Past performance is no guarantee of future results?

WARNING: This thread contains discussion about something you might consider as stupid. You don't have to read it if you don't want to. And you really don't have to post if you don't want to. It looks stupid to do something you consider as stupid.

***

This is a continuation of my thread. Previously we discussed about the non-randomness movement of price. Now I want to discuss about one of the most famous disclaimer in the forex world:

Past performance is no guarantee of future results.

This is basically saying historical data doesn't have its value because the system might have done great in the past, but might fail in the future. Maybe, if you are using technical system. Maybe not, if you trade with understanding how the big boys trade.

On one hand:

Due to the ever changing nature of the market, the past is the past. There is nothing we can learn from it. Curve fitting past data into current trading won't work. There is no pattern that is consistent enough for us to exploit. Live price is everything. Past price is nothing.

On the other hand:

Due to the cyclical nature of the forex market, historical data gives tremendous value for the study of this cycle. It is essentially the footprint of the big boys. Market pattern, not candlestick pattern, or any other kind of pattern, happens daily. There is absolutely nothing new in the market. It's only a variation of a theme. Something we can all exploit consistently.

Does historical data have any value for live trading?

As usual I'll talk more after you share your opinion.

past performance is by no means indicative of future results. unless you gather all the past performance.

Now THIS is a very good topic, and also very key.

Now this is the big disclaimer that every EA vendor should have in font 14 bold at the bottom of their website.

"Past performance is no guarantee of future results"

my take:
choosing between the 2 hands...I'll go with hand number one.
past is the past.

but...once in the market...the very recent past should be paid attention to.
looking forward to hear from (almost) everyone else

Well you got to start from somewhere I guess, have a reference of some sort to build an understanding of how the markets generally moves. Using past data to lay down some general, adaptable rules to follow is not data mining or curve fitting. It can prove very useful to study the past but ignoring the present context is a lethal mistake. It is my belief that the movement of price cannot be integrated into one algorithm or at least if I am wrong and it can be done, I know for a fact that I cannot achieve such mathematical perfection. What I know can be done though is to derive vague and general guidelines to trading from past data. Those guidelines are by no means absolute truth but they help you making sense of what is happening in the market right now. As for the statement that says that past data does not guarantee future performance unless you find some arbitraging opportunities you cannot legally or ethically guarantee anything in trading or investing. For all we know even the so called risk free rate is not riskless.

"Past performance is no guarantee of future results"

Does historical data have any value for live trading? But People would backtest their system to judge whether it is worth trading or not.(I think this is more psychological in nature).

From my point of view Past performance will never guarantee profitability in the long run however if recording (Trading Journal) is done it does provide an increase in chances of being profitable.


-By the way Trading Journal is a form of recorded Past Performance which is given emphasis by almost all traders that can increase one chance in the market!!.

I don't understand this. Are you saying that if you gather all past performances, it helps with your future results?

Yes, exactly like I said. It's one of the most common disclaimer.

***

Disregard the guarantee part guys. I'm focusing on the value of studying historical data. The value of backtesting anything.

Think about it. If price is random, would any study of historical data be worth it? No. What's the point of studying something that is proven to be random, and proven to stay random in the future. BUT! We've proven beyond any doubt now that price is not random. Therefore studying historical data is worth it. As a matter of fact, it's the very thing that either make you or break you. Once you've studied the charts thoroughly, deep enough, and long enough, you can identify market patterns, and you will never ever think again that price is random. If the market pattern is not repeatable, I will not trade. Never. It's too risky. But when the patterns runs like clockwork (doesn't mean I always get it right), then trading becomes worth it.

At this point in time, your subconscious begin to trust these patterns, and you can begin to think like the market makers. And this is when your trading makes a leap forward. Exactly what happens in my case. These days, I open my terminal, within 5 minutes I can roughly tell where the big guys are out of 21 major and crosses pairs (I don't trade NZD pairs, 21 are more than enough already). This is btw, a game of cat and mouse. You have to identify which pairs are being worked on, because the pairs that are not being worked on will produce false signals. Banks don't have unlimited amount of equity they can commit to foreign currency. They can only handle a certain number of pairs at one time. Mind you investments banks don't only handle foreign exchange. They have to handle options, stocks, futures, indices, etc. Foreign exchange is merely a part of their portfolio.

Out of these 21 pairs, then I pick the 3 that are most likely to be used by the market makers, in my opinion. Determine the entry and exit, then the job is done for the day. Tomorrow, do the same thing. The market pattern is ALWAYS there, every day.

if you find stuff that work since the beggining of markets it is more likely that this is linked to the market itself and could be the best answer for the future. the probabilistic laws of charts .

For the reason I've explained in my previous thread, I will no longer teach or share my understanding of the market because by doing so, I'm risking myself getting cut off from this community that I love.

You keep talking about the "big boys" doing this and doing that. Large hedge funds, bank traders etc. speculate like we do - here's a hint, those guys aren't the big boys. The real big boys are managing your superannuation fund, and they have a completely different set of rules. Speculative traders can move price around on a 5 minute chart no problem but to really move price hundreds, even thousands of pips you need major investment trends on board. When you look at charts of historic price movements you are effectively tracking the past tendencies and capital flows of these guys - it doesn't indicate future results because they are not looking at the same picture they were then.

On the wider point (and simply put) - Technical Analysis is predicated on the central assumption/belief that what happened in the past can give us a clue to what the greater probability of what may happen in the future may be - ie that patterns repeat and these patterns of market behaviour repeat. Now, as we all suspect - no one can predict the future with any degree of accuracy (*) but repeating patterns (howsoever you determine them) can give a heads up as to what the greater probability is and this can be used to gain an edge in the market. It follows therefore that when such a pattern (s) does not repeat it is not possible for us to ascertain what the greater probability may be so it is best to stay flat until it/they do.

* There may be some that think they can predict the future and there may even be some that can...ie those whose ability to predict future price movement gives them an edge, and like any edge - one that is proven over may thousands of examples on the t/f's of interest to them in the market conditions they wish to trade and that the edge has been tested in. It may even be (like all edges) that a prediction strike rate of lower than 50% (winning predictions as a % of total predictions) is sufficient to make money across any given future sample because the gain available on the occasions of a winning prediction is sufficient to covert the losses on the losing ones and still provide an acceptable margin of profit (howsoever measured - it's relationship to risk encountered for eg.)

G/L