Cutting your losses is liberating

[Linus van Pelt]

The chart snippet above what is its code?

It took me a while to find it again.
ELO, circa 2020-03-23.
I really was just looking for one with a long tail.

 

[Dona Ferentes]

Yeh ---I did this with a wife once ---very liberating.--

but a mathematical conundrum resulted; Divide things 50:50 and you won't end up with half !

Keeping with the thread. ,.. I don't employ stop losses.
"Cutting losses is liberating" .... and, anecdotally, most stocks when sold aren't worth revisiting.

For me, holding on too long is the biggest challenge (but that is another story).

 

[Linus van Pelt]

Keeping with the thread. ,.. I don't employ stop losses.

For me, holding on too long is the biggest challenge (but that is another story).

I promise I'm not trolling...honest question. Why not then? Would stop losses help with your biggest challenge? I intend to use them, at least in my AFL if not in the GTC market.

P.S.: I also have the same challenge as you!

 

[Dona Ferentes]

I promise I'm not trolling...honest question. Why not then? Would stop losses help with your biggest challenge? I intend to use them, at least in my AFL if not in the GTC market.

I just don't. Any purchase represents an investment, of time, money and mental focus. I find I carry the status, and am sufficiently engaged to make adjustments along the way. (hence a SL would be an artificial constraint). Some things are out the door quicksmart; others I have held for decades.

Works for me.

 

[tech/a]

Similar to DF but very different timeframe.
iuse stop levels as a position sizing tool
and ratchet it UP to B/ E then it’s gone!

 

[rnr]

I just don't. Any purchase represents an investment, of time, money and mental focus. I find I carry the status, and am sufficiently engaged to make adjustments along the way. (hence a SL would be an artificial constraint). Some things are out the door quicksmart; others I have held for decades.

Works for me.

Hi @Dona Ferentes,

My question is based on the premise that you don't mind sharing your thoughts on this process!

How do you establish your position size in a trade if you don't use an iSL?

Cheers, Rob

 

[Joe90]

Problem with that is that you can’t be sure some conditions will be met on close
Eg if a price breaks above a certain level
Or above an indicator level

you’ll get amazingly false results if you set to zero on most systems you can’t look ahead until you can see ahead

a system will record everything that works but none that don’t it won’t see them

Norgate Data Updater provides scheduled updates, you can trade MOC based on the scheduled 3.00PM prices, available at 3.20PM.

 

[qldfrog]

I was put onto this thread by a friend and have read it "cover to cover".

One thing I was planning to do in all my future buys is put an ISL order into the market when I open the trade. However, in this thread I read some mixed reviews on that strategy.

I suppose this approach can't be properly backtested. For example, say you have a weekly trading system and your ISL triggers on a Wednesday. That could never be backtested. Ok, perhaps it could with some "fancy" coding: a daily system that only trades on a particular day but triggers stops daily, with the GTFO stop triggered on the low. But even that doesn't account for the fact you probably only have EOD data feed.

But is there anything inherently wrong with setting a INITIAL stop loss, based on your buy price, into the market? You hope your stock goes up, and the trailing stop loss would prevent the ISL/GTFO loss to ever trigger. But if it tanks say 10% (or whatever your risk planning allows) straight away, then you made a bad purchase and you're out. If you find your in-the-market stops often get you out of eventual winners, is it possible your ISL is too tight?

I really want to implement proper risk management strategy so my "dog" stocks are shot as puppies (that image is for you @frugal.rock along with your festering cankers thread!).

In summary, what are your thoughts on entering an in-market, good-til-cancelled ISL/GTFO stop loss for every buy?

Edit: I did find this related post: https://www.aussiestockforums.com/threads/stop-losses-when-to-use.33459/.
Still, if you have anything to add re: my post above, please do so.

SL in a weekly system can be backtested:

Algo in words:
if low of the week is below SL,then sell at SL or if antong the worst case scenario sell at low of week...

 

[tech/a]

Norgate Data Updater provides scheduled updates, you can trade MOC based on the scheduled 3.00PM prices, available at 3.20PM.

I can trade live data
I can take an intelligent punt on the close and will get it right most times
but if you set a system to zero lag the results will be 100 %.correct as it
won’t show those that failed —- because it knows the closing price!

Takes a while to understand this but when you do it will be
like turning on a light.
in the meantime start trading a zero lag system you’ll soon be
scratching your head and toping up your equity.

 

[qldfrog]

SL in a weekly system can be backtested:

Algo in words:
if low of the week is below SL,then sell at SL or if antong the worst case scenario sell at low of week...

antong the worst ??
I mean, if you want to use the worst case scenario, just sell at low of the week, and you have a conservative backtest in AB of your weekly system using SL triggered during the week.
But as many have pointed, these SL destroy your average return, yet removing these before the feb crash costed me in the 5 figures.
I found going from weekly to daily is the best option if you have the time to run it

 

[Linus van Pelt]

I can trade live data
I can take an intelligent punt on the close and will get it right most times
but if you set a system to zero lag the results will be 100 %.correct as it
won’t show those that failed —- because it knows the closing price!

Takes a while to understand this but when you do it will be
like turning on a light.
in the meantime start trading a zero lag system you’ll soon be
scratching your head and toping up your equity.

I really like Bandy's stuff. It appeals to my numerate nature, even if I don't get it the first (or third) time. And he's highly respected, so his algorithms and approaches aren't just for my pointy head, but actually work.

But I sure wish he sold his books in digital format. It would make it so much easier for me to paste in here. Of course, not selling in digital form likely protects his intellectual property from piracy.

Putting on my data entry clerk hat, and forgive any typos...

Chapter 7, Entries

Market on Close of Action Bar
My research shows that the signals from a good system should be acted upon as soon as possible. If a system reliably forecasts the direction of price from the close of one day to the close of the next day, about one-third of that price change happens in the overnight market. Entering MOC of the action bar captures that profit - entering NDO misses it.

<Then elsewhere he describes a "MOC Order". But then later he says...>

Invariably, someone will point out that many of the signals depend on the closing price, and it is not possible to trade at the closing price if the close has already happened. Quantitative Trading Systems discusses this in more detail, but there are several ways to accomplish trading at the close of the action bar:

• Use a real-time data feed, note the price just before the close, perform the calculations, enter a market order or a limit-on-close order.
• Wait for the close, perform the calculations, trade in the after-hours market. Many EFTs and surrogates for them trade 15 minutes longer that regular hours.
• Pre-compute the price at which a signal will be generated. Depending on the data required for the calculation, this can be done just before the close or, or as early as the previous evening if you are using daily bars. Place conditional orders to be executed if the closing price satisfies the conditions. See the section in Anticipating Action below.

It is easy to get confused about what is known at what time, and what happens when.

So, forgive my misunderstanding that an actual "MOC Order" was "a thing". And for not understanding Dr. Bandy's definition of an "MOC Order" before my original post.

Finally, I assume my brief quote above does not cross the line toward reprinting copyrighted material. If anything, I will say that I've gotten a lot of great information from Mean Reversion Trading Systems and I recommend it.

If I have crossed that reprinting copyrighted material line then @joeblow feel free to edit or delete this post (not that you need my permission )

 

[tech/a]

Yes I agree
Brandy and I concur!

 

[Trav.]

The stop loss must be placed at a level that compliments your trading strategy. If you're a pattern trader then the SL is placed where the pattern fails. If you're a BO trader then place the SL at a level that indicates the BO has failed to go on with it.

Initial SL's must be placed outside the normal price noise for the time frame you're trading.

I think that this point by peter2 might have been missed by some, or at least this resonates with me.

The SL has to compliment your strategy so you should think about what you are trying to achieve when placing the SL, ie. if that BO fails then there is no point having a 30% trail stop in this strategy. You would more than like have it sitting at a predefined support level, or your calculated trade risk level.

Sounds obvious but as usual peter2 is on the money.

 

[tech/a]

Peters advice is sound
Personally I’ve found that if a trade doesn’t respond straight away then chances are it will either fail dismally or wallow around between buy and stop levels

I’ve found ratcheting up over a max 3 day period I get taken out a lot earlier than if I held my ISL

my reward to risk has increased substantially
But has taken a while to perfect

 

[Linus van Pelt]

or your calculated trade risk level.

I've still got my training wheels on but that's what I'm thinking right now. But it would depend on the style of trading I'm doing, plus supported by backtests.

At this point in my learning, I'd rather give back some modest gains and avoid big losses.

Based on the comments so far though, I don't think I'd put the stop loss order in the market, unlike my early sentiment, due to intraday lows. Instead, I will carefully monitor the trade in AB and execute the ISL on a NDO if the close hits the stop.

 

[aus_trader]

I lost an average of 30% on those trades. If I had left them all to run until last friday, I would have made an average of over +100% with only one loss out of the seven.

Problem with that analogy is that most bounced back hard this year.

But one has to be careful generalizing because in a bad year when the markets may continue to decline month after month, you'd be glad to have cut the losses and limit the damage to 30%.

 

[Warr87]

Problem with that analogy is that most bounced back hard this year.

But one has to be careful generalizing because in a bad year when the markets may continue to decline month after month, you'd be glad to have cut the losses and limit the damage to 30%.

the time that it doesn't is the time you are particularly hurt. also, that no longer becmes risk management but gambling .....

 

[qldfrog]

And the most liberating: getting rid of a stock which has stopped trading months ago and is under bankrupsy.
I have one now AJM: report value:7c real value:0 and can not get rid of it.
before july, will offload for tax purpose using one of these 'donation site' at a small fee

 

[jbocker]

From my younger naive stupid days on stocks buy and sell I have held a doozie that have merged and amalgamated, transformed business That current price is 0.02c and for me to recover the original sum would need to sell for $12.44 a share. It has now again merged. from Quantify shares (which were originally 1 000 000 WHL shares)
To recover costs these now have to sell for 154.56 per share. (beat THAT!)
Liberation day is coming as there is a largish capital gain happening this year, now in my older (still stoopid) days.
I cant wait to bask in that liberating feeling....

 

[Jack Aubrey]

Thanks Jack.

By last Friday I assume you mean 27Nov, not "the" last Friday after your stop loss hit. Depending on when you bought, that could be a LOT of time to recover.

This is all great in hindsight. But at the time of the stop, you didn't know if the stock would have continued downward, so the stop could have saved you from big losses. After all, you had those orders in the market for a reason. I assume it was to prevent a loss greater than you had tolerance for in your risk management.

If you don't mind doing a tiny bit of work, could you look at your charts for those 7 positions that were sold, and let me know if they sold due to an intraday low or due to a close below the stop? IOW, if you had manually executed that stop at EOD after market close?

Note the difference between the two example stops in my screenshot

View attachment 115569

These were just the trades I was "kicked out of" on intraday drops and, yes, I was comparing my outcome with the price at close on 27 November. I can see that my auto-stops were set far too tightly for spec stocks and that my trading "strategy" was hopeless. I was getting caught out on quite normal pullbacks and spec volatility. I also had too many stupidly small positions in a misguided effort to spread my risk. Even now, I would not have let all of those losing positions run (although I do regret bailing out of FYI at that stage). These days I have fewer, larger positions and don't set technical stops. Instead, I look at why I entered the trade and assess whether a drop means anything has materially changed in terms of the fundamentals or the market. I am also having a lot more good luck.