl because it influences the Bund so much. You might see one guy offering huge in the Bobl just to get his bid filled in the Bund. He knows that momentum is down and that he is unlikely to get filled on his fake order in the Bobl, but he does want to make sure he gets his bids in the Bund filled.

 

Sometimes outright traders just enter huge bids or offers to get a feel for the market or fool others. They offer big to pull it on the first trade. Make it look like a fake offer while in reality they want to get filled. Or they offer just to see if anyone wants to buy. If no one does, then it confirms their position.

 

Sometimes outright traders just enter huge bids or offers push it over the edge. Sometimes a big offer is all a market needs to make the stuck longs puke which then cause a quick long covering frenzy. They don't necessarily want to get filled on that big offer, but the risk/reward might be worth it if the market cracks a few ticks because of that offer.

 

Sometimes a guy is long 1500 in the Bund and it goes a couple ticks against him. He is still bullish but wants to hedge it just in case by selling 500 in the Bobl because there might be a big bid in the Bobl and he doesn't want to drive the Bund against him. If he times it right, he might even push the Bund a few ticks his way when his covers his short position in the Bobl.

 

I could go on for pages, there is so much going on and there could be so many reasons. It's all psychology and who has the most money.

 

I now just pay attention to how the depth changes and how much trades and not necessarily how much is bid or offered.

[cunparis 10]

I should re-iterate that perhaps the biggest problem is with how tradestation and multicharts implement certain functions (asynchronously). This is before you even consider how you are going to utilise the tools.

 

I use the Ninjatrader DOM for trading. I'm not sure if there are any issues with it.

[BlowFish 10]

I use the Ninjatrader DOM for trading. I'm not sure if there are any issues with it.

 

Not that I am aware of. The issues I mention are documented earlier in the thread and are largely due to the ELA functiions insidebid and insideask. They return the current best bid and ask rather than the one that was in force at the time of the tick you are processing, This makes them prone to race conditions.

 

Random thoughts 38.... I have always thought it would be pretty interesting to analyse pulled orders versus filled orders. Seems to me that when you get directional movement it is as often as not due to orders being pulled as it is the dominant (aggressive) side 'consuming' them.

[paolfili 10]

First of all Thanks.

Not so much traders share methodology and suggestion in such a free way.

So,if you, want, let' s focus on an analysis method for the Bund's DOM (and Bobl's DOM).

Some questions arise from your statements.

Are you sure your bid/ask feed is not aggregated? (i.e. IB aggregate data in 300ms windows)

Are you sure your collect all bid/ask data? (i.e. Zenfire UDP feed CAN loose data)

I assum the reply will be positive and continue.

 

i)For example, it goes +34, +76, +22, +89, etc. which of which the sum then (surprisingly or not) ends up to be exactly 500, 1000 or 2000

This is a "black box algorithm searching" edge.

Despite your (positive) experience an analysis in "Exploratory Data" context is probably useful.

 

ii)On how many months of "good" trading experience is your method based?)

 

ii)Can you detail your trade managment rules (Ex. TP+5/SL-2).

 

iii)Can you add some words on the "big traders plays" you note in the DOM and in particular have you designed a "scanner" software for this plays?

 

Thanks

 

 

You're obviously smart and you've put a lot of thought into this so yes, the logic is not flawed if you fully understand what's really going on in the market, which you do. I just know there aren't many people like you. They think "buy if delta positive, sell if delta negative". But of course, it's never as simple as that.

 

 

 

I base all my trading decision based on what I see in the DOM and visualizing the individual depth changes had definitely helped and I can see whether a large order is real or just there to push the market. But I don't think they really use them to screw over traders leaning on price, the risk/reward is not worth it. What follows is what I understand from watching the Bund, so other markets might be completely different. Many fake orders are entered using an algorithm that quickly enter a random number of contracts until the desired number of contracts appear. For example, it goes +34, +76, +22, +89, etc. which of which the sum then (surprisingly or not) ends up to be exactly 500, 1000 or 2000. This then usually get pulled very quickly without even trading or upon the first trade. I think they enter it using smaller contracts to avoid the risk of getting completely filled if they just flash 1000. Sometimes I've seen it happen, where some guy flashes 1000 a couple of times to push the market one way and another trader who then just waited for that took at the entire offer as soon as he flashed his 1000 again. He must have used a special software though because it got taken out before I could even see the offer with my eyes. I could only see in my software that it did happen. If you have two large traders like that, it will be the one with the most money who wins. Just last week, I saw the market going down to an important support, as soon as that price went offer, all offers were swooped up. This kept going until 12k contracts traded at that price alone without the inside market even changing which is a lot in the Bund for just one inside market. So this looked very bullish and I went long. It traded down the next 5 ticks for another 12k contracts (I got out after it went 1 tick against me). Then it reversed and went straight up like 30 ticks. You obviously never know these guys risk tolerance or profit targets. The long guy might have been unlucky that momentum was against him and he might have cut his losses which might have the reason you saw another 12k contracts trade on the next 5 ticks. Or, he might have been a massive trader and he might have just bought another 12k making sure that it wouldn't drop much further against him and thinking that the shorts would take profits if they see so much more buying and loss of momentum. If he did that, then he obviously cleaned up that day.

 

Sometimes it's just large spreaders that put of huge bids or offers hoping to get filled while the opportunity is there for them. If it disappears they pull them because it's not favourable for them to get filled any more. I pay a lot of attention to the Bobl because it influences the Bund so much. You might see one guy offering huge in the Bobl just to get his bid filled in the Bund. He knows that momentum is down and that he is unlikely to get filled on his fake order in the Bobl, but he does want to make sure he gets his bids in the Bund filled.

 

Sometimes outright traders just enter huge bids or offers to get a feel for the market or fool others. They offer big to pull it on the first trade. Make it look like a fake offer while in reality they want to get filled. Or they offer just to see if anyone wants to buy. If no one does, then it confirms their position.

 

Sometimes outright traders just enter huge bids or offers push it over the edge. Sometimes a big offer is all a market needs to make the stuck longs puke which then cause a quick long covering frenzy. They don't necessarily want to get filled on that big offer, but the risk/reward might be worth it if the market cracks a few ticks because of that offer.

 

Sometimes a guy is long 1500 in the Bund and it goes a couple ticks against him. He is still bullish but wants to hedge it just in case by selling 500 in the Bobl because there might be a big bid in the Bobl and he doesn't want to drive the Bund against him. If he times it right, he might even push the Bund a few ticks his way when his covers his short position in the Bobl.

 

I could go on for pages, there is so much going on and there could be so many reasons. It's all psychology and who has the most money.

 

I now just pay attention to how the depth changes and how much trades and not necessarily how much is bid or offered.

[AgeKay 10]

Are you sure your bid/ask feed is not aggregated? (i.e. IB aggregate data in 300ms windows)

 

Well, bids or ask are never aggregated but I guess what you mean is that you don't see all depth changes, which can certainly happen because you only get the current snapshot of the order book when you request it. I explained the reason for this above, you usually don't want stale order book data. The X_Trader API offers a subscription method that guarantees that all trades are received but they might still be aggregated if your broker or exchange chooses to do so, can't do anything about that, but I don't see at as a problem. It doesn't really matter whether someone sells 2 x 100 or 100 x 2, it's still the same amount of trade unless you're looking for a certain pattern.

 

Are you sure your collect all bid/ask data? (i.e. Zenfire UDP feed CAN loose data)

 

No, as I said, it can certainly happen. You can reduce lost data though with a good connection and offloading all processing to a different thread than the one receiving the events to prevent blocking the thread from receiving updates.I've looked a long time on what's exactly happening in the order book of bund but I found no edge. It might be because it is so strongly correlated to the Bobl. A lot of what you see in the order book in the Bobl is just because something happened in the Bobl. I haven't analyzed the order book of the Bobl though, I might do it in the future, but I am actually working on looking at less, not more.

 

i)For example, it goes +34, +76, +22, +89, etc. which of which the sum then (surprisingly or not) ends up to be exactly 500, 1000 or 2000

This is a "black box algorithm searching" edge.

 

Call it want you want, but this algorithm clearly tries to manipulate the market without risking more than it has to.

 

ii)On how many months of "good" trading experience is your method based?)

 

ii)Can you detail your trade managment rules (Ex. TP+5/SL-2).

 

 

This is not a method or strategy I am presenting. This is just what happens. I do have very specific rules but it's very discretionary. I just developed them over time. I don't use indicators or set targets so there is no "trading system" to build around it. I try not to risk more than 2 ticks. A lot of times I just get out with a 1 tick loser or break-even if it doesn't do what I expected it would do or if I see something where I would reverse my position. Sometimes I reverse just because I am wrong and I know it is going to move a lot so I make back my loss and more. Sometimes I get out just because momentum slows down or I've reached my daily target. Sometimes I get out because I really have to take a piss

 

In general I like to enter at very low risk entries (i.e. where I can risk only 2 ticks or less and I know I am wrong), get the "edge" (be filled on a limit order or use a market order just before the price leaves) go with the trend/momentum, buy support, sell resistance and follow the largest traders.

 

iii)Can you add some words on the "big traders plays" you note in the DOM and in particular have you designed a "scanner" software for this plays?

 

You can't automate this. All the large short term traders are discretionary so they don't always use the same methods, but they do use some pre-programmed actions. They use what makes sense. Sometimes they sweep the offer only because they know everyone will puke if it moves another tick having entered their targets (offers) in advance and they are happy with just a few ticks on some quick (puking) action.

[cunparis 10]

This is not a method or strategy I am presenting. This is just what happens. I do have very specific rules but it's very discretionary. I just developed them over time. I don't use indicators or set targets so there is no "trading system" to build around it. I try not to risk more than 2 ticks. A lot of times I just get out with a 1 tick loser or break-even if it doesn't do what I expected it would do or if I see something where I would reverse my position. Sometimes I reverse just because I am wrong and I know it is going to move a lot so I

 

I am slowly coming to the same conclusion about the stop. My current goal over the past few days is to limit my losses to 3 ticks. I'm curious if you have an automatic stop in at -2 or if you use a mental stop? I'm currently using 7 tick stop on my DOM strategy but I try to cut it at 3. I'm debating if I want to move it to 3. Sometimes I am totally wrong and the market hits my stop very fast and I don't have time to get out. But other times it'll go -5 or -6 and come back to -3 and I bail at -3. So I'm trying to find the best way to do this.

 

Also are you scalping? What kind of profit do you get? I'm currently going for 5-10 ticks and once I get that consistent I want to try for more with my 3rd target.

 

Many people tell me 3 tick stop is too tight but once I started doing that I started doing much better so I think I'm on the right track. Learning to swallow my pride and accept a 3 tick loss was (and still is sometimes) very difficult for me so I feel like this is a major step for me.

 

Thanks for sharing.

[AgeKay 10]

It depends a lot on the market. If you're market is volatile, you're stops might need to be larger. What market do you trade? I don't use stop orders anymore, there were way to many times that I got stop out when a lonely 10 lot order traded. I just exit manually. It does happen that I lose more than 2 ticks but it doesn't happen very often and if I don't manage to get out so do others.

 

Bund is not that volatile and I managed to have only 1 tick losers yesterday (I was on fire). I am scalping. I usually go for 3-4 ticks, but I also take only 2 if that is all the market is going to give me. Sometimes I enter and the market just keeps moving my way so I stay with it for as long as it will go. Yesterday I caught the sudden move up at the news but I panicked when it moved back 2 ticks so exited way to early for a 6 tick profit. If I hadn't been so eager to take profit, I could have made 15 ticks on that move.

[cunparis 10]

It depends a lot on the market. If you're market is volatile, you're stops might need to be larger. What market do you trade? I don't use stop orders anymore, there were way to many times that I got stop out when a lonely 10 lot order traded. I just exit manually. It does happen that I lose more than 2 ticks but it doesn't happen very often and if I don't manage to get out so do others.

 

Bund is not that volatile and I managed to have only 1 tick losers yesterday (I was on fire). I am scalping. I usually go for 3-4 ticks, but I also take only 2 if that is all the market is going to give me. Sometimes I enter and the market just keeps moving my way so I stay with it for as long as it will go. Yesterday I caught the sudden move up at the news but I panicked when it moved back 2 ticks so exited way to early for a 6 tick profit. If I hadn't been so eager to take profit, I could have made 15 ticks on that move.

 

I trade dax for the european morning and then I switch to CL for the US morning. For both I really focus on a low-risk entry and I use a 6 tick hard stop but I try to be out with -3 ticks or better. It has taken a lot of practice but I'm getting a lot better.

 

My main concern now are commissions. I'm currently trading on simulator and when I factor in commissions they really eat into my profits. So I'm working on my exits and would like to stay in for some of the bigger moves. I trade 3 contracts and I'm out of the first at the first sign of hesitation. Stop goes to -2 ticks. Then the 2nd I'm aggressive too and when it is out stop is BE and I try to get 10 or more out of the third.

 

Have you traded dax? I'm curious how the bund compares. I'm used to high volatility markets such as Dax & CL so when I look at bund or fesx it's hard for me to adjust so I just stick with dax & cl. But maybe bund is better for scalping I don't know.

[cunparis 10]

To expand on my current challenge, here are two CL trades. My 3rd target is +10 which I will bring in if price is stalling. On the first trade I got out and missed the huge move. on the second I got out just in time.

 

I am coming to the realization that trade location is really important and I'm trying to learn when to take 10 and when to go for a bigger move.

[BlowFish 10]

I trade dax for the european morning and then I switch to CL for the US morning. For both I really focus on a low-risk entry and I use a 6 tick hard stop but I try to be out with -3 ticks or better. It has taken a lot of practice but I'm getting a lot better.

 

My main concern now are commissions. I'm currently trading on simulator and when I factor in commissions they really eat into my profits. So I'm working on my exits and would like to stay in for some of the bigger moves. I trade 3 contracts and I'm out of the first at the first sign of hesitation. Stop goes to -2 ticks. Then the 2nd I'm aggressive too and when it is out stop is BE and I try to get 10 or more out of the third.

 

Have you traded dax? I'm curious how the bund compares. I'm used to high volatility markets such as Dax & CL so when I look at bund or fesx it's hard for me to adjust so I just stick with dax & cl. But maybe bund is better for scalping I don't know.

 

3-6 points on the DAX is pretty impressive. Mind you it is a very versatile instrument, you can nickel and dime it (scalp) or capture some pretty nice 100+ tick intradday moves depending on your approach. There is an old thread somewhere about DAX trading only a few pages with peoples thoughts on it.

 

Actually DAX is not bad for commissions because of its fairly large €€€€ per tick (and volatile nature) in fact taking 1 tick on 3 contracts will leave you a good €25 Euros profit, not that shabby for a single tick! Be careful when you move off sim you do get slippage at times. Recently it seems to have been playing pretty nice.

[BlowFish 10]

Heres a good example of the DAX flexing its muscles. The spike into the high was 40 ticks in a couple of seconds, it came back almost as quick. Sometimes it does 40 one way back 40 the other and then resumes trading where it was! It usually seems to be around about 4o ticks for some reason. Actually i guess you saw this but this is not atypical behaviour (to use a double negative).

[cunparis 10]

Heres a good example of the DAX flexing its muscles. The spike into the high was 40 ticks in a couple of seconds, it came back almost as quick. Sometimes it does 40 one way back 40 the other and then resumes trading where it was! It usually seems to be around about 4o ticks for some reason. Actually i guess you saw this but this is not atypical behaviour (to use a double negative).

 

Was there news at that time? I stay out around news times and I only trade from 9am-noon german time. After noon things get slower and more volatile.

 

Today I stopped around 10:30 because it was very dead and then all of a sudden would get very volatile. I couldn't trade it. I'd enter and then the market would just sit there. So I spent my time observing the dom and reading a market profile book.

 

Check out this spoof I saw on the dom. The images were taken about 1 minute apart but the market was dead so there were not many trades executing, certainly not 50+. The market eventually broke out to the upside.

[AgeKay 10]

I trade dax for the european morning and then I switch to CL for the US morning. For both I really focus on a low-risk entry and I use a 6 tick hard stop but I try to be out with -3 ticks or better. It has taken a lot of practice but I'm getting a lot better.

 

I've never looked at CL, but I looked quite a bit at DAX a few years ago. I wouldn't trade it without looking at the FESX or ES, like I wouldn't trade Bund without looking at Bobl. I don't know if it's still the case but based on the ever existing 99% correlation of FESX and DAX, when I looked at DAX, it moved 4 ticks for every tick the FESX moved. Back then you could anticipate DAX moves based on how FESX traded or its depth changed. It was similar to the Bund/Bobl relationship. Bobl might go bid 500 from 1000 without any trade and Bund trades 1 tick lower. Same with FESX/DAX. It's great for getting in at the right prices, risking as little as possible. The reason I mentioned ES is that I had the feeling that ES would sometimes lead the FESX the same way FESX would lead the DAX. So you could basically look at ES to trade the DAX. It could be related to what time you trade, I don't remember. In any case, don't look at more than 2 DOM's at the same time, it will be too much information for you to process, especially since FESX/DAX are a lot more volatile than the markets I am used to.

 

My main concern now are commissions. I'm currently trading on simulator and when I factor in commissions they really eat into my profits.

 

Commissions are always a concern for short term traders. Unless you get really bad commission from your broker, they shouldn't eat too much into your profits since you're trading pretty volatile markets. But at the same time, you should always get the best commission deal you can.

 

But maybe bund is better for scalping I don't know.

 

Maybe, but I am scalping not because Bund doesn't much that much (which is the case) but because my decisions are based on things that require me to get out of a position after just a few ticks a lot of times. If I am long and think that it should go higher, but see that it will go down now, I get out, I might even reverse. I can always get back in long when it looks like it goes up again. No reason to give back profits and possibly let a winning trade turn into a losing trade. Of course, the risk of missing out has a much higher impact in markets that are volatile, so it's always up to your trading style.

[AgeKay 10]

This is what a good day should look like (this was today from 10:08 - 10:35 CET). Less than 30 mins of trading, no losing trade, 4 winning trades, 1 scratch. With the exception of that scratch, market didn't even go 1 tick against me. Winning streaks don't go on forever and it's Friday, so I just call it a day.

 

Entry and exit on my first trade really sucked though (the one that ended up being a 3 tick winner). I got in 2 ticks too late (cause I just sat down to trade and I don't like trading right away without looking at it for at least 5 mins) and got out 2 ticks too late (hadn't had time to analyze the volume since I had just sat down). My entries generally sucked today, I could have gotten in 1 tick better on each trade if I didn't insist on getting filled on the limit prices.

[AgeKay 10]

Cunparis, there was this guy on YouTube who would occasionally record his DAX trading and post it there. He was a phenomenal DAX trader. I remember one time where he had over 20 winning trades in a row and he risked less than he made on each trade so that is pretty insane for a volatile market like DAX. If I remember correctly he looked at the FESX chart and somehow always knew where good entries and exits were. He deleted all of his videos though so the only one that is left is this one: [ame=http://www.youtube.com/watch?v=-WEvm9u1MOI]YouTube - The DAX Scalping,Oct 19 USA[/ame]

[BlowFish 10]

Thanks for the reminder. If it is the guy I think it is then it's well worth a look.

[daedalus 10]

That a great video... truly one of the best i've seen in a LONG time. Applies almost same logic I do (entries are a bit different) but still, the LH/HL logic is the truth I chase personally.

[AgeKay 10]

I started working on IQFeed again. It doesn't explicitly give you the side that the trade traded on. Instead it sends a message (they call it "Update Message") that contains the current trade price, bid price and ask price. I can see how chart software developers that implement this feed made the mistake of comparing the trade price to the inside market of that message. It initially did the same and I noticed trades and the inside market that are contained in the same message are not synchronized which is the reason for "trades between the bid or ask". It would report a trade at 123.22 while it bid was 123.21 and the ask was 123.23. If you had looked at the DOM though you would have seen that 123.22 went bid and as soon as it went bid, it got taken out. So that trade "between the bid or ask" was on the bid. What must be happening internally with IQFeed is that they update the inside market before the trade so the bid is already 1 tick lower before the trade is reported.

 

Instead of comparing against the current inside market, you should keep track of the market side of each price using an associative array (dictionary, map, hash, index or whatever your programming language calls it). This will always correctly determine the side of a trade independent of the how the feed reports the data. Here is what an implementation in C# would look like:

 

 

//enumeration for market side

public enum eSide {

None,

Bid,

Ask,

}

 

//static dictionary taht keeps track of market side at price

private static readonly Dictionary<decimal, eSide> _marketSideByPrice = new Dictionary<decimal, eSide>>();

 

private eSide GetTradeSide(decimal bidPrice, decimal askPrice, decimal tradePrice) {

//set side at bid

if (_marketSideByPrice.ContainsKey(bidPrice))

_marketSideByPrice[bidPrice] = eSide.Bid;

else

_marketSideByPrice.Add(bidPrice, eSide.Bid);

 

//set side at ask

if (_marketSideByPrice.ContainsKey(askPrice))

_marketSideByPrice[askPrice] = eSide.Ask;

else

_marketSideByPrice.Add(askPrice, eSide.Ask);

 

//get side at trade price

eSide tradeSide = marketSideByPrice[tradePrice];

 

return tradeSide;

}

[cunparis 10]

It would report a trade at 123.22 while it bid was 123.21 and the ask was 123.23. If you had looked at the DOM though you would have seen that 123.22 went bid and as soon as it went bid, it got taken out. So that trade "between the bid or ask" was on the bid. What must be happening internally with IQFeed is that they update the inside market before the trade so the bid is already 1 tick lower before the trade is reported.

 

I understand your example and I understand the algorithm of your code, but I don't see how (or when) your code will know that 123.22 went bid.

 

are you programming to their API directly? Is their API C#?

[taotree 10]

I understand your example and I understand the algorithm of your code, but I don't see how (or when) your code will know that 123.22 went bid.

 

are you programming to their API directly? Is their API C#?

 

It's tracking each price and whether it was most recently a bid or an ask. So... when 123.22 went bid, it would be tracked as bid. When 123.21 went bid, it's tracked as bid. 123.22 is still in the dictionary as bid. So if a trade comes across at 123.22, it will be assigned as bid because that's the most recent value for that price.

[cunparis 10]

It's tracking each price and whether it was most recently a bid or an ask. So... when 123.22 went bid, it would be tracked as bid. When 123.21 went bid, it's tracked as bid. 123.22 is still in the dictionary as bid. So if a trade comes across at 123.22, it will be assigned as bid because that's the most recent value for that price.

 

From what I understand we only get the update message on each trade.

 

Let's say first trade is at 150.05 on the bid and we've never been higher. Then the bid goes 150.06 but there are no trades. and now you get an update message with bid 150.07 ask 150.08 and a trade at 150.06. we wouldn't have 150.06 in our dictionary?

 

If the bid/ask changes were sent independently of the trades then everything would make a lot more sense. Please forgive me if I'm missing something really simple.