ProRealAlgos? Real or fake

• 02/23/2021 at 10:40 AM #162410
  zilliq

  No EUR/USD 1 mn if I remember but it doesn’t change a lot

  The stop loss is the strategy (you can read the code) is very important so with a little loss you can lose many gains

  If you see the picture of prorealalgos you can see the average loss is 4 times the average win, who is very big (Ok win/loss is high, but as we all know the past is the past, and not the future)

  That’s just my opinion

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  02/23/2021 at 1:15 PM #162431
  snucke

  what are you even trying to say zilliq ?

  everyone who sees the results of the strategy understands that it has big losses and smal gains

  

  snucke

  Participant

  Topics: 9

  Replies: 143

  Been thanked: 21 times
  02/23/2021 at 2:36 PM #162441
  Monobrow

  zilliq wrote:

  If you see the picture of prorealalgos you can see the average loss is 4 times the average win

  4x and a 90% win rate?

   

  where do I send my money 😬

  

  Monobrow

  Participant

  Topics: 15

  Replies: 157

  Been thanked: 23 times
  02/23/2021 at 6:52 PM #162471
  zilliq

  snucke wrote:

  what are you even trying to say zilliq ?

  everyone who sees the results of the strategy understands that it has big losses and smal gains

  Hi

  Exactly what you say

  rare big loss and frequent small gains

  Average loss = 4* average gain according to the picture of Prorealalgos

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  03/04/2021 at 2:57 PM #163133
  jmf125

  Quick question to the forum, at the end of the day has anyone enrolled and purchased one or all algos ?

  I just registered for the next trial starting in April. The algos have a very big loss/win ratio and the back test does not always reflect the figures listed on the web site.

  However, these guys are trying to be as transparent as possible it seems, and provide far more info than most other algos being sold on the market.

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
  03/04/2021 at 3:05 PM #163134
  jmf125

  I guess you did not take the plunge.  Did you keep monitoring after that ?

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
  03/04/2021 at 4:55 PM #163137
  zilliq

  Yes they are more transparent than most of website but, in my opinion, we haven’t all answers and elements to conclude

  Notably, there is a very big stoploss. So you can have many winning trades but with only a few you can lose all

  I use these algo on forex (yes I know it’s not exactly the same thing), BUT i see that it works well with some pairs and very bad with others

  Have a nice day

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  03/04/2021 at 6:03 PM #163143
  Mattias

  Completely agree, I think they are transparent. I took the trial, but did not get the same results as they (partly my fault). I have not bought any algos, and I have no intention doing it either. Way too big risk. And I believe Zilliq has posted a question or two they have not replied to (yet).

  1 user thanked author for this post.

  Bossebus

  

  Mattias

  Participant

  Topics: 9

  Replies: 76

  Been thanked: 19 times
  03/05/2021 at 10:15 AM #163180
  jmf125

  GraHal wrote:

  All my Trial PRA+ Algos traded … see attached.

  Hi Grahal,

   

  Are you still following these algos or have you abandonned ?

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
  03/05/2021 at 10:27 AM #163183
  Vonasi

  ProRealAlgos wrote:

  I see that the position sizes differ from what we recommend. We recommend using factorized position sizes i.e. to set position size so that the risk is equally divided by the algos. I will use the same example as I did earlier in the thread: Running 1 contract on DOW at ~30000 is a 15 times bigger position size than running 1 contract on OMX at ~2000

  I’m not a ProRealAlgos user or even tester but it does seem clear from what they have previously said that position sizing is very important when running their algos on different instruments to even out the risk/diversification. If anyone is not doing this and/or not using the algos on the correct instrument or time frame then how can they be in a position to compare or comment on the results?

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  03/05/2021 at 10:35 AM #163184
  zilliq

  Hi Vonasi

  I understand that diversification was used by them, but it doesn’t change the fact that each algo must be profitable after weeks or months

  Diversification is only another way to fit curve

  As I said I test on forex, on different pairs. Some works very well, some others not

  And as i previusly said, I don’t say they are bad algos, no (And thanks for you work Guys), but, we need some other datas to conclude, and personnaly I’m afraid by the very big stop loss (And so a very High win/loss ratio as usual)

  Have a nice day

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  03/05/2021 at 10:48 AM #163186
  GraHal

  jmf125 wrote:

  Are you still following these algos or have you abandonned ?

  Yes I will be taking up the next Trial and I will ensure I run position size as per the PRTA code.

  I had (incorrectly) changed (position size) through force of habit … as when folk on here try to elevate returns on for example on DJI at Price 30,000 .

  The PRTA Algos factorized position sizes is the correct way to go on, for example, OMX at Price 2,000.

  

  GraHal

  Participant

  Topics: 214

  Replies: 8826

  Been thanked: 2198 times
  03/05/2021 at 11:12 AM #163190
  Vonasi

  zilliq wrote:

  personnaly I’m afraid by the very big stop loss (And so a very High win/loss ratio as usual)

  …and as has previously been explained the win size/loss size ratio is completely irrelevant. What is relevant is that in a large sample of trades a strategy wins at least one penny more than it loses because then it can be classified as a winning strategy. It matters not one bit if it does this by winning 10% of all trades or by winning 90% of all trades.

  It seems that a strategy that has a high win rate and the occasional big loss is not your cup of tea…. so I would suggest looking for a strategy that wins 50% and loses 50% and all the wins and losses are the virtually the same size. Perhaps coin tossing?

  zilliq wrote:

  Diversification is only another way to fit curve

  Diversification is not curve fitting it is simply a way to spread risk and not put too many eggs in one basket and in a strategy that has lots of small wins and the odd big loss it is very important to reduce position size on instruments that are expensive to buy a piece of because otherwise one big loss on them could wipe out all your tiny wins on all your other strategies running on other cheaper instruments.

  Curve fitted diversification would be increasing position size on expensive instruments just because they did well in a back test just to make an equity curve look better.

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  03/05/2021 at 11:27 AM #163193
  jmf125

  GraHal wrote:

  jmf125 wrote:

  Are you still following these algos or have you abandonned ?

  Yes I will be taking up the next Trial and I will ensure I run position size as per the PRTA code.

  I had (incorrectly) changed (position size) through force of habit … as when folk on here try to elevate returns on for example on DJI at Price 30,000 .

  The PRTA Algos factorized position sizes is the correct way to go on, for example, OMX at Price 2,000.

  I have also registered for the next trial, and I am currently reviewing the algos. I think it is worth an evaluation, some of algos seem pretty good and specially in conjunction of each others.
  I am  currently looking at correlation of the algos, and so far it does not look too bad. But of course all in the past. And yes some have a big ratio loss/win.

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
  03/05/2021 at 5:19 PM #163221
  Nicolas

  Creating a strategy in the aim to win at the end is already curve fitting, just sayin’ 😆

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
• Author
  Posts