We're all wrong - Apparently

Blimey, didn't think such a mild swear word would be edited out. For the record, it wasn't the f word, it was an alternative description for posterior and rhymes with farce, which links nicely back to the article.
 
"Chet Currier is a Bloomberg News columnist. His opinions are his own."

Perhaps he should keep them to himself if that article's the best he can offer!!
 
What Planet?

A very large one a long way from the Earth perhaps.


:cheesy:
 
Maybe the article was badly written but to be fair buy and hold has worked well enough for Warren Buffett and he's only the second richest guy in the world as a result.
 
I believe it was Warren Buffet who said the only time to sell is.....
NEVER :cheesy:
 
Warren Buffet

I know this is not a "Warren Buffet" discussion but it has been well acknowledged that he had a nack of picking undervalued companies so it is more his nack of picking the right companies that lead to him being successful, not just a blanket policy of buy and hold any old share forever. I think that it is possible to make a lot more money by carefully picking shares either through TA or FA or by using both (which is how I do it), letting them run and then selling when the run stops. Then do the same with the next share.
 
my 2 cents

Obviously there are people out there that can time the market and do it reasonably well enough to make a living. I didn't read the market but if he thinks that nobody can time the market then he is dead wrong. However, if he argues that the average Joe cannot time the market and is better off sticking to index tracking funds with a 15-20 year investment horizon and low management expenses then he is right.

Have a look at some of these articles for interests sake
http://faculty.haas.berkeley.edu/odean/Current Research.htm

One article actually looks at the trading habits of tens of thousands of US households and basically attributes losing behaviour to overtrading and therefore excess brokerage.

So we just need to minimise brokerage, increase our win size and win percentage. Simple stuff really (said with tongue firmly in cheek). Actually I am sure that many of us are doing really well at this and hence the indignation at his suggestion.

Personally I have chosen futures trading to save up the capital for longer term investing sometime down the track. Simply less work but much more capital needed.

dant
 
LTBH as per Buffet is of course the only way to trade correctly. The fact that Mr Buffet made his cash a long time back and has since increased it further most of the time clearly shows that this is the only way to do it surely? Otherwise you'd have the preposterous situation where there was more than one possible solution to a problem, and solutions from the past would (even more ludicrously) not necessarily work for evermore... and if that happened where would it end? Streets full of redundant chimney sweeps, boot blacks, sedan chair operatives, heaven knows where it'd all end.
Dave
 
"..day-traders all went bust.. " Shadower quoting Bloomberg.

Sure, right enough, its very very easy to jack away money on day trading. Is the oft repeated line true that 95% of traders lose?

The more relevant question is, are many or any members of T2W losing?

The reference to timing in successful day trading in this thread would presumably be when set-ups and/or suitable entry conditions can be utilized. Actually what is acutely important is the actual time .. the time axis measured in minutes when a particular market (ie Dow) is in session.

There is usually little or no reference to the role of actual time in all the huge output on charts and in all the great and continuous literature from market gurus. The best that occurs might mention wider periods of time .. whether they are better or not.
 
......Timers cannot get by on one or two good decisions; they must make hard choices correctly over and over again.

The odds of calling a coin flip right once are 50-50, and twice consecutively 25 percent. By seven times in a row, your probability of success is down to less than 1 percent.



This is contrary to just about everything I have ever experienced or read about trading. Trading is not about getting it right time after time, in fact it isn't even about getting it right 50% of the time, it is about having a strategy which has a net positive expectation over many trades (in other words making a net profit). This may require only 1/3 of trades to be 'correct' but if the gains from the 'correct trades' are on average 3 times the losses from the 'incorrect trades' you will profit overall.

In fact it is wrong to think of individual trades as correct or incorrect, this misses the point entirely, it is only the overall strategy which has to be correct, losing trades are an inevitable part of any succesful system in the stockmarket.

Short term trading (which is obviously what he he targeting) is also very little about timing, it is 90% money management.

This guy is obviously talking up his portfolio
 
I have been working for the past two years on the maths of trading. I asked a large audience a couple of years ago for their help in analysing their figures but very few people stepped forward which led me to believe that the following may be true:-

People who are succesful don't want to share it?
Maybe there are just very few success stories out there.

Of course there are probably many reasons but I was suprised at the lack of response.

I thought it may take me 12-18 months to resolve some fundamental issues but every time I dig deeper I uncover another problem to be solved!!!

One of the problems with the expectancy theory requiring low entry point success is that the expectancy appears to be directly propotionate to the stop loss. There fore you can not move one without affecting the other. Also the degree of certainty shows that you could have an adverse run which could cause you pretty severe problems.

If for example you have a succesful trade 25% of the time, To ensure with a 95% degree of certainty that this occurs you will need to perform a range of at least 11 trades (Roughly). In 100 trades at this rate you are likley to have a losing streak of 16 trades and a positive run of only 3!! That is enough to test the mettle of anybody.

The only solution I have found to this is to trade longer term where the rewards are potentially far higher.

But this also shows how you need a large range of historic trades to understand if you system is going to work at all.

I have so far been unable to scale all of these down to an intra day level but am attempting to. Would hope to meet a few people on these boards that will explain their intra day strategy so I can see if my "Magic" forumlue will fit into this scale.
 
Hi Scrip,

You may well be right about the successful not being prepared to come forward because I am also not prepared to reveal exactly how I manage a trade which is the key to successful trading in my view. What I will say is, and as a rough guide, that 67% of my trades result in a no loss situation. This is very much down to how the trade is managed from point of entry and has only a small amount to do with stop losses and that is all I am willing to say.

Cheers


Paul
 
Scrip and Paul

I am experiencing a similar dilemma at the moment.

I am currently tading a system I have developed to trade several FX pairs. It is not intraday, more position trading with trades lasting typically 1 to 5 days.

I have not traded it long but I have had 10 losing trades in a row which is starting to worry me a little. Also, I am unable to back test with any degree of accuracy as entries are discretionary and I don't have the equipment to do it anyway.

However, I plan to review my trading rules every 3 months to see if subtle changes could have improved my performance- both profits and drawdowns.

I am not too concerned as the system is overall profitable and only 4 of my "losing" trades resulted in actual losses. ie the others were breakeven or close to breakeven.

I don't know if my system will continue to be profitabe in future, nor do I think backtesting with a computer can tell you that either. However, if I review my rules regularly, this I beleive will give me the best chance of achieving trading success long term.

What interested me about this post was scrips comment about 25% success rate. This just happens to be the % of trades that have hit my profit target as of this morning! spooky!

Also, Paul if you don't mind me asking of the 67% of trades that have no loss, what proportion actually result in your targetted profit levels? Again, the reason I ask this is that overall, this figure is remarkably similar to my own! Spooky again!

Cheers

Darren
 
Scripophilist,

I think it's more to do with not wanting to sell the silver. I run a stat arb desk, very heavily math based, rolling longs and shorts over variable periods of a couple fo days to a couple of weeks. We've been very successful in it, last year being actually our worst, 150ish% ROI. Because we make our cut by playing the inefficiancies in an 'efficient' market, if others started doing what we did we'd cut our own legs off. The arb opps would get smaller and smaller and it wouldn't be worth it.

I think also that if someone has sunk a good couple of years into the r&d of a big system, they might wanna talk in general terms but never at the code level.

As for drawdown and bad runs - seen 'em, you can't avoid them. You can reduce their impact through money management, portfolio hedging, etc but I've seen one of our models lose a third of it's value in a month. Now we've got a pretty cooking hedging algorythm but .......

Intraday can be done but, as it looks like you're finding, when you zero in on the timeline, all the bugs get much much bigger.

Best o luck

Dave
 
DarrenF

mmmm, maybe reviewing every three months, or changing the rules base every three months, not sure mate. In what we do ( YMMV ) we've got several model all running at once. One is obviously our live trading system but if the backtests and parrallel paper trading models look promising we maybe switch across. You might wanna try and have a couple of systems watching your chosen market all at the same time - then you'll be able to see when a particular model works best in an up trend or down trend or through a range.

Then again, I've got a crap load of hardware here that I can backtest against and run parrallel model's, so I guessing I've got it a tad easier than your set up. I'd definately look at doing some backtesting before putting a model live - you may have worked out a rule set for that current market moment but which wont live beyond the market horizon.

Best of luck

Dave
 
It would be quite easy for me to record trades on an alternative method or two and run a comparitive profit/loss comparison.

To be honest, I'm only talking about tweeking stop levels, profit targets and position size so I think I might give this a go.

Thanks for the idea.

My logic behind reviewing every three months was to observe and record optimal stop and profit targets from a sample period and then adjust accordingly for the next three months, thus taking account of the most recent market conditions. Does that make sense?
 
Scripophilist

you may have to be careful of the logic of looking at longer trade periods if shorter periods do not seem to work - might be best to assume the underlying trading logic is wrong - not the time scale
 
Top