Well good luck on the trading with real money. I’m going to put that podcast into my rotation of finance related topics, so thanks!
Alright, I did it. I took my trading live Jan 11. I don’t have any detailed results to post because I have only take 13 trades since I started. This is because I know my trade plan has still not been vetted enough for me to fully trust it so I only occasionally take trades just to face the fear and start getting practice in. For the most part I watch the market during my session, take notes, and journal about trades I could have taken. I am being caution since I have set a maximum draw down of $300. If I hit it, back to sim. I would rather not have to do that so I am holding myself back. This way, I can continue to learn while still live. So far my account is down by about $30 (including commissions)
To be honest, I am struggling and I’m miserable. Every day I face that hard right edge and only twice has my decisions rewarded me with a winning result. 75% of that is me making mistakes as I practice implementing a rather complex discretionary plan. 25% of my lack of success is my plan still being in development. It is really painful! Sometimes I just watch the market for my whole 3 hour session and won’t see anything set up. I will feel like a failure for not trading. Sometimes I get so wound up in a particular bias I will take a sloppy set up just to prove to the monkey in my head that at least I have the guts. Then the trade fails. Sometimes I do everything right, get scared, take loss against my rules, and then it will got strait to my win target.
So, why should I still do this? Well, I’m actually not supposed to be good at this yet. In fact, at this stage in the game if I can just keep about even, I am on track. My trading hero, Morad Askar, tells of when he first got into trading. He quit his project management job and went to work at a prop group called the SOES Bandits. He says for 3 months strait, trading full days, he just lost money. It was not until the 4th month that he could even figure out how to “stop the bleeding”. That was in a professional environment with lots of crazy good traders all around and helping. Since I am trading half sessions, comparing myself to him, I should be on schedule to be a break even trader after 6 months… or February. That gives me hope 
In addition I am part of a little online community which includes a few retail guys who actually do manage to make a living. I have learned much from them and it creates a somewhat professional environment to interact in.
Even though my trading has been frustrating lately. I have to lift myself up to recognize how far I have come. Just how I view the market has come a long way. My original trade plan was very simplistic and the only reason it was successful in my early sim trading was because I was not being completely honest with myself (my earliest trades were actually just little lines on the chart, did not even have a demo simulator hooked up yet) It was really easy to say “oh, I was just thinking about taking that one”… after it start losing. Anyway, now my plan is very specific and includes collecting information from order flow, volume profiling, and chart formations. It is a 3 step process that starts with context and then seeks specific setups that honor risk/ reward. More importantly, instead of just looking at chart patterns, I look for places where other traders are wrong. In general, my trading is more sophisitcated. Also, now that I am live, there is nowhere to hide. My P/L does not lie!
If anyone is curious about my trade plan or the tools I use to collect information I would be happy to share. These days I focus only on Bund futures (FGBL) and I still trade a session from 2:30am to 5:30am.
Meanwhile, I just wanted to post an update and mention how much it sucks right now… because later, when I’m good at this, you can see what it took to get there. 
it can be consistently profitable but you have to do it “professionally”… it wont work if you are not committed 100% and full time… “retail” day trading is an oxymoron because retail implies some ridiculous commission/ fee structure that will rape you… Interactive Brokers is as good as it gets for “retail” but professionals have access to rebates and even lower fees…the real game is options trading and market making… firms like Jane Street and Optiver have real prop trading desks with pros pulling in real money from real people but its all options… the money in “stock” trading is all HFT/ quant stuff.
Dont worry ive gotten slaughtered the last few days as well. It happens, keep at it, keep adjusting assumptions & parameters and eventually you may catch on to something.
Fee structure is not so critical if you are not a scalper. My eventual target is to find about 3 round trip opportunities per day. One of the nice surprises about going live is I found out my round trip commissions+fees for FBGL are only $2.60. Nice! I have no other fees at my broker but I do pay for my data (extra for eurex) and my charting/ analysis software(investor RT)… about $150/month.
I am 100% committed to trading, but I have to be patient in my development. I will get there but since I only have 25 hours a week to give, it will take longer. I have to keep my regular job while I learn. The reason for the crazy hours is I wanted to make sure it was a time I could set aside and have a consistent routine. Anyway, the key is deliberate practice and consistent performance review. I definitely have those down… it will just be spread out over a longer time.
But yea, industry money is going to be way more attractive than what an independent trader can expect. Good thing I’m not expecting a whole lot.
On prop trading equities outside of HFT: I know a guy at Trillium that’s been making money consistently for several years. Successful equity traders do exist, although I hear there are full days where there are basically no opportunities for humans to earn, so they just hang out.
I don’t think you should put a number on finding opportunities.
thanks!
yes, plus my friends in my online community have been great. I am always asking them “what made you enter the market there?” Or “how could you tell size stepped in, the DOM just looks like a freaking mess to me!” They are really helpful. I have made a lot of progress with order flow and understanding what to do with “size” because of them.
well, you are right. Sometimes there will be none… or more like 5. It is more a statement about how specific my setups are. I look for the confluence of a few different things to happen in one place and I would like to get at least 8-12 ticks, which is much more specific than a scalper who is less specific but only targeting 3 ticks… and could trade hundreds of time in a day.
I appreciate you mention your strategy is “discretionary” but can your set ups be formalised into a set of logical rules? If so, being able to code and back-test your strategy opens up a whole new world. Half the problem with discretionary strategies is that you’re trading and developing your rules as you go along with no real knowledge of whether they even truly give you an edge or not - So you doubt yourself every step of the way. It’s extremely mentally frustrating!
At least with backtested results you can gain an insight into the long term results of your core strategy and build from there. It gives you that little bit of mental sanity to fall back on.
If there is anyone on this forum that gets what “discretionary” vs. “systematic” really means, it is you. I actually started out last March as someone who was looking to develop an algo and trade completely non-discretionary. I was working with my friend who is a programmer and a trader. What I kept discovering over and over was that trading systems are “dumb”. Because they can’t have too many degrees of freedom, you can only make them so complex. You can’t get them to read context. They hunt setups, but those setups may or may not make sense given the bigger picture. You can add things like filters, but I found they did not really improve the results. Because of this, they essentially get their edge by assuming the market is random and finding a standardized way to enter and exit. Because of that you trade profitability for stability. You can get stable results, but I think as a skilled discretionary trader that includes context as part of the process, you can get better return on capital as well as stable results.(by taking less “dumb” trades) On the other hand, this is why I can probably appreciate the results you presented in the other thread more than anyone on here. Developing an algo that is stable AND profitable is really hard. I presume you do “walk forward” testing and optimization?
However, this is all theory, It is not like I am setting the world on fire trying to “read” the market. I have been able to achieve stability (larger draw ups than draw downs). Now, I just need to be right a little bit more often Lets see if I can accomplish this…
Yes, I do have a set of logical rules, you just cannot back test it. Here is why…
My process
-
context… pre session I go over the big picture and come up with some theories as to where price may be attracted to and where it may be rejected or interrupted. I also think about what kind of path it may take to get there.
-
evidence… once the session begins I start taking note of things like trending behavior, momentum, and order flow characteristics. I use the “evidence” to help me decided which of my “theories” from my pre session is likely playing out
-
set ups…Once I have a directional bias, I look for at least 2 triggers happening in one place to take a trade which will put me in play in line with my pre session theory.
so yea, try to program that! That’s the trade off though. I get to always be in tune with the market ( in theory) and don’t have to fear the blow up because my “edge” vanished. My trading should change with the market. But, I don’t get the quantitative reassurance that you get with algo. All I can do is get the screen time in and keep notes on stuff that is working and stuff and consistently screws me over. I hope it pays off…
What if all of your pre session theories are wrong?
They won’t be. I have a plan set up for if price drives to the upside, just chops all day, or drives to the downside. The useful information comes in as I describe the levels and zones that I expect to influence price in each of those directions.
So during my session if I see a set up to go short, but it is at a level where I was expecting a bounce, I won’t take that trade. Instead I start to look for evidence of bottoming out and a set up to go long… for example.
And obviously, as you can tell by my frustration I am wrong about these thing a lot! Even at best I still expect to be wrong frequently. Thats just the reality of this game. However, the next few months should have me making the correct call more often and that will get me profitable. This will be a make or break time in my development. Deciding to just study and trade one product was a step in the right direction.
I saw this meme on FB and had to laugh because it sums up my semi pathetic position I find myself in as a trader. This month I had only scratches and losses. No wins. I’m actually getting worse! I really should move on… become a plumber or something. Looking at my performance trajectory in a logical manner, I would say my failure is statistically significant.
mehhh… I’m not going to do that. I don’t feel done. Even though I traded so poorly, I managed to not hit my $300 max draw down yet. (kind of an accomplishment…I guess?) I feel like I have solved many of the pieces to the equation of a trading process but I have not been able to the put it all together yet. Here is how the month went down…
I started out feeling really excited because it seemed I finally had a comprehensive plan for discretionary trading. This was my trade plan.
1_) context…. pre session I go over the big picture and come up with some theories as to where price may be attracted to and where it may be rejected or interrupted. I also think about what kind of path it may take to get there._
2) evidence…… once the session begins I start taking note of things like trending behavior, momentum, and order flow characteristics. I use the “evidence” to help me decided which of my “theories” from my pre session is likely playing out
3) set ups……Once I have a directional bias, I look for at least 2 triggers happening in one place to take a trade which will put me in play in line with my pre session theory.
The problem that arose was that this turned out to be far less specific that I thought it was. Step #3 was really clear. I have specific set ups that are discrete. It was step #1 and #2 that was getting me in trouble because I was never really sure how to apply information. You will always see reasons (“evidence”) to take a trade… or not. It is always somewhat of a mixed message. So, what I was doing was I would see a set up happening but I would never be completely sure if it was supported by context and evidence. This is really bad. No matter what, intraday trading needs to be about consistency. The only way you realize your edge is by repeating the same process over and over. I found I could not apply context with any sort of consistency. What would happen is I would second guess everything, and when I finally decided to pull the trigger it happened to be a loss. I would get extra nervous, sit out of winning trades… finally take one… repeat. The main point, I was never really sure if context supported my set up or not. I could not be consistent.
I became so frustrated that I said… forget it! I will go back to algo. Forget context! So, I started manually back testing some of my set ups ignoring context. I ended up focusing in on one of them and I actually got fairly robust results based on 200 samples. Roughly it came out to 1/3 wins, 1/3 loss 1/3 scratch. That does not sound to exciting but it was using a 1 to 3 risk reward structure. So yea, that’s pretty great! The problem is that the set up sometimes only occurs a few times in a month (given my session is only 3 hours). I suppose I could test some other setups and play them too. However, then there is the problem of the algos being correlated and therefore likely to all have draw down at once. That is too unstable for my taste.
Now I have come full circle. I need to be a discretionary trader who applies context. I need to fix the problem I originally had with ambiguity in my trading decisions. I am currently trying to develop a context heuristic. I want to identify exactly what information I will use as context and then have a standardized way that I apply it to identify zones where setups are valid. Then I can have a have a way to apply context with consistency. With that in place all I have to do is follow my rules because I will know exactly what they are.
Well, off to my charts to start working on my heuristic. I’ll be back next month… probably to whine about more failure… or maybe not 
The struggle is real!
it is impossible to make money for yourself through trading because - if you are making money consistently you will consitsently increase your positions and will become the richest person in the world, and we know that no one has become ultra rich just by trading for himself
sadly i realised that too late - every time you loose, you try to identify a mistake and vow not to repeat it but all you are doing is to identify a pattern where there is none
This is inaccurate.
Yeah spelling of second consistently is wrong.
While I am far past getting defensive over jaded remarks, I will say that you have nailed my current issue with accuracy. However this is exactly the problem I am trying to solve. My trading decisions need to based on a repeatable set of conditions and not questioned after the fact. The reason that my trading is currently failing is that I have, apparently, not yet been able to accomplish that. I cannot afford to question each losing trade and ask, “what about this was wrong”. Losses are to be expected, and ideally, not questioned. Key is for losses to be accepted as part of the plan and stay consistent. It is not about being right each time. It is a win/ loss ratio vs. risk/ reward structure that works out over the set of results.
and BTW… there are definitely patterns and repeatable behaviors in the market. Otherwise, no algo would ever work. There is just no such thing as something that happens “every time”. There are only things that work, like I said before, under the function of win rate vs. risk/reward structure. Sorry you were not at it long enough to discover any.