Technical Analysis: Is it voodoo? Or does it work?

[Logic 10]

Yes, Blowfish. Each method of recording data will highlight something slightly different about the market and provide the trader with information that he can use to develop strategies to take money from the market. Unfortunately, the unsophisticated egos that frequent these threads need to exclaim that all that matters is what they choose to look at because they have looked at a zillion years or data and concluded years ago that all that matters is what they look at and what you look at can't possibly matter because you are a retail trader and his way is better than your way even though he does not have the guts to demonstrate all he claims before the fact.

 

It's a great lesson in unadulterated myopia that new traders should examine and attempt to weed from their personalities so that they can develop into successful traders.

 

Do you trade; intraday, swing or postion?

[MightyMouse 12]

Do you trade; intraday, swing or postion?

 

I trade certain markets, oil and gold, intra-day and others such as ES on a short term basis. Basically, 1 to 10 days.

[BlowFish 10]

Steve I was talking about sample rate (bar 'length' if you like) as opposed to sample size (number of bars). As you point out sample size (how much historical data you look at) is important too. This is what one does when they pick a bar size (plus type) and also choose how many bars to look at. This is likely to be influenced by the size (and or duration) of moves that a trader is trying to target. Even for discretionary traders how you slice and dice the data effects what is highlighted or obscured.

 

As an example few scalpers would work from a monthly chart, in fact the 'context' of the monthly chart is unlikely to be of much concern to someone scalping for a couple of ticks a trade. This probably seems obvious but from some of the questions you see (or even advice offered) perhaps it's not.

[steve46 12]

Yes I was thinking about your comment.

 

We call that problem "aggregation"...and it requires a comment relating to quality not quantity

 

My bad...sorry

[TimRacette 11]

Technical analysis is really just a way to identify the trend. The edge emerges when you combine your strategy for entry with cutting your losses when the trade goes against you therefore resulting in manipulating the odds in your favor.

 

HOWEVER, if you do want some interesting theory and empirical evidence to support that the markets are random Nassim Talebs' book's the Black Swan and Fooled by Randomness will do just that, I recommend the audio versions.

[Logic 10]

Technical analysis is really just a way to identify the trend. The edge emerges when you combine your strategy for entry with cutting your losses when the trade goes against you therefore resulting in manipulating the odds in your favor.

 

HOWEVER, if you do want some interesting theory and empirical evidence to support that the markets are random Nassim Talebs' book's the Black Swan and Fooled by Randomness will do just that, I recommend the audio versions.

 

Nice book to read, once, after you are consistently profitable.

Have you ever met and had a lengthy conversation with Nassim?

Something you ought to do, if you haven't . . . one time as well.

[TimRacette 11]

Nice book to read, once, after you are consistently profitable.

Have you ever met and had a lengthy conversation with Nassim?

Something you ought to do, if you haven't . . . one time as well.

 

Wow, that's cool you got to speak with Nassim. I've heard some interviews he's done on EconTalk, quite the eccentric character, but there is a lot of weight to his arguments. Pretty deep stuff.

[SIUYA 11]

As I read it the problem is if you a true believer in the black swans and unforeseen events (as an ex option market maker who loved being long volatility and making money out of disaster I do), then you might fall into the purists view of markets are unpredictable and therefore technical analysis is voodoo.

Yet, I still think there is value in following the road map, and at the same time allowing for the black swans....through such things as stops, money management, options, diversification.

 

Interestingly enough depending on which black swan event, some I have seen argue that via technical analysis these actually happen with the major trends already having been in the direction of the black swan event movement. (I have no opinion on this as everyones trend definition is different)

[Logic 10]

As I read it the problem is if you a true believer in the black swans and unforeseen events (as an ex option market maker who loved being long volatility and making money out of disaster I do), then you might fall into the purists view of markets are unpredictable and therefore technical analysis is voodoo.

Yet, I still think there is value in following the road map, and at the same time allowing for the black swans....through such things as stops, money management, options, diversification.

 

Interestingly enough depending on which black swan event, some I have seen argue that via technical analysis these actually happen with the major trends already having been in the direction of the black swan event movement. (I have no opinion on this as everyones trend definition is different)

 

I can attest that every Black Swam event I ever participated in was pre-empted by a confirming TA oscillation or oscillations including but not limited to; the London bombings, 911, the Japanese earthquake and the Flash Crash.

 

Taleb's book is a good read but far from being an accurate assessment of what TA is capable of showing an individual.

[joshdance 10]

I can attest that every Black Swam event I ever participated in was pre-empted by a confirming TA oscillation or oscillations including but not limited to; the London bombings, 911, the Japanese earthquake and the Flash Crash.

 

Taleb's book is a good read but far from being an accurate assessment of what TA is capable of showing an individual.

 

I'm sure you're not saying that a "TA oscillation" occurred before these catastrophies (9/11, london, earthquake) and in some way gave warning of a market reaction before the actual event occurred?

[Logic 10]

I'm sure you're not saying that a "TA oscillation" occurred before these catastrophies (9/11, london, earthquake) and in some way gave warning of a market reaction before the actual event occurred?

 

??

 

I'm not saying that a chart oscillation foretold 9/11, the London bombings, the Flash Crash or the Japanese earthquake. I'm saying that confirming oscillations put us short the equities about a month prior to 9/11 on our longer term charts and about 10 days prior on the shorter term charts. Josh, support levels in most, if not all of the Equities had made new lows the Friday (9/07) before. The news media was announcing after the attacks that they were going to investigate everyone that was short the markets on 9/11. A whole bunch of us contacted CNBC (Maria B.) and told her to bring her camera and her dumb fat butt and we would explain to her how the charts actually worked. Needless to say she never returned our emails.

 

Two of my Australian students posted beautiful short set ups on the NIKKEI two or three days prior to the earthquake that remained valid till many days after the quake.

 

The day before the London bombings, going into the close, we triggered sells in the Equities that reversed and confirmed bottoms prior to the market open. By the end of the same day we were making new highs. (The intraday equity charts were all in confirmed bull trends).

 

The day of the Flash Crash, strength on our equity charts had all turned down in the pre-market and into the early morning triggering shorts. We had sell signals on ever bounce up the whole way down. It was not a straight drop down as the ignorant media had you believe. There were buyers in that drop the whole way down.

 

Just last week in Crude on Tuesday afternoon we triggered a sell at about $113 that exited two days later, late in the afternoon at about $97. That is over a 10% pullback in 3 days and according to Taleb qualifies for a Black Swan as well. I posted that chart and many of our traders caught it perfectly (the ones swing trading Crude that is).

 

You can't do that with time based charting.

 

These are all objective rules based oscillation set-ups that we have programmed. These are not subjective discretionary art trades made by some idiot with a magic 8 ball, tea leaves and an aluminum foil pyramid hat.

[Logic 10]

I'm sure you're not saying that a "TA oscillation" occurred before these catastrophies (9/11, london, earthquake) and in some way gave warning of a market reaction before the actual event occurred?

 

You requested infomation from me. You had the opportunity to get my newsletter for free, get my indicator for free, ask me questions for free and test it for yourself for free. If free is too expensive for you I can't help that. Others in my thread on here did the same thing to prove to themselves it worked. I don't need to prove anything.

[TimRacette 11]

Logic does make a great point, most recently the flash crash, there were technical short setups before the crash even took place.

[joshdance 10]

You requested infomation from me. You had the opportunity to get my newsletter for free, get my indicator for free, ask me questions for free and test it for yourself for free. If free is too expensive for you I can't help that. Others in my thread on here did the same thing to prove to themselves it worked. I don't need to prove anything.

 

I said, "I'm sure you're NOT saying ..." ... I wanted to be clear on what you said, nothing more, nothing less. I'm not asking you to prove anything--you won't prove anything anyway, so I wouldn't ask you to. I was asking for clarification on a statement that was not clear to me. I didn't think that's what you meant, that's why I phrased the question that way. Are you looking for ways to take offense?

 

Free isn't too expensive for me, but several thousand dollars paid to someone who will happily sell me "the system" is. Or is it several hundred?

 

Meanwhile, I dusted off my programming skills and wrote a simple little program that will post my live trades to my twitter feed. I just wrote it yesterday, and it simply reads the NT log, and parses the lines where a trade is executed, and posts the relevant data to twitter within 2-3 seconds, not 2-3 hours or days as seems to be the trend with pretty much everybody, particularly vendors and "institutional" traders. It posts the trade execution only, not the limits or stops. The twitter page is http://twitter.com/joshtrader ... So far only those posting in the "reading charts in real time" thread do something similar, as did MM in his thread as well. Whether my trades are winners or losers is already a matter of public record on another site where I have a trading journal, so it's really nothing new, only these are posted as they happen. My thought, Logic, was, since I have nothing to prove just like you, then why not just put the trades out there? It will prove just what I said--that I have certainly been struggling the last few weeks with oil!

[Logic 10]

I said, "I'm sure you're NOT saying ..." ... I wanted to be clear on what you said, nothing more, nothing less. I'm not asking you to prove anything--you won't prove anything anyway, so I wouldn't ask you to. I was asking for clarification on a statement that was not clear to me. I didn't think that's what you meant, that's why I phrased the question that way. Are you looking for ways to take offense?

 

Free isn't too expensive for me, but several thousand dollars paid to someone who will happily sell me "the system" is. Or is it several hundred?

 

Meanwhile, I dusted off my programming skills and wrote a simple little program that will post my live trades to my twitter feed. I just wrote it yesterday, and it simply reads the NT log, and parses the lines where a trade is executed, and posts the relevant data to twitter within 2-3 seconds, not 2-3 hours or days as seems to be the trend with pretty much everybody, particularly vendors and "institutional" traders. It posts the trade execution only, not the limits or stops. The twitter page is josh (joshtrader) on Twitter ... So far only those posting in the "reading charts in real time" thread do something similar, as did MM in his thread as well. Whether my trades are winners or losers is already a matter of public record on another site where I have a trading journal, so it's really nothing new, only these are posted as they happen. My thought, Logic, was, since I have nothing to prove just like you, then why not just put the trades out there? It will prove just what I said--that I have certainly been struggling the last few weeks with oil!

 

I post Crude nightly. Let me know if you use NinjaTrader, Tradestation or MultiCharts and I will send you the indicator. I'll repost the trade set ups in tonights email so you have it too.

 

Like any method, if you expect to learn this in a couple weeks, don't bother. I say this because you already have established habits to overcome and that in itself is a wall you have to overcome.

 

Regarding Twitter and programming, I don't do either. My programmers are integrating the feature to post trades to Twitter in the new program they are building for me. There is no way I will ask them (at $150 per hour) to separate that feature out and I can't anyway because it isn't useful to me without the new program.

 

I really wasn't trying to be argumentative but you stated your question like it was a physical impossibility for TA to preempt moves on a chart and that just isn't correct.

[Logic 10]

Crude triggered at 8:18 am EST this morning, 2 hours and 12 minutes BEFORE the report, is currently up $1.50 and in an area an exit will trigger but nothing yet.

[joshdance 10]

Giving the status of a trade already in the money 150 ticks doesn't really say much. I'm not doubting you took the short, the market was definitely weak overnight and near the open, but posting that you're short at 8:18, instead of 10:35, would perhaps give a little more weight to what you said, particularly as this would be before the inventories.

Edited May 11, 2011 by joshdance
"particularly" not "particular"

[Logic 10]

Giving the status of a trade already in the money 150 ticks doesn't really say much. I'm not doubting you took the short, the market was definitely weak overnight and near the open, but posting that you're short at 8:18, instead of 10:35, would perhaps give a little more weight to what you said, particularly as this would be before the inventories.

 

As I stated previously.

All the trades I take are strict objective rules based. In other words, they can't be fudged on my charts.

I don't need to prove anything. If you would have had the charts, the indicator and rules you would have seem the same trade in real time. Again that isn't my fault. I've given you the ability to validate it for yourself. Other than that I can do no more.

[BlowFish 10]

Wow, that's cool you got to speak with Nassim. I've heard some interviews he's done on EconTalk, quite the eccentric character, but there is a lot of weight to his arguments. Pretty deep stuff.

 

I met him at some do that Wilmott (the quant training guys) organised, I got that impression that he does the odd speaking gig.

[Logic 10]

Interesting article today.

 

Oil crash pits floor veterans versus computer algorithms | Reuters

 

Pit oil traders are pissed that us new "technical traders" have an advantage over them. (I consider myself a pure technical [algorithmic] trader). I created my own algorithms just like they could have and I consider myself no smarter than any of them or have access to any more resources than they do. That breaks my heart. Just like any industry; keep up with technology or suffer for it ignoring it. I know many many pit traders that have migrated to the charts, the open minded ones that is. Of those, some succeed and most fail.

 

Still waiting on that email Josh.

[SIUYA 11]

Interesting article today.

 

Oil crash pits floor veterans versus computer algorithms | Reuters

 

Pit oil traders are pissed that us new "technical traders" have an advantage over them. (I consider myself a pure technical [algorithmic] trader).

 

While I generally agree with your sentiment, where do you get from the article that pit oil traders are pissed with "new" tehcnical traders?

[Logic 10]

While I generally agree with your sentiment, where do you get from the article that pit oil traders are pissed with "new" tehcnical traders?

 

Sorry for the disconnect. Two thoughts going on when I posted. I talked to a group of energy pit traders that were margined out on Wednesday and were blaming HFT's, algorithm traders and general speculators for their predicament.

[DavidTurner 10]

...we are not really trading the markets. We are trading other people

 

^^ That's the problem right there. It's an accurate description of technical trading, and also an accurate description of why it's a bad game to play in.

 

The returns to investors ultimately derive from the value created by the companies that receive their capital. On average stocks would not grow if the underlying companies were not, on average, growing. So if you're not trading the markets then you do not have those returns available to you. Instead you are competing in a win/lose zero-sum game - quite literally, you are gambling, and the rake taken by the brokers is pretty steep.

 

In fact I'd go so far as to say that any trade with a horizon less than a year is likely to be a form of hunt-the-sucker. I know there are a lot of A-types out there, but the truth is that statistically half of them will be the suckers. Not great odds.

[joshdance 10]

^^ That's the problem right there. It's an accurate description of technical trading, and also an accurate description of why it's a bad game to play in.

 

The returns to investors ultimately derive from the value created by the companies that receive their capital. On average stocks would not grow if the underlying companies were not, on average, growing. So if you're not trading the markets then you do not have those returns available to you. Instead you are competing in a win/lose zero-sum game - quite literally, you are gambling, and the rake taken by the brokers is pretty steep.

 

In fact I'd go so far as to say that any trade with a horizon less than a year is likely to be a form of hunt-the-sucker. I know there are a lot of A-types out there, but the truth is that statistically half of them will be the suckers. Not great odds.

 

So David, when did you give this a shot, and fail at it?

[DavidTurner 10]

So David, when did you give this a shot, and fail at it?

 

Ha ha, yes, very funny. Actually I've only ever been a top-down investor and occasionally a stock picker, and I've done pretty handsomely - especially after trading costs are accounted for.

 

There's an interesting study (sorry, too lazy to look it up but I think it may have been HBR) which says that if you'd simply picked the market index or bonds at the beginning of each month for the last forty years (and been right), you'd have managed a CAGR of something like 40%. Now that is something that I think I could at least have a crack at from an analysis perspective.

 

Anyway the basic truth remains - investing is about buying good companies at good prices.