The BS sim journal...Add your own BS!

[qlai]

Here you hold through a 17 point drawdown and end with just a couple of points profit.

You did a nice analysis! I was looking at MAE but it doesn't seem to reflect that for some reason.

with real money trading, these can be detrimental.

Yeah it's hard to deal with this emotionally, to me anyway. But I think lots of automated systems do this ... Increase the stops to bump up win rate and hope that the losses don't cluster.

 

[qlai]

I was looking at MAE but it doesn't seem to reflect that for some reason.

Oh, it's an average.

 

[Gotcha]

But I think lots of automated systems do this ... Increase the stops to bump up win rate and hope that the losses don't cluster.

Hope is the big problem. A human trader will be affected by this. Maybe increasing stops can make mathematical sense in certain situations, like a bad entry, high volatility so you need to use a bigger stop than usual, etc., but there has a be a reason to do it, and proof that it will be profitable in the long run. Based on what I see here, the large stops are just a way to increase the win rate.

Plus, as I'm reading between the lines of what Overnight says, he is scared to go live, and reluctant to, even though he seems to be doing well in sim. I think the voice inside of his head is saying that what he is doing in sim he won't be able to do live, and I think he is right. When you see these trades, you see how close he comes to it being a losing day. When trading live, I think the outcomes of some of these trades would be different and the PnL would not be nearly as high.

 

[Overnight]

...

I took the liberty of plotting your trades so there is something to discuss. Here are the first two trades. Red dot is a short, black dot is an exit, and in further charts, green dot is a long. Your first short you held on for a very long time until you exited. Seems like you were just hoping and praying. On your second short, this too went against you almost 10 points.
...

Yes, thanks for the plotting.

Because I use Unirenkos, time is of zero import to me. So what may seem to be a long time on a time-based chart translates a lot differently visually.

Here the trade period you've covered in your plot with what I see on this end.

1st half..

The problem with screenshots is they cannot show the screen action at the time of working the trade (obviously.) It all depends on how the market is looking at the moment, combined with the current FA of the day, other chart-size action action, etc. When I made the decision to close that first trade for a loss, the bar looked like it was about to close higher. After a few minutes when it then looked like I was going to be wrong and we might be starting a 20-point cycling motion, I said welp, get in on the cycle, so I shorted again. What happened next with that 10 point move against me? It happened in the space of a minute and stalled. 3 minutes later it was back at BE and went into a choppy area. A hard stop there for 10ish points would have knocked me out, but seeing the speed of the move and having seen it before, I "discretioned" it.

Jump ahead now to the next trades. Trade 3, the short, why such a small profit here? Unless there was a clear reason, you're going from one strategy to the next it seems where you hang on for a long time, or just take little nibbles.

Yeah, that's exactly what I did, and do. Sometimes I just like taking little bites out of the market if it seems that is all it will give me. My daytrading strategy is not fixed for the entire day, it is what I think the market will give me at the time I enter a trade, and monitored during it's lifespan.

Trade 5, the long, is a huge problem. Here you hold through a 17 point drawdown and end with just a couple of points profit. Sure it looks good on paper since its a win, but some of these never come back, and with real money trading, these can be detrimental.

Again, that is another case of "you had to be there to see it." The drop happened in the space of about 2 minutes, stalled, and then resumed upwards 30 seconds later. The target was just another nibble. Just because my initial target is 2 points instead of 20 means I can't let it go against me for a bit? Yes, some of them don't ever come back that day. But where my target is is not going to affect that. If my target was 200 points and market pulls back 10 from entry, goes back up 5 points from entry, and then drops 100 more points and I close at a loss of 100 points, then what difference did it make that I took a 100 point loss on a 2:1 ratio trade? I risked 100 to make 200, but I still lost 100. But what if you take the profit at 5 instead of 200? Did you really have a r:r of 40:1?

Trades 6 and 7 are now just little nibbles, and if you ask me, indicative of fear, so you take what little profit is offered. The rest is kind of all over the place as well.

Now don't get me wrong, I don't know your strategy, so maybe its all good from your end and as intended, but huge losses, huge drawdowns and small wins isn't profitable long term.

Nope, they are not indicative of fear, just looking for little nibbles

Also, I see problems with your data or how your platform handles sim trades. Because you provided time stamps down to the second, it was easy to plot your trades. The trouble is some trades happen outside of where the market was at the time. If you're using market orders, this will of course make a huge difference. And if you're using limit orders, perhaps in the real world, you wouldn't have gotten filled. Notice how in the chart below, these two long trades are off by several seconds, as are the exits. I put the dot exactly where your time stamp indicated, but the market just wasn't there. For other trades, the time stamp lined up nicely with where the market was, but perhaps this was just because price was in a tiny range so even though it was off, there was still time for it to fill at that price.

I know my data is good, and I would expect your data is good too, but why your time stamps are so far off I don't know.

I know for a fact that Ninja's sim engine leaves much to be desired as far as fills. The datafeed is spot on as it's CQG, but the engine simply can't handle a decent fill when the tick data suddenly floods in as it did on those errant trades outside of where the market actually was at that moment, which you can see on the chart. The little sideways triangles point to the bar, and at what price (drawing a horizontal line to the right from it's tip) the trade occurred. Blue for buy and pink for short.

My entries here are all market orders, as are the exits that say "close". Those are market-order exits. The other exits are all limits. This is why I always say that I get better fills in real than sim, because of this fact. In real trading, the difference is negligible in the outcome for me, for over the average, bad fills go both ways...To the plus and to the negative, and they cancel out. Therefore sim can be taken as an average as well.

Thank you much Gotcha for taking the time to do your plotting and start a decent discussion here. That is exactly why I post my journal logs, sim or real, with that information...So someone like you who has an interest can simply do what you did to see for themselves what is what, rather than settle for the subjective re-telling of stories about a person's trades. (*Yawn*, yeah, I got out of the trade at around 10:20AM, the instrument was at X. (Meanwhile, within that 60 seconds, the person posting that doesn't realize that a LOT can happen in those 60 seconds these days. ))

 

[Overnight]

Also, in what on your plots would be trade #8 and 9, is exactly the sort of thing that burns me about stops, mental or hard. I entered into the start of a chop zone, and decided to get out when it looked like it was about to break on the downside for it's next run. But lo'! It went back up! So back in for 10 points. Lost 6 points, made 10. Net 4.

Or, if I had just left it alone, LIKE I USED TO BE ABLE TO DO IN THE CALM WATERS OF 2017, I could have just made the 10 net which is what my original target was.

Alas, they are so necessary now, it sux. Unless you went long on Dec 26th of course.

 

[Gotcha]

Because I use Unirenkos, time is of zero import to me.

I actually didn't mean to imply the time factor, although I'm sure its my fault for the misunderstanding. What I meant was that you took a profit, a small profit, even though often you held through a big loss. So its not so much that its a quick profit, but rather, a small profit. Stops need to be in proportion to targets in my opinion.

Now don't get me wrong, what you describe sounds like a legitimate trading plan and since its discretionary, who am I to say that its wrong. But what worried me was how you came to reap the rewards you got. I think that over the long wrong, luck wouldn't be on your side. Those areas where you hope to take a scalp trade, might not give you the scalp, and because you're liable to hold through a deep drawdown, it might never come back.

But I also don't want to highlight that a positive PnL is still the most important variable. The question is if you can do this consistently lets say 4 out of 5 days, and, just as importantly, if this type of trading will translate in live trading. I know from personal experience that when you have all these different variables, (ie. taking small nibbles, staying in during a deep drawdown), bad things happen when trading live because you get the combinations all wrong. When you change just one variable, everything falls apart. But you really only know this when you go live.

Thank you much Gotcha for taking the time to do your plotting and start a decent discussion here.

No problem. I think these days, the only discussion that has real value are actual trades, even if they are just sim trades. Guys calling out where the markets will turn with nothing on the line is such a waste of time that there is no point in getting into the discussion.

At the same time, someone else's trades are also kind of pointless as well because what they see and do has absolutely no bearing on what I'm doing. Took a long time to learn this. But at least talking entries and exits makes it more real.

 

[Gotcha]

Or, if I had just left it alone, LIKE I USED TO BE ABLE TO DO IN THE CALM WATERS OF 2017, I could have just made the 10 net which is what my original target was.

I did careful analysis on very many trades and what I saw was that my trade management was the weakest link in the entire chain. Even shit entries sometimes would hit a profit target if just left alone. No matter how much I think price will do something, it easily does the opposite half of the time. For every trade I exited at break even, thinking I'm preventing a loss, there would be another that would hit full target if I just let it.

What all this taught me was that once the trade is entered, its best to just leave it alone. Unless I statistically prove to myself that I am able to see which trade will work or not as its unfolding, then no point in trying to pick and choose which trade to hold or which to exit. What is most important to me is putting the trade where I want it, and making sure that my stop is just outside of the noise (usually the most recent swing high or swing low). Then the trade either works or it doesn't. If I have a good combination of risk and reward, then profits should take care of themselves.

But I do agree with you that the volatility of the past few months has really been screwy. It really means having to adapt. The stops have to be much bigger, but then so can the target be, if we just leave the trade alone to hit the targets.

 

[Overnight]

..What is most important to me is putting the trade where I want it, and making sure that my stop is just outside of the noise (usually the most recent swing high or swing low). Then the trade either works or it doesn't. If I have a good combination of risk and reward, then profits should take care of themselves.

But I do agree with you that the volatility of the past few months has really been screwy. It really means having to adapt. The stops have to be much bigger, but then so can the target be, if we just leave the trade alone to hit the targets.

Yep. Exactly. And when we go from snoozing on a daily range of 40 points in summer of 2018 to wild 200 point swings on the daily, there is no time to adapt with the hard stops, in my head. My hard targets and stops don't work anymore. So I have to go all mental.

The noise in summer 2018 was in a 10 point range. The noise now is in a 20-30 point range. The noise from Oct to Dec 2018? Good God man, it was 50 points at a clip. WTF.

Today was quite a respite in the NQ as compared to the last few weeks, we ranged about 40 points today. Feels a bit like old times. But I do not expect one day a new volatility range to make, not with you-know-who licking his own golden shoe.

 

[volpri]

Scaling in when price moves a traders way is generally a bad tactic when price is in a range UNLESS it is a very large range that allows for 3 or more legs in a move. Most ranges should just be scalped.

However, averaging down can be very profitable in a range and it doesn’t have to be a really wide range. A trader sells at top and if he gets an adverse move the trader sells more if the adverse move at least meets the traders minimum scalp size. Each time the adverse moves continues the trader can add to the position. Usually once, twice, or thrice is usually enough. Then as price reverses one can exit part of the position at BE and hold the rest for profit. Or if appears to be a good reversal hold all the position for more profit. This type of buying and selling has to be done with limit orders and not stop entry orders. Because one wants to sell at the top when taking short postions and add on adverse moves to the short and take long positions at the bottom and average down on adverse moves. Novice traders do the exact opposite and sell on a BO’s out of the bottom of the range and scale in by selling even more on further movement south. Then the reversal comes and they find themselves stopped out. Novice traders buy on bull BO’s out of the top of the range and scale in with even more as price moves in their direction. Then suddenly price reverses and they are stopped out with a loss.

The reason this strategy works is because around 75% of BO’s from a range fail within 5 bars and price goes back down into the range if the BO was at the top, or up into the range if the BO was on the bottom end of the range.

A word of caution: this strategy can generally be used several times AS LONG as a range continues to show range price behaviour..i.e. overlapping bars..many reversals...large up bars followed by large down bar. Large down bar followed by large up bar...tails on tops and bottoms of bars..bad follow through on moves up or down in the range or bad FT after big range bars. However a trader needs to be aware that all ranges at one point or the other will break out into a successful trend; bear or bull. So, a trader needs to be able to read the pressures as the range is forming and continues to try and determine the most likely direction of the coming successful BO. A trader can learn to achieve this by watching the things mentioned above and by counting the bull/bear bars in a range. And by taking note of how many series of bear or bull bars there are in the range and their size.

Try it (on a sim of course!) before knocking it!

Volpri

 

[volpri]

PS there really is no noise in the markets. What appears to be noise on one TF is simply trends..PB’s ranges..etc on another time frame.