Day trading / Swing trading

[Duritz]

Anyone have experience in them, advice, thoughts?

Am going to begin researching them, in attempt to come up with some methods of my own to predict share price changes from end of day (eod) to end of day, but I'm also interested in knowing if there's reliable methods out there already, or software. I know there's a lot of software on the market but no doubt a lot of it is bs.

Thoughts on it as a whole? Potential for profitability? Pitfalls? Positive leads, etc?

I don't think day trading so much really suits me, is holding for a few mins and selling, but I like the idea of buying them for a day, or a few days, and trying to move on with at least a few percent profit.

 

[androidtommy]

I have no experience or particular knowledge but I often look at the 5 worst performed stocks for the day and most of them tend to rally back in the next day or two. IE with SGH when they had their really big fall pre-Xmas I thought they would go back up. It went something like down 30%, down 20% then up 30%.
I don't have the guts to buy into those 5 worst performed stocks in case they go down again, but I think half of them would rally back to similar position, some move only a little up or down and some have another big fall

 

[Duritz]

I have no experience or particular knowledge but I often look at the 5 worst performed stocks for the day and most of them tend to rally back in the next day or two. IE with SGH when they had their really big fall pre-Xmas I thought they would go back up. It went something like down 30%, down 20% then up 30%.
I don't have the guts to buy into those 5 worst performed stocks in case they go down again, but I think half of them would rally back to similar position, some move only a little up or down and some have another big fall

Yeah, I wonder if there's a detectable pattern there. If a stock drops more than X% in a given day, it has a Y% likelihood of regaining Z% of the lost ground within a week, or whatever.

 

[George Washington]

Anyone have experience in them, advice, thoughts?

Am going to begin researching them, in attempt to come up with some methods of my own to predict share price changes from end of day (eod) to end of day, but I'm also interested in knowing if there's reliable methods out there already, or software. I know there's a lot of software on the market but no doubt a lot of it is bs.

No, there is no easy magic secret to making money. Anyone who tells you there is lying and is probably after your cash.

The beauty of the market is even if there was some kind of reliable 'system', it would very quickly become priced in by the market, making it no longer profitable.

The road to making money in the sharemarket requires a lot of reading and it takes a long time (the answer no one likes to hear..).

 

[Duritz]

No, there is no easy magic secret to making money. Anyone who tells you there is lying and is probably after your cash.

The beauty of the market is even if there was some kind of reliable 'system', it would very quickly become priced in by the market, making it no longer profitable.

The road to making money in the sharemarket requires a lot of reading and it takes a long time (the answer no one likes to hear..).

Thanks for the honest reply. BTW I am happy to do a lot of reading, in fact in the last couple of weeks I've read an enormous amount about the subject. I also come from a really strong analysis, probability assessment and gambling background in horse racing so have a bit of a head start on a lot of stuff, especially as regards technical analysis and money management. There's a lot of crossover between the two forms of speculation.

What I don't seem to be able to find is a middle ground between day traders and swing traders. I'm interested in a single day approach, ie buy at start of day, sell at close. This is a real racing type approach, longer than the minutes of the day trader but shorter than the swing trader. I can't seem to find info about that as a method.

 

[ash_1050]

You'd need to be doing it with a decent stack of cash otherwise trade fees (approx $15 each buy or sell) are going to eat away at your profit. Do trading on paper for 2-3 months before hand and track things meticulously to see if it's worth doing for real, chances are it won't.

 

[Duritz]

You'd need to be doing it with a decent stack of cash otherwise trade fees (approx $15 each buy or sell) are going to eat away at your profit. Do trading on paper for 2-3 months before hand and track things meticulously to see if it's worth doing for real, chances are it won't.

buffet called it in the short term a popularity contest. Surely that indicates inefficiency

 

[ash_1050]

buffet called it in the short term a popularity contest. Surely that indicates inefficiency

There is inefficiency and it can definitely be profitable but unless you're willing to invest the large amount of hours and cash to do so it's likely only going to end in tears.

 

[Duritz]

You'd need to be doing it with a decent stack of cash otherwise trade fees (approx $15 each buy or sell) are going to eat away at your profit. Do trading on paper for 2-3 months before hand and track things meticulously to see if it's worth doing for real, chances are it won't.

I thought more about this post this morning and did some maths. You're right in that a $15 fee is a lot if you're outlaying small amounts. EG at 1k per buy its 1.5%, that's a lot to overcome if I intended on selling after just one day.

A $15 fee on a 20k purchase is 0.075% of cost, so less than one thousandth of a percent. In that scenario the fee is negligible. Therefore a model purchasing 20 companies shares at 1k for a one day hold is vastly inferior to a model that does one purchase at 20k, assuming they all acted equally, which of course they don't!

It occurs to me that a way around this is leverage. A 1k purchased leveraged 5 times cuts the fee to 0.3% of outlay, but that does introduce greater risk obviously being leveraged.

Answer - I'll run more numbers yet...

Question though while we are on it: what are good and low fee websites to use to trade this way, on the ASX?

 

[ash_1050]

I thought more about this post this morning and did some maths. You're right in that a $15 fee is a lot if you're outlaying small amounts. EG at 1k per buy its 1.5%, that's a lot to overcome if I intended on selling after just one day.

Question though while we are on it: what are good and low fee websites to use to trade this way, on the ASX?

There's a $15 selling fee as well so you would need to make at least 3% just to break even on a 1k trade. Leveraging can be handy, it can also cause drastic losses though if you're not wary. As I mentioned earlier please paper trade this for some time before jumping in the deep end, otherwise it will be an expensive way to learn.

Not sure about resources for day trading however I use NABtrade and it does a decent job. Commsec is also pretty decent from what I hear.

 

[Duritz]

There's a $15 selling fee as well so you would need to make at least 3% just to break even on a 1k trade. Leveraging can be handy, it can also cause drastic losses though if you're not wary. As I mentioned earlier please paper trade this for some time before jumping in the deep end, otherwise it will be an expensive way to learn.

Not sure about resources for day trading however I use NABtrade and it does a decent job. Commsec is also pretty decent from what I hear.

Are they around 0.1% each side for fees?

 

[ash_1050]

Are they around 0.1% each side for fees?

Not sure for commsec. NABtrade is a flat $15 for every buy/sell trade and more for options trading.

 

[Duritz]

Not sure for commsec. NABtrade is a flat $15 for every buy/sell trade and more for options trading.

That's the one I want then, the flat rate, cheers.

 

[George Washington]

That's the one I want then, the flat rate, cheers.

It's not a flat rate, most brokers will be around 0.1% above 20k, think CMC is the cheapest for small parcels at $10 up to $10k

 

[ash_1050]

It's not a flat rate, most brokers will be around 0.1% above 20k, think CMC is the cheapest for small parcels at $10 up to $10k

Ah apologies, I've never traded above 20k

 

[Duritz]

Ok no prob, thanks. My most likely trade is to spend about 5k on a buy, leverage, I dunno, maybe 100 times, then buy a mansion.

 

[George Washington]

Ah apologies, I've never traded above 20k

One day mate.

Duritz, you're going down a path of losing money, but I see this a lot and inevitably the person has to see it through and lose some cash before they realise it!

 

[Duritz]

One day mate.

Duritz, you're going down a path of losing money, but I see this a lot and inevitably the person has to see it through and lose some cash before they realise it!

I'm pretty sure I'll win right away though and never look back.

 

[Old M8]

One day mate.

Duritz, you're going down a path of losing money, but I see this a lot and inevitably the person has to see it through and lose some cash before they realise it!

Yeah agree with this.

Sorry Duritz, probability assessments aren't going to help you here. It's not like horse racing.

You've already made a mistake thinking there is a market pattern.

There is no such thing as a market "pattern." There is market history, but the market doesn't follow patterns. There is no script followed.

Unless you're outlining tens of thousands, it's not going to be very lucrative either.

 

[Duritz]

Yeah agree with this.

Sorry Duritz, probability assessments aren't going to help you here. It's not like horse racing.

You've already made a mistake thinking there is a market pattern.

There is no such thing as a market "pattern." There is market history, but the market doesn't follow patterns. There is no script followed.

Unless you're outlining tens of thousands, it's not going to be very lucrative either.

I'd be enormously surprised if there wasn't a predictable pattern. I will research the guts out of it anyway, to find out. If you're right and there's no pattern then I won't spend a cent on it.

 

[George Washington]

I'd be enormously surprised if there wasn't a predictable pattern. I will research the guts out of it anyway, to find out. If you're right and there's no pattern then I won't spend a cent on it.

Prepare to be enormously surprised.

The key leap in logic is: if there is a predictable pattern, the nature of a market is once that is known, the pattern is no longer predictable.

For example, if everyone knows the price of ABC stock will go to $11 tomorrow. It will go to $11 today.

 

[Duritz]

Prepare to be enormously surprised.

The key leap in logic is: if there is a predictable pattern, the nature of a market is once that is known, the pattern is no longer predictable.

For example, if everyone knows the price of ABC stock will go to $11 tomorrow. It will go to $11 today.

If that's the case, no one could ever make money in the stock market. Therefore, how do you explain warren buffet?

 

[androidtommy]

If that's the case, no one could ever make money in the stock market. Therefore, how do you explain warren buffet?

Warren Buffet would have bought at $5 when he had a reasonable expectation that the business was profitable, had a solid financial scaffold and was well run, giving it every opportunity to hit $11.

I believe everyone can do that too if they have the patience to research properly and wait for an opportune time to buy. I'm not that patient yet though

 

[Duritz]

If all the talk is right about an impending recession then there's going to be plenty of value buys in a while

 

[George Washington]

If that's the case, no one could ever make money in the stock market. Therefore, how do you explain warren buffet?

You absolutely can make money in the stock market. First you have to accept prices move in a random walk in the short term and are unable to predict though.

Buffett isn't exploiting 'patterns' in the market. He is exploiting the power of compounding returns and at times the short term irrationality of the market.