Dom's notes

Despite having been hammered for posting my sincere feedback re. ATW in the appropriate thread (or may-be, because of ...) I am starting this thread for the benefit of all waanabee traders.

I have no intent of sharing my exact trade setups, however I will share as much as I can - the good, the bad & the ugly, and if someone learn something from this then it won't have been a waste of my time.

I have been a student of the market for over 4 years now. I actually started my education with Investools PhD program (which was a lot of money for not much outcome in my case), trading stocks then options, then started to focus on the SPX (directional Calls/Puts then credit spreads), at some point I discovered the futures / ES in particular and that was a "revelation" ... 1/4pt spread vs 2pt spread for the SPX options, I was sold in no-time.

I purchased Todd Mitchell's system (TradingConcepts), which was a lot less money than Investools but no more outcome in my case, at least it triggered my interest in Fibs (back then, retracements only) and I did a lot of research / backtesting using Fibs - but at the end of the day, the real challenge resides in figuring out which fib level will "hold", and to this day I have to admit I have not solved it.

Anyway, I then spent a lot of time in a couple of "free" trading rooms
learning pure price action, the person that was offering his time teaching PA free for several months later offered a paying program, which I took, but I still couldn't make money live. I took another mentor, recommended on a free forum, and despite he being a nice guy (and why wouldn't he be when students are paying), this also was a failure.

I was about to throw the towel when I found ATW, started with the 101A education, added 3 weeks later the mentorship education, and have been studying / backtesting a lot for now a full year. I reached a point where I can "consistently" make fake-money on sim, however the transition to live trading is a real challenge for me, because of fear I pass on many trade setups and I deviate from my trade management rules.

So, that's the background. I trade CL (Crude Oil "big" contract) which is very liquid, has 1 tick spread throughout the European & US session, and is nicely volatile (day range is on average 250-350 ticks, some days even more - like today : 450 ticks). I find CL to have a lot of momentum, it shows very repetitive "patterns" (at least, for the fib user that I am).

I don't use any indicator, I trade price action using a lot of price projection techniques (best book IMO on this topic is Robert C. Miner "Dynamic Trading").

If I have one advice for new traders, it is look for another way of making a living ... but if you are truly in love with the markets, then I believe it is a must to 1) find an excellent education (this is way more than just reading books, and frankly, there is probably no one-stop education shop) and 2) find a mentor to accelerate your learning curve.

Now that this is out of the way, a brief summary of my week :

Tuesday - tried 1 trade early morning, entry hit no fill, that made me mentally sick for the rest of the day, I passed on 2 setups (both wins), the last setup I tried but my entry wasn't even close to be hit.

Wednesday - 2 setups no-fill, then 1 small winner (got out at 1/2 of my target for pseudo-good reasons - really, lack of discipline), and I passed on the last one (another wouldabe winner :( )

Thursday - passed on 5 setups (4 wins / 1 loss), took 1 small winner (1/2 target again, same lack of discipline), and missed the best setup of the day by being away for 5min

Friday - passed on 1 setup (win), no fill on next 2 setups, then I couldn't focus & called it a day.


Bottom-line - only 2 trades this week, I made ~10% of what I should have made if I had the discipline to follow my plan.

Only 1 trade this week for my other system ... 1 win net +575. I did lack discipline on this trade & bailed out as price spiked toward the target & missed it by 5-ticks.

On the R&D front (reversal system), I took some time to re-visit my 1st target parameters ... It is extremely frustrating that some trades do miss that 1st target by a few ticks, and end-up net losers (I do not ever move stop to breakeven, I only move it under last low upon a higher-high in an uptrend). But the bottom line is that the current target-1 parameter is truly the best one for P&L, but not for frustration or drawdown. I identified another setting that is certainly worse for P&L (that target-1 P&L is reduced by 10%) but far better for win%, frustration level & drawdown ... the win% at target-1 goes from 59% to 65% (on 500 trades, this is significant) and the drawdown levels are significantly reduced ... I have yet to make a final decision on this, but I am thinking I should use that reduced target-1 when I restart trading this system, as it is easier to be disciplined with a higher win% and lower drawdowns.

The other item I worked on this week is setup reactivation idea. The very 1st attempt at it was quite crude, but generated a large enough number of trades to identify a couple of required changes in setup definition & trade management, and then define 2 sub-patterns to explore separately. I worked on the 1st one today, and got rewarded at my 1st attempt: it generated 26 trades with a win% of 69% (at the "max P&L" setting for that target-1), and an average per trade about 1.5 times the basic system average per trade. I then explored the performance at various settings of the 4 parameters of that sub-pattern, interestingly I had them all correct on the 1st try (I must start to know something about trading CL ?!), but I refined the testing of these parameters (essentially < vs <=, and as always fighting with Ensign floating-point errors in comparisons). The final version has only 23 trades in the official backtesting period (plus 2 in CLH12, which I count as forward testing), this is certainly too low to make a definitive judgement on the performance level, but I feel safe to state it is probably as good as the basic system.

Next week I'll address the 2nd sub-pattern. I wouldn't be surprised for it to be a fail, that's what backtesting does best: showing what doesn't work.

Indeed, it only took 1/2 day to verify that the 2nd sub-pattern is a plain fail.

I am pretty close to the finish line now on this system, I started detailed backtesting this afternoon (tick-by-tick market replay, to catch any "real-time" bug, which are relatively frequent using DYOs, even for the "expert" that I have become), that's going to keep one of my PCs busy for the next 4 days (!), meanwhile I am going to revisit the initial-stop setting (I believe I can do better than what I currently have), and then I will start checking the detailed backtesting results (each trade is logged in a file, which I can access from another PC, so I can verify the trades for the dates already replayed).

Originally posted by dom993


Indeed, it only took 1/2 day to verify that the 2nd sub-pattern is a plain fail.

I am pretty close to the finish line now on this system, I started detailed backtesting this afternoon (tick-by-tick market replay, to catch any "real-time" bug, which are relatively frequent using DYOs, even for the "expert" that I have become), that's going to keep one of my PCs busy for the next 4 days (!), meanwhile I am going to revisit the initial-stop setting (I believe I can do better than what I currently have), and then I will start checking the detailed backtesting results (each trade is logged in a file, which I can access from another PC, so I can verify the trades for the dates already replayed).

dom993
Do you use standard deviation in your calculations

It depends ... for example, when I was evaluating the setup reactivation idea, I did not, as the data sample was so small (but I keep an eye on the win% of target-1, at least to ensure it is "compatible" with the basic system).

Where I use std-dev is really for deciding of the max drawdown I am allowing the system ... for that I use the trade distribution out of backtesting, run 10,000+ MonteCarlo sims for 1 year (typically) worth of trades, then use the mean & std-dev max-drawdown to compute the maximum allowable drawdown for the system. Because trading results are NOT (never) normally distributed, using std-dev to that purpose (or any purpose in trading) requires to anticipate those "fat-tails" which are much more frequent than black-swans. I currently use mean+5*std-dev (1 year) for the max-drawdown allowable - but to be honest, I have no rationale for using the 5* multiplier (vs 3.3* or 6* or 10*), aside from wanting to make sure - as much as financially possible - not to stop trading the system in the midst of a "normal" drawdown.

I use similar techniques to monitor the early performance of the system in live testing, this time using the exact number of trades executed by the system.

Not the best week for my other system, but it comes out in the green, and on a marginal new equity peak (+100 vs January 9th): 2 wins / 1 loss ; net +515

I have to say the loss came first (-875), and as at that time I was looking for a win & a new equity peak, it did hurt quite a bit my mood. Luckily the next 2 trades were full win, and I finally get that new equity peak (even if only marginal) which ends an almost 2 months drawdown period.

On the R&D front (reversal system), I did spend a couple day revisiting the initial stop calculation, and even though it is currently sub-optimal (because it is capped at 30-ticks, even when the size of the reversal pattern would call for a larger one), I really don't like the exposure of an uncapped initial stop (largest pattern in the backtesting calls for a 83-cents initial stop) ... I might twist that cap a little bit, but anyway it is a compromise and will always end-up frustrating one day or the other.

Last couple days have been dedicated to the review, trade-by-trade, of the detailed backtesting (tick-by-tick replay). So far I checked 19 months (out of 28 total), found & fixed 2 bugs, and thanks to an oddity in Ensign I am fully aware of how sensitive this system is - it uses 200-vol bars, with a number of things requiring a min number of bars (pivots, reversal leg "duration"), and there is always the possibility for something marginal happening (either see or miss a pattern for a few contracts) ... I just have to admit that it will happen, and hope this isn't too often.
On that note, I am thinking of testing a "voting" mechanism, where I would run the system on 3 timeframes (185 / 200 / 215 vol. bars) and only take trades when 2 systems agree. But I have no other way to test that than detailed backtesting (tick-by-tick replay), which takes a lot of time ... I must finish the current one first.

I wish you all a great week-end!

Just lost my post - wasn't logged :(

Too frustrated to do it again. 4 BE this week for my other system (net -140 on the week), and a lot of issues uncovered re. the reversal system.

Finally, a new equity peak for my other system! Although nothing spectacular, but it had been 2 months that I was waiting for it! 2 wins / 1 loss / net +685 on the week.

I finally received & started reading the "Evidence-Based Technical Analysis" book by Aronson. I am part-way through it, but I can already vote it as the best trading book I read in many years! Chapter 2 (mostly about psychological flaws) is nothing really new but well written & documented, chapter 3 makes a great job at describing the basics of the scientific method (even though I am an engineer & have had a 25 year career in high-tech R&D, I have to admit I was totally unaware of it), chapters 4 & 5 provide a sound statistical basis to test the predictive power of a trading rule/signal/system. Funny enough, the tool I was using to decide on the max drawdown to allow to a system (MonteCarlo simulation) is the right tool (but used in the wrong way), and it was a matter of seconds after reading the book to use it in a much more effective way.

On the R&D front, I have been working on some of the issues identified last week, most of them pertaining to pivots identification - making this more reliable, especially when using slightly different timeframes (I was using V200, now also working with V180 / V220 & V100). I also identified & fixed a good number of bugs (1 more to go), and generated a couple new CRs regarding fine-tuning the entry placement & trade management. Overall I believe I am making progress, but I am still nowhere close to a decision point re. taking this system live again.

Another positive week for my other system: 2 wins / 1 loss / 1 BE ; net +510

More of the same on the R&D front ... I almost exhausted my todo list for this system, I should be done with it within a couple weeks (I hope - lol).

I finished reading "Evidence-Based Technical Analysis". The last part of the book is actually a pretty sad flop, as it describes the systematic testing of over 6,000 signals on the S&P500 daily, of which the "best" one doesn't pass the statistical significance test for data-mining. But I am not terribly surprised, as I doubt he would have published a signal of real value. All in all, it just confirms that identifying a systematic (programmable) edge statistically significant is of the utmost difficulty, and also highlights how "easily" one could get led to wrongly believe he has found something of value.

A milestone week for my other system, despite taking a large loss : 3 wins / 1 loss (-815) ; net +550 on the week. The milestone is breaking the $10,000 mark on cumulative P&L since I started trading this system in July 2011.

Stats: 116 trades, 68 wins (58.6%) / 46 losses (39.7%) / 2 BE (1.7%), P/F : 1.46, avg/trade: +88, avg-win: +474, avg-loss: -479. The P/F & avg/trade are lower than the prior 2.5 years in backtesting (P/F 1.85 & avg/trade +135), the main difference being the win% (~65% backtesting vs 59% live). (all figures for 1 contract, which is my current trade size for this system)
2 significant drawdowns, not by their depth but rather by their duration: -2750 / 3 months (Jul-21 to Nov-7, 42 trades) & -2390 / 2 months (Jan-10 to Mar-2, 22 trades).

Discipline stats: 13 discipline errors, which cost me a total of -545 (9 errors costing me money for 4 errors that gained me money). Monthly: March (0) February (1) January (0) December (1) November (3)October (1) September (0) August (2) July (5).


On the R&D front (reversal system), I am feeling I made some serious progress this week. I started the week being sick of porting changes from one version (200-volume chart) to the other (100-volume chart), and decided to bite the bullet & create a flexible version, self-adapting to whatever-volume timeframe it is running on. Most of this work pertained to the pivots identification mechanism, which now takes its parameters as Volume for the left & right side, and converts these into the appropriate number of bars. Of course, this didn't make for any performance improvement, only saving future efforts. The first CR I tackled after that led me to a series of improvements, which can be summarized by saying the system P&L on 29 months went up +5700 (92k to 97.7k - for 2 contracts) while the number of trades was going down by 29 (545 from 574). It would be pointless to describe these improvements, but in a nutshell I removed the cutting of a couple corners (simplification assumptions made in the early stages of the prototyping of this system), allowing the system to better identify & interpret price-action.

That said, the last 6 months' performance remains very low (+4850 for 119 trades), and I am going to take yet another look at these.

Not the best week for my other system, 1 lone trade & it's an almost max loss ; net -835 for the week.

Some progress on the reversal system, went back to one filter idea that I couldn't get to work a few weeks ago, this time my brain worked better :) Also resurrected a much older filter, which I had discarded because, despite a sensible performance improvement on target-1, it kills too much of the runner performance (big runners are not always coming from the picture-perfect/textbook setups). It still has the same effect, but I have a renewed interest in boosting the target-1 performance as much as I can (see below).

I am now convinced that CL changed behavior somewhere around mid August, as I can clearly see that no matter how much improvement I put in this system, the last 7 months performance is just a shadow of the prior 22 months. For example, using that resurrected filter, the avg/trade on the prior 22 months is $292 (2 contracts) for 257 trades, while the avg/trade on the last 7 months is $74 (2 contracts) for 95 trades. Without that filter, it is $237 avg/trade for 392 trades, vs $53 avg/trade for 142 trades. I think it is about time for me to revisit those last 7 months with a different goal, understanding what new behavior CL exhibits now for reversals.

That said, it the runner's performance which has been decimated in the last 7 months ... the target-1 performance is much reduced, but still showing $57 avg/trade (win% 62%) with the resurrected filter. So at this point I am thinking about taking that version live with 1 contract - at least if I can convince myself that the current level of performance has a chance of not dropping further moving forward. I still have some work to do before that, but that could be the goal for the start of trading of CLN12 (around mid-May).

Happy Easter!

Dom that equity curve looks too good to be true so I'd be scared but hope it trades like that in real time....what became of your trading contest...? how did Chuck Huges do ? He wants me to spend almost 4 k on options recomendations