Bitcoin Forum
November 18, 2019, 10:57:28 AM *
News: Help collect the most notable posts made over the last 10 years.
 
   Home   Help Search Login Register More  
Pages: « 1 ... 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 [273] 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 ... 328 »
  Print  
Author Topic: Martin Armstrong Discussion  (Read 617638 times)
luckyplate
Newbie
*
Offline Offline

Activity: 46
Merit: 0


View Profile
June 05, 2019, 07:49:47 AM
 #5441

it seems like Scotland might break away from UK? They are different than the city folk in London?

What does the latest private blog Dow Jones in June says ??
1574074648
Hero Member
*
Offline Offline

Posts: 1574074648

View Profile Personal Message (Offline)

Ignore
1574074648
Reply with quote  #2

1574074648
Report to moderator
The Bitcoin Forum is turning 10 years old! Join the community in sharing and exploring the notable posts made over the years.
1574074648
Hero Member
*
Offline Offline

Posts: 1574074648

View Profile Personal Message (Offline)

Ignore
1574074648
Reply with quote  #2

1574074648
Report to moderator
1574074648
Hero Member
*
Offline Offline

Posts: 1574074648

View Profile Personal Message (Offline)

Ignore
1574074648
Reply with quote  #2

1574074648
Report to moderator
MA_talk
Member
**
Offline Offline

Activity: 189
Merit: 10


View Profile
June 05, 2019, 07:31:13 PM
 #5442

Armstrong has been using this construct to "predict" several things now:
From https://www.armstrongeconomics.com/world-news/war/cycle-of-war-religion/

Quote
It appears we are building toward 2021/2022 when things are going to begin to get heated on a religious basis.

First of all, it "appears".  He is not 100% sure, but it appears.

Then he attempts to give a date, but the date is in YEARS, and for TWO YEARS.

Then he said "going to begin".  But for all these kind of events and trends, nobody can ever precisely define the beginning of such trend.  Did it begin in 2021, 2022, 2023, 2024, 2025?

1. If a bigger magnitude event happens in 2021/2022, then Armstrong will tell you that he is RIGHT ON.
2. If a bigger magnitude event happens in 2023/2024/2025, then Armstrong will tell you that he is right about that things "begin" in 2021/2022.
3. If nothing significant happens even until 2026/2027, then you can bet that Armstrong will NOT mention a thing about his forecast failure, but instead he will be writing about any current interests at that time.  And you think any of the readers will remember his failed calls at that time?  No, all of them will be focusing on the current events at that time instead.

In essence, Martin Armstrong tries to catch on to any emerging trend here.  He is just a trend observer.

Anything coded in computer algorithm requires capturing the linguistic term in the form of some precise numerical values.  A beginning can be defined as some value exceeding 15% threshold, etc.  Of course, Armstrong is NOT giving you how he defines as "heated" religious issues, nor any threshold values.  Most likely, he is just trying to BS his way to another "excellent call".

That is exactly what he did with "peak of government confidence", and also plague peaking at 2019.  Well, except forecasting the plague from 2014 until now, with nothing much happens (instead of international intensity), he is choosing option#3 above, for the escape.

And the crowd/readers' feedback keeps applauding all of his "success".
psp777
Jr. Member
*
Offline Offline

Activity: 73
Merit: 5


View Profile
June 06, 2019, 03:09:35 AM
 #5443

So Armstrongs 1% rule worked , but it didn't if you wanted to trade it. Let me explain. Last Friday we elected two weekly bearish reversals on the DOW. The closing price was greater than the 1%. His rule simply states that if the election of the reversal is greater than one percent, prices will snap back to test the reversal. The idea then is to go long or short (depending on the direction of the elected reversal) with a stop on the other side of the reversal. Well, we snapped back to test the two weekly bearish reversals and any one with common sense would have been stopped out of the trade today with the upside move we witnessed.

So interesting the we did snap back, but disappointing that the weekly reversals were not stronger and did not act as resistance.

Side note: I do feel we will see some more weakness in the market and continue to fall completing C of the downward ABC pattern.
Alex4711
Newbie
*
Offline Offline

Activity: 17
Merit: 0


View Profile
June 06, 2019, 11:09:55 AM
Last edit: June 06, 2019, 02:21:06 PM by Alex4711
 #5444

You may need to look to the next reversal which is valid this week. This gives you an idea of the expected trading range. MA recommends in his video not to start trading with a closing > 1% but looking first what the market is doing.
etoimene
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
June 06, 2019, 11:47:53 AM
 #5445

Thanks Alex!
Which video exactly?

It is sooo difficult to spot all those details, and each seems important.

EDIT:
This one, I guess.
https://www.armstrongeconomics.com/armstrongeconomics101/training-tools/instructional-video-reversals-1-rule/

EDIT2:
No, not that one.
Can you post a link?

In than one he suggests entering on a retracement to reversal (but typically). You also have to look at timing perspective. Sad

DanB1
Newbie
*
Offline Offline

Activity: 65
Merit: 0


View Profile
June 06, 2019, 12:23:49 PM
 #5446

Regarding reversals and the 1% rule.

Today I asked the helpdesk the following question:
According to the 1% rule, if the closing is >1% than the elected reversal, the reversal will be retested.
Sometimes more than 1 reversal gets elected at a certain time frame, especially at the daily level. If that's the case will all reversals be retested again, or only the last elected one?
So for instance, in a market there are 2 bearish reversals at 101 and 98. The index closes at 95, will both reversals be retested because of the 1% rule or only 98?

Their answer was:
hard to say sometimes in a weak market only the last gets tested in a stronger market the highest. There is no Rule for that just experience
Alex4711
Newbie
*
Offline Offline

Activity: 17
Merit: 0


View Profile
June 06, 2019, 01:21:16 PM
 #5447


Yes, listen at minute 2:20 "wait for the reaction back to the reversal... before taking any type of action".
nattybear
Newbie
*
Offline Offline

Activity: 33
Merit: 0


View Profile
June 07, 2019, 06:36:10 PM
 #5448

Well, this week is turning into a groundhog event.

I've been following MA since November last year after a good friend extolled his accuracy in calling movements in the market. I'm now done with this little experiment of MA and Socrates.

In the last 6mths, we've had TWO significant market run-ups which MA has not only failed to identify but has also been bearish immediately before them.

In Dec (maybe Jan blog - I can't recall) there was a strong "DO NOT ATTEMPT TO GO LONG THIS MARKET" warning... The Dow subsequently put on almost 5,000 pts in a historic upwards movement... To borrow one of MA's catchphrases - that run itself (with some leverage) could be considered a 'trade of a lifetime'. He missed this one big time... Too bad if you were sitting out or even potentially short based on his perspective... Any follow up,  correction, or admission of getting that wrong? Nope...

More recently, across several blogs in May (I limited it to May because April's posts didn't have much info that was tradeable), MA claimed that if the weekly bearish reversals were elected then we should expect to see at least 23,700 (there were several 27K numbers - ~700 was the most common). Even if I'm generous with which day I choose to benchmark this against (i.e. from last Friday's close and not Monday's low) the Dow has run up ~1,200pts since then - or ~800pts from the reversals. So again, if you followed his advice you were either sitting out waiting for a 'restest' of the weekly reversals, OR more likely, short (with a significant run against you) plus the missed trading opportunity.

Rinse and repeat?

Granted - with the weekly reversals he has given two possible dates for a low to form in response to those being elected. In the May blogs, he highlighted either 10th June or weakness into July (you're allowed to pick which one for yourself - 50/50 odds - not bad). With the Trump/Mexico/China drama unfolding it's entirely possible that it falls apart and we see a sharp drop in the market... However, if you follow his approach then you have at a minimum missed a great trading opportunity in this week's quick run up and the subsequent decline, or even worse, nursing a sizable trading loss, praying that Xi publicly gives Trump the middle finger.

This short prediction now gets even more complicated because (as of Friday 2.21pm - trading at 26,023) there are presently 4 bullish daily reversals that could be elected today AND in MA's 6th June blog he screws you over further by stating that a close above 25,725 would signal a further rally...

Now - I'll give credit where it's due. The daily reversal levels have interestingly led to some trading opportunities and proven to be sticking points in price movements, however in my experience, they have not provided much more value than trading Decision Points.

It's been an interesting experiment but after 6mths - I'm done with it.

Note: FFS Marty - get your all knowing AI computer to figure out basic English grammar and use a damn spell checker... There are basic pre-written APIs out there that you can download... OR try out grammarly.com - it's free...
AnonymousCoder
Jr. Member
*
Offline Offline

Activity: 168
Merit: 2


View Profile
June 08, 2019, 05:31:30 AM
 #5449

Hi,

I have been Socrates subscriber for many years since the service started, including the more recent Pro level. I have also kept record of thousands of Reversals on the Daily, Weekly and Monthly level. I have back tested them all. Even under consideration of the most sophisticated but complicating "Superposition Events" where Reversals are negated by signals that come out of the blue even as acknowledged by his staff, my back tests across all these signals reveal that the accuracy is 50% which is worthless. So they are effectively random signals, although they are based on logic as they are trend following signals that are positioned near support and resistance values. I have also back tested about 100 instances of another daily level signal "False Reaction" that catches a bounce on a daily level that appears to go too far back up from the previous temp high. It failed in roughly 50% of the cases.

When confronted with these issues, Armstrong staff has a simple strategy which is backed up by the overwhelming ambiguity of the system: They basically say that following the reversals is just not enough, one has to consider all the other factors and signals such as timing, energy (which is a mysterious signal). So there is no computerization in spite of Armstrong's AI claims.

So not only as others posted does one need to dig deep into the details to get information - more so the fact is if you do that then you just waste more time to find out the inevitable that such technical systems simply do not work regardless of their complexity. I can say that because I have followed their advice and then the system still fails.

Armstrong claims that he is the inventor of Capital Flow Analysis. For this analysis, the Socrates service has provided from day one a geographic heat map that shows current and projected capital flow daily, weekly and monthly, projected and previous. So one would expect that this prominent feature, which is on the member home page, is at least of some value. But unfortunately, every school kid with some JavaScript knowledge can check out that the logic behind it is bogus (has been for years). If you care, get a trial membership and check out the JavaScript links. The animation effect of changing capital flows between current and projected on the same daily level comes from the error that the script swaps between weekly and daily data where you think it is both on the daily level. It is a fraud, completely bogus. The complicatedness is overwhelming, and the server back-end script that writes these automated reports produces a lot of conflicting garbage.

Kiwibird
Jr. Member
*
Offline Offline

Activity: 45
Merit: 2


View Profile
June 08, 2019, 06:52:59 AM
 #5450

Did anyone else see it??!

https://www.armstrongeconomics.com/markets-by-sector/interest-rates/trump-v-federal-reserve-why/

He's deleted his final paragraph from this blog post..I definitely read it 12 hrs ago and now I can't find it. 

From memory it said something to the effect that '..I should warn you that the Govt are trying to silence me again, this time once and for all.  They can kill me - it does not matter as it will not affect the outcome...'

I am definitely worried for him.  He should leave the US and move to Asia..!
Alex4711
Newbie
*
Offline Offline

Activity: 17
Merit: 0


View Profile
June 08, 2019, 07:55:23 AM
 #5451

From my side I can only states that the bearish move on Nasdaq was properly predicted by the election of the weekly reversal of Mai 17th, the downturn move stopped at the border of another weekly reversal (June 3rd) which acted well as exit signal (not electing it) and now elected a new bullish weekly reversal that has been elected by more than 1% (risk of counter move is there). I have  now followed the reversals on Nasaq for one year and at least for me worked :-)

It is fair also to say that based on technical analysis alone the bottom of June 3rd would have been expected as it was exactly the minimum of March 8th. So no reversals would have been required to predict a proabability of bouncing.
nattybear
Newbie
*
Offline Offline

Activity: 33
Merit: 0


View Profile
June 08, 2019, 08:28:39 AM
 #5452


Armstrong claims that he is the inventor of Capital Flow Analysis. For this analysis, the Socrates service has provided from day one a geographic heat map that shows current and projected capital flow daily, weekly and monthly, projected and previous. So one would expect that this prominent feature, which is on the member home page, is at least of some value. But unfortunately, every school kid with some JavaScript knowledge can check out that the logic behind it is bogus (has been for years). If you care, get a trial membership and check out the JavaScript links. The animation effect of changing capital flows between current and projected on the same daily level comes from the error that the script swaps between weekly and daily data where you think it is both on the daily level. It is a fraud, completely bogus. The complicatedness is overwhelming, and the server back-end script that writes these automated reports produces a lot of conflicting garbage.


Thanks for the tip. I always discounted any of his Capital Flow Analysis or comments. There are no data points that he can use to track accurately on  a daily basis...

As with many of areas, even if we extend some faith here, his claims that international capital (he seems to like Euros) was flooding into the US and that was propping up the Dow is disconnected from reality and industry. Unless all this capital was parked in Europe and invested in Dow tracking investments there's simply no logical case that all the UHNW investors/institutions would sell off investments in Euros, wear the FX risk, and plough it into the DJIA.
nattybear
Newbie
*
Offline Offline

Activity: 33
Merit: 0


View Profile
June 08, 2019, 08:35:35 AM
 #5453

From my side I can only states that the bearish move on Nasdaq was properly predicted by the election of the weekly reversal of Mai 17th, the downturn move stopped at the border of another weekly reversal (June 3rd) which acted well as exit signal (not electing it) and now elected a new bullish weekly reversal that has been elected by more than 1% (risk of counter move is there). I have  now followed the reversals on Nasaq for one year and at least for me worked :-)

It is fair also to say that based on technical analysis alone the bottom of June 3rd would have been expected as it was exactly the minimum of March 8th. So no reversals would have been required to predict a proabability of bouncing.

Agreed - this is what kept me engaged for more than the first couple of months. The reversals do seem to provide some useful price points.

As time went on I saw several weekly reversals calling for Dow movements which turned out to be completely wrong. So while the daily levels were most useful as a confirmation tool I didn't find them any more insightful than decision points.
DanB1
Newbie
*
Offline Offline

Activity: 65
Merit: 0


View Profile
June 08, 2019, 10:36:14 AM
Last edit: June 08, 2019, 10:47:48 AM by DanB1
 #5454

From my side I can only states that the bearish move on Nasdaq was properly predicted by the election of the weekly reversal of Mai 17th, the downturn move stopped at the border of another weekly reversal (June 3rd) which acted well as exit signal (not electing it) and now elected a new bullish weekly reversal that has been elected by more than 1% (risk of counter move is there). I have  now followed the reversals on Nasaq for one year and at least for me worked :-)

It is fair also to say that based on technical analysis alone the bottom of June 3rd would have been expected as it was exactly the minimum of March 8th. So no reversals would have been required to predict a proabability of bouncing.

Agreed - this is what kept me engaged for more than the first couple of months. The reversals do seem to provide some useful price points.

As time went on I saw several weekly reversals calling for Dow movements which turned out to be completely wrong. So while the daily levels were most useful as a confirmation tool I didn't find them any more insightful than decision points.

You are right, the reversals are useful up to a point and often you see support/resistance around these price points. I have often traded these both moving with the flow when elected or as a counter move when not elected. From my experience around 60/70% of the time it works (which is okay but nothing more than that). So I combine it with macro data I buy (in this case from Hedgeye, also not perfect but gives me more insight on where we are) and my own technical analysis.

My experience is not to trade too often on reversals, only when the reversals say so, in combination with the arrays (they do change all the time but offer some insight), macro info and technical analysis.
So Socrates is more like an extra tool but not leading.

Regarding MA blogs. He always say what COULD happen. The annoying part is that when he's wrong he never rectifies himself. I know from his staff that a lot of people contact them on those occasions, but they like to see it as hate mail rather than people asking for clarification after an incorrect statement.

I have also requested them several times to make a video where Martin makes some live trades, and share it afterwards, just so we can learn from his decision making.
It might be a nice idea to share some trade/investing info so we can learn from each other. For instance, I trade, amongst others, the SP500 and EUR/USD so if there are others who would be interested we can share thoughts on this.
bikefront
Member
**
Offline Offline

Activity: 428
Merit: 21


View Profile
June 08, 2019, 03:31:02 PM
 #5455

I was going to do a Reversal analysis but if things are being deleted from his site, there is no point, as it puts the validity of the data to question. I think Armstrong/StrikeEagle needs to go to a doctor before going to Asia. I don't mean it in jest, but he really seems to believe in his own outlandish claims. A lot of the data he uses seems to be based on faulty premises, or has holes, and so on. He may even think we're government agents trying to discredit him, or try to spin it that way. Also, he called for an Apple high some years ago (based purely on his opinion) and was wrong... Anyway, I'm mostly posting on Reddit now, under the name 'priceforecast' with live trades, setup explanation, forecasts etc. There are a lot of great traders on there to learn from and discuss things, so yeah. If cycles don't work consistently, and Reversals don't either, then there is just nothing that has any practical use. He did say that they intend to bring out something that will automatically trade for traders, so who knows what that'll turn out to be, if it ever does come.
Alex-11
Newbie
*
Offline Offline

Activity: 112
Merit: 0


View Profile WWW
June 08, 2019, 08:38:13 PM
 #5456

So Armstrongs 1% rule worked , but it didn't if you wanted to trade it.
Maybe it doesn't work each time, but if it worked 7 out of 10 times, it would still be good, right?
Also you could loose 7 times small amounts and have only 3 win's, but they are big enough to cover the 7 loosing trades and make enough profit on top, it would be fine. I haven't tested this with the 1% rule, but this is often times a strategy in trading which is supposed to work with certain methods. It "only" need to be observed long enough to know if it works on not.


Did anyone else see it??!
yes, I did.


all these signals reveal that the accuracy is 50%
Maybe your testing methods has some problems. I've done a reversal back test on 50+ samples and it had more than 50% success rate. This is only on Reversals. Very black and white. Buy and sell signals.
http://armstrong.forumprofi.de/showthread.php?tid=78
How did you backtest the reversals?  Only on elected reversals (like I did when you follow the link)?


It is fair also to say that based on technical analysis alone the bottom of June 3rd would have been expected as it was exactly the minimum of March 8th. So no reversals would
have been required to predict a proabability of bouncing.
Maybe, but reversals could give you additional confidence to enter/exit/omit a trade.
s29
Jr. Member
*
Offline Offline

Activity: 71
Merit: 1


View Profile
June 09, 2019, 11:58:03 AM
 #5457

Well, this week is turning into a groundhog event.

I've been following MA since November last year after a good friend extolled his accuracy in calling movements in the market. I'm now done with this little experiment of MA and Socrates.

In the last 6mths, we've had TWO significant market run-ups which MA has not only failed to identify but has also been bearish immediately before them.

In Dec (maybe Jan blog - I can't recall) there was a strong "DO NOT ATTEMPT TO GO LONG THIS MARKET" warning... The Dow subsequently put on almost 5,000 pts in a historic upwards movement... To borrow one of MA's catchphrases - that run itself (with some leverage) could be considered a 'trade of a lifetime'. He missed this one big time... Too bad if you were sitting out or even potentially short based on his perspective... Any follow up,  correction, or admission of getting that wrong? Nope...

More recently, across several blogs in May (I limited it to May because April's posts didn't have much info that was tradeable), MA claimed that if the weekly bearish reversals were elected then we should expect to see at least 23,700 (there were several 27K numbers - ~700 was the most common). Even if I'm generous with which day I choose to benchmark this against (i.e. from last Friday's close and not Monday's low) the Dow has run up ~1,200pts since then - or ~800pts from the reversals. So again, if you followed his advice you were either sitting out waiting for a 'restest' of the weekly reversals, OR more likely, short (with a significant run against you) plus the missed trading opportunity.

Rinse and repeat?

Granted - with the weekly reversals he has given two possible dates for a low to form in response to those being elected. In the May blogs, he highlighted either 10th June or weakness into July (you're allowed to pick which one for yourself - 50/50 odds - not bad). With the Trump/Mexico/China drama unfolding it's entirely possible that it falls apart and we see a sharp drop in the market... However, if you follow his approach then you have at a minimum missed a great trading opportunity in this week's quick run up and the subsequent decline, or even worse, nursing a sizable trading loss, praying that Xi publicly gives Trump the middle finger.

This short prediction now gets even more complicated because (as of Friday 2.21pm - trading at 26,023) there are presently 4 bullish daily reversals that could be elected today AND in MA's 6th June blog he screws you over further by stating that a close above 25,725 would signal a further rally...

Now - I'll give credit where it's due. The daily reversal levels have interestingly led to some trading opportunities and proven to be sticking points in price movements, however in my experience, they have not provided much more value than trading Decision Points.

It's been an interesting experiment but after 6mths - I'm done with it.

Note: FFS Marty - get your all knowing AI computer to figure out basic English grammar and use a damn spell checker... There are basic pre-written APIs out there that you can download... OR try out grammarly.com - it's free...

Great commentary, but it seems Marty's fans disagree with you:

https://www.armstrongeconomics.com/armstrongeconomics101/ecm-armstrongeconomics101/the-next-cycle-in-the-ecm-beginning-january-2020/

Quote
QUESTION: Martin,

I am a huge fan of yours and have followed your blog for probably 8 years now, I watched your many predictions using the AI models and have been amazed by their accuracy.

https://www.armstrongeconomics.com/armstrongeconomics101/economics/the-future-which-door-to-enter/

Quote
QUESTION: Good day Mr. Armstrong, I have been a student of yours since 2001 and am always amazed at your work.

 Cheesy
AnonymousCoder
Jr. Member
*
Offline Offline

Activity: 168
Merit: 2


View Profile
June 09, 2019, 01:46:00 PM
 #5458

Some have commented regarding "back testing". I must admit that it is actually the bad term because his system cannot be back tested by us because the system as such is secret, not available for inspection by a 3rd party. The next best thing we have is to take all the predictions that are available and check them after the fact. Hundreds if not thousands. Build an open source database of his reversals. So far, I have not seen such analysis. So far, I have not seen anyone doing this. It is a full time job for years.

Why is it a full time job? Because not only is the system secret, but as others pointed out, it is not possible to obtain historical data. So one has to build historical data. Collect from Socrates every day and, depending on the subscription level, especially every Thursday and Friday the Weekly Reversals before and after election. Use some 25 or more symbols, otherwise this is a waste of time.

The data we get is in the form of textual reports at best. From these reports we have to extract the reversal values and the close values, because Socrates uses proprietary integer format not the actual market close value.

Then after a couple of years worth of full time work, we have a database of a few hundred or thousand reversals that we can check against the next period close. That is still not good enough because we need to also discount the 1% rule and the superposition events.

With all that, we will find that the reversals have a success rate of roughly 50%. Anything else is cherry-picking.

Now here comes another twist. Armstrong will say you can't do testing like that. He will say one has to use the daily timing array to optimize the closing of the trade before the end of the period. Can you see that this is all nuts? Whenever one tries to be systematic, there is a catch for evasion. So let's do that. If we close the trades not on the end of the trading period but before, then the success rate will NOT improve because the daily forecast arrays are noisier than anyhing else, and they are ambiguous in a way that there are often multiple turning points per week. So we are back at 50% at best, and that is where the system stands. Anything else is wishful thinking.

There is an opportunity for Martin Armstrong to build a system that is better than that. But he hasn't done it. Now I am giving you an example that illustrates this fact. Martin Armstrong disputes the Global Warming dogma as pseudo science which is a cheap thing to do. He is actually right. But he is misleading us when he says that his computer predicts his climate theory. His computer created a Weekly Bearish Reversal 4260 for Wheat Futures /ZW on Friday 2019-05-10 elected at 4246 clearly within the 1% rule. That was the major low. Future close 2019-05-17 was 4650 up 9.9%, total disaster. One can confidently say that Wheat is up since then in spite of a prior global glut, due to crop failure in the US from colder than normal weather conditions.
pereira4
Legendary
*
Offline Offline

Activity: 1568
Merit: 1156


View Profile
June 09, 2019, 02:44:24 PM
 #5459

It would be great if Martin Armstrong open sourced Socrates. What's the big secret about it? If it's so good and it works, let people compile it themselves, and he would end up making MORE money than he would ever make by charging subscriptions into some obscure future-predictor. Pretty ridiculous if you ask me. Believing he will keep being right in the future is delusional at best. I think he is just a decent fundamental analyst, that's all. I don't believe in Socrates, Hyperwaves, and other so called "methods" to predict the future. Open source it or bust.
bikefront
Member
**
Offline Offline

Activity: 428
Merit: 21


View Profile
June 09, 2019, 03:55:02 PM
 #5460

Some have commented regarding "back testing". I must admit that it is actually the bad term because his system cannot be back tested by us because the system as such is secret, not available for inspection by a 3rd party. The next best thing we have is to take all the predictions that are available and check them after the fact. Hundreds if not thousands. Build an open source database of his reversals. So far, I have not seen such analysis. So far, I have not seen anyone doing this. It is a full time job for years.

Why is it a full time job? Because not only is the system secret, but as others pointed out, it is not possible to obtain historical data. So one has to build historical data. Collect from Socrates every day and, depending on the subscription level, especially every Thursday and Friday the Weekly Reversals before and after election. Use some 25 or more symbols, otherwise this is a waste of time.

The data we get is in the form of textual reports at best. From these reports we have to extract the reversal values and the close values, because Socrates uses proprietary integer format not the actual market close value.

Then after a couple of years worth of full time work, we have a database of a few hundred or thousand reversals that we can check against the next period close. That is still not good enough because we need to also discount the 1% rule and the superposition events.

With all that, we will find that the reversals have a success rate of roughly 50%. Anything else is cherry-picking.

Now here comes another twist. Armstrong will say you can't do testing like that. He will say one has to use the daily timing array to optimize the closing of the trade before the end of the period. Can you see that this is all nuts? Whenever one tries to be systematic, there is a catch for evasion. So let's do that. If we close the trades not on the end of the trading period but before, then the success rate will NOT improve because the daily forecast arrays are noisier than anyhing else, and they are ambiguous in a way that there are often multiple turning points per week. So we are back at 50% at best, and that is where the system stands. Anything else is wishful thinking.

There is an opportunity for Martin Armstrong to build a system that is better than that. But he hasn't done it. Now I am giving you an example that illustrates this fact. Martin Armstrong disputes the Global Warming dogma as pseudo science which is a cheap thing to do. He is actually right. But he is misleading us when he says that his computer predicts his climate theory. His computer created a Weekly Bearish Reversal 4260 for Wheat Futures /ZW on Friday 2019-05-10 elected at 4246 clearly within the 1% rule. That was the major low. Future close 2019-05-17 was 4650 up 9.9%, total disaster. One can confidently say that Wheat is up since then in spite of a prior global glut, due to crop failure in the US from colder than normal weather conditions.

Anon, you bring up some good points. A few things I wanted to mention:
-Other than the project of getting and inputting Sorcrates data daily, we would have to go by the predictions Armstrong makes on his blog. So using the ones he posted on his blog is do-able, even though it would take a bit of time. This would give some degree of accuracy, no? As the Reversals he posts are in a variety of market conditions and seem to be a large enough sample size.

-Perhaps the superposition events may be discounted (maybe not, depending on the timeframe- eg Weekly Bearish elected and a Monthly Bullish- so it may be right on both) but the 1% rule only means a return back before moving again in the predicted direction. So we would be able to still give it a yes or a no if it got to the predicted point.

-Armstrong has actually said that the Reversals can be used standalone, without the arrays. However, the arrays cannot be used without the Reversals. He also says you leave money on the table by doing so but you catch the main move.

-I stopped being a subscriber some time ago, but last I saw, the Weekly Reversal numbers change in the Socrates textual analysis. So I don't know if they were wrong or invalid, or what.

There was another user who claimed 8/8 trades using a certain method with the Socrates Pro version- I wonder how it is doing now since.
Pages: « 1 ... 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 [273] 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 ... 328 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!