natureboy1618

....the whole point is - a butterfly is testable. No its not TRADING. Its a price behavior. This thread is about randomness in price. Trading is a whole other activity than analysis - totally. You can make money trading with zero statistically verified analysis. Its uncommon, guys have done it. At any rate, the whole point of studying TA is to find the edge to act on. As far as trading circles go, i've spoken with several people who trade successfully for a living and they do take chart patterns seriously. Its not tea leaves man. Andrew Lo - MIT - .... studied Double Top/Bottom and others... Best grip you can have is a repeating price behavior.. Bull excrement is a harsh vote.. In the light of modern research, i'd say its an uninformed vote. TA can work - TA alone won't make you money, yeah that's right.

TA needs to be objective. Period. Otherwise its unverifiable. Here's the mindbender for me: It is said price is somewhat random {Andrew Lo, A Non-Random Walk Down Wall Street}. It is said many technicians can't tell a real price chart from a randomly generated one. The bender: check out the Wikipedia page for Ito Calculus. There is a graph on an Ito Integral of Brownian (random) motion. Tell me that's not a Bullish Butterfly pattern in the middle (the low) of the chart at T 1.5. I find these Butterflies (also called M7s) in other time series that could be easily thought of (if inaccurately) as random ... such as total solar irradiance, crime rates, barometric pressure, % of men with beards, etc.. The weird thing is that M7 patterns do NOT have a random outcome.. they precede reversals in the time series 60%+ of the time (from what i have measured, far from the population).. even in 'random' data! Someone please explain that to me

I'll back this up with some performance reports if i can figure out how to post an image that's large enough to read

why is a large image i attach showing up as a thumbnail??

I use wave patterns for entry; as with practically all chart patterns the pattern can be used for stop loss placement. Say for an ABCD entry, entering on a reaction from D my stop loss will go 12 pips (not incl. spread) beyond D.

Definitely good numbers to know. I had a person who works with hedge funds tell me the idea for functionality is to have a total return/max DD(peak to valley) > 3 So i guess total return should be considered too.

ok, i'm not programming that one.

I think talking $$$ made is not as good as % on account. right? only makes sense, i could make a million trading with a billion, that is still the same as making 0.1% on a trade. All that said... biggest $ ever... five figures, stupid risk. Biggest loss ever, also five figures..:haha::doh: Losses like that are humility trades, and thankfully (and mindfully) behind me. On a percent return level - i see 2% - 20% on a single trade (i scale in, risking 0.25% per entry, up to maximum heat of 2%, trail out)

SR.. There's that newer volatility based S/R. I like swings and Fib clusters.. The best/most reliable is a special case - a lot of you have seen it - where price is moving in a channel (i notice it particularly in bear channels) and price will come up to the resistance TL.. often couples with candles, momentum signals, even cycles. That's probably the best trade, just not terribly frequent. Generally... s/r for me is TL s/r. INDICATOR. Ah, right. In the indicator case... i hear market profile is worth looking at?? Maybe getting some order book numbers archived and correlated to prices over time?

That's using commissions incl, the 2 days in a row close < open exit.. 1,000 shares/trade Entry - we can redo with a better definition of trend. i know one when i see one.... in hindsight. We must have rules for moving forward. This test used ADX(14) > 25, ADX(14) > ADX(14) for the last 3 bars, and current close greater than the close 15 bars ago(a stab at 'rising prices'). Better ways to define a trend probably exist, this was quick and dirty. Optimizing between 1 and 30 bars (days) for close to close comparison (to define the uptrend), best results over this 20 year period were from the 10 day lookback. Actually did turn out profitable on SPY, though choppy Ok... that was intriguing so i carried over same settings, no optimizing to past 15 years EURUSD Daily, and to my surprise it worked. The return was only 40% since 1998, but hey Ridiculous, can't believe that worked. Hell, that was easy lol - my hat's off to you sir

simple, not easy. how do you define uptrend? I'll code it up, see what comes out.

I sleep around FX. I'm in the EST zone, so i'm up at midnight, asleep 4-5pm.

As others have stated, it comes down to what you are practicing. Agreed. To maximize profit, sanely, you must act on opportunity as often as possible and protect capital.. goes without saying that you must have a mathematically valid way to profit trading.. And IMO you must utilize multiple timeframes/price action levels. For example, I have resistance on 1hr, signs of a turn using 20m data, then trend change on a 5m - i will start shorting using patterns on a 2 pip kase bar chart... My approach is to scale in, and get all out (either trailing or target). Scale in, all out in traded this way definitely 'maximizes profit' to my satisfaction.... my two cents

1) Wild West.. If you have the compound to billions drive, its the best market - liquidity in the true interbank, i mean 2) Use currenex, icap, etc 3) pay to play.. but if the broker is ripping you off - switch 4) there is orderbookfx i think they use the EBS data which is something like 1/5 of the interbank if you aren't that highly capitalized i feel your pain. If you've got 10k USD at least - check out IB. I've been happy with the fills in FX with them, vast majority of the time, good spreads too. NOT AFFILIATED, IB, etc... just a trader Peace

ABCD and MW - trade with trend, buy low sell high. These patterns (and a few others) get you in near the turns in price, thus allowing better R:R

Hi, Pyramiding can work, and can beat a non-pyramid version (during a trend). So the issue becomes manifold: 1 - How good is your trend ID? How many stop outs are part of trading trend situations that work? How many stop outs indicate the trend ID setup is probably going to fail? This can be a money mgmt stop loss/stop trading factor 2 - What timeframe is your trend on? Are you using multiple timeframes? For example this morning in the EURUSD, using currency strength to identify up and down swings on 15m data, there were 3 trade worthy swings from 315 EST. Pyramiding entries on a 20 tick chart gave about 3% return on capital on each one potential. The buy and hold was around 1/10th of that. 3 - On what basis are you pyramiding? Meaning, what is your edge? If you are just pyramiding off ATR entries, what kind of stop loss range are we talking about? It is preferable to buy low and sell high, you want each entry to have built in reward to risk advantage in pyramiding. If your reward to risk is no better with pyramiding, then there is no point, which may explain the OP's findings

Better still, combine chart patterns and the contextual filters that ForexTraderX has listed. Using patterns like ABC or ABCD within the context of say buying a support level or VSA confirmed breakout, etc will make systems with different probability distributions than if you just traded the patterns alone. Generally pattern allows a tighter stop loss, thus skewing reward to risk on successful trades. More stop outs can be part of the package. More important though is money management.

Chart Patterns are definitely still useful - all the more if they are able to be identified via computer. The Diary of C. Trader book is good. The author records his trades for about one year, earns about 60% Return using classical patterns and discretionary trading. The fascinating thing is his accuracy was really low - near 10% at times if I recall.