MTPredictor

[Euro_d]

Interesting that after esignal took over the reviews are very negative. And the GET guro, Tom Joseph is not Dynamic Trend.

Again not cheap at $250 / mth. Am not sure this includes the data feed.

Time to pause for thought!

 

[Pat494]

MTP do post on their website their previous day's letter and comments, so you can see how they did. As it's free they probably wont mind if I post the address. It is:-
http://www.mtpredictor.com/pricing/FreeReports.html
They have admitted recently that their strike rate is low ( about 1/3 ) but claim that they get the big moves and thus make profits. I am still undecided though.
This is the only company I've ever heard of that has had the courage ( or ability ) to post recent results
( I have no connection to them )

 

[Jack o'Clubs]

After a long break for holidays and thinking time, I've returned to my desk and bitten the bullet, buying MTPredictor. I shall update with how successful I find it over the next month or so.

 

[Pat494]

After a long break for holidays and thinking time, I've returned to my desk and bitten the bullet, buying MTPredictor. I shall update with how successful I find it over the next month or so.

Great - I look forward to seeing your results.
By my calculations they did make a profit in August

 

[Jack o'Clubs]

A couple of early observations: First, customer service from MTP and Paritech (through whom I ordered MTP and picked as my data provider) are first rate - both followed up my orders with calls to make sure I'd got what I wanted, had installed it and got it up and running, and whether I had any questions. Both made it clear they were very happy to provide ongoing support.

There are something like 550 pages of training manual plus a stack of video tutorials on how to get the best from the software, so it's going to take a while before I feel completely au fait with it. One thing to make very clear is that it is not a complete 'black box' which provides set-ups requiring no thought from the trader. What it does is quickly filter for potential set-ups, delivering a list of those worth investigation. As an example, a trawl through the US market EOD data this morning brought up maybe 20 'possibles'. Of these, just one clearly ticked all the boxes and was a trade. About 15 were obvious duds. But a handful sit in a grey area between a 'classic' set-up and a clear 'no'. It's at the trader's discretion whether to act on them or not. Therefore one person using MTP is likely to have a very different P&L and list of trades to another - you still have to make decisions as a trader!

Another point is that if you're looking for the main set-ups, they don't crop up very often (on EOD data) - therefore you'll either need to consider a very wide range of markets, or allow more leeway in the interpretation of results which the software and training guide gives plenty of guidance around. Personally I'd rather search a bigger data sample and take the best quality trades, but my thinking around this may change as I work through the manuals. I'm running scans on futures and US and UK stocks and over the last two days, scanning maybe 15,000+ instruments, I've come up with two cast-iron trades.

So far? Too early to have a view on performance, but I am feeling quietly confident!

 

[Pat494]

Am I going barmy or what ?

Having a look at MTPredictor's website of the previous day's forecast and how they did - i was amazed to notice that their figures for the S&P and Dow seem to be substantially different from my own eg for 7/9/2006 they show in their graph for the S&P
O-1308.7 H-1313.7 L-1304 C-1307.4
Whereas I have
O-1300.21 H-1301.25 L-1292.13 C-1294.02
Please could someone check their figures before I query it with MTP ??
( dont want to look a complete twat )

 

[s&ptrader]

Looks like they are showing the new december contract and you are looking at
the September contract.

Regards

 

[Pat494]

Looks like they are showing the new december contract and you are looking at
the September contract.

Regards

Yes thanks. I was looking at the S&P500.

 

[Jack o'Clubs]

Some backtesting...

A quick follow up on some back-testing I've been doing. The program has a good back-test facility, although it's time-consuming and subjective as you have to assess the set-ups it filters (see post #25). Hence this should be read as my back-test results - another person would likely have chosen different trades.

I've back-tested 7 months of data from the start of 2006 on 100+ futures contracts. That's not a huge time-frame, but it does take a long time to do... This initiated 29 trades (ie about one a week), of which 7 weren't filled. Of the remaining 22, 9 were stopped out for a loss and 3 at breakeven. So, 10 winning trades, a higher percentage than the generally quoted 33% hit-rate for the software. This might either be because of my fairly strict set-up acceptance criteria, or because my data sample is relatively small. Encouragingly, if X is my average loss, the average profit was 3.5X (incl. slippage and comms). Average timespan for winning trades is three weeks. This seems to me to be the crux of the method: money and risk/reward management. I'll be very happy with it going forward if I can use it to generate 45% winners and 3.5x risk/reward on winning trades. We'll have to see.

No meaningful results from my 'live' system yet - I've selected five trades since last Wednesday. Paper-trading for the time being until I get a good feel for the set-ups. Plan (assuming it all seems to work) is to go live at the start of next year after some thorough testing, since with relatively few set-ups and successful trades averaging three weeks it clearly needs a bit of time to feel comfortable with its results, and then I'll run it alongside my existing method. For similar reasons I won't update this post with ongoing results for a while.

 

[Pat494]

Thanks for testing the MTP software. I am still undecided. I normally just look at the S&P 500. I imagine I would have had very few trades, say in the last six months, If 100+ futures only produced 29 trades. One index might not produce a single one ? I wonder what the win ratio is if the criteria is more relaxed to get at least one trade per week ?
Anyway I look forward to seeing your results.

 

[*JDR*]

Can't help but point out this on their website:

"Identify low risk/high return trades"

They then go on to say...

"There is no hype with MTPredictor"

Really? What was the other comment if not "hype" ?

I think anyone promising low risk with high returns is making promises they cannot keep. Risk needs to be managed by the individual.

Sounds to me like any other Prediction trading software promising easy money. Nothing that cannot be done with a chart, a pencil and your own risk / position sizing calculations (there are plenty of good ones on the Internet)

 

[Jack o'Clubs]

Can't help but point out this on their website:

"Identify low risk/high return trades"
They then go on to say...
"There is no hype with MTPredictor"
Really? What was the other comment if not "hype" ?

Well, that is what it does. It idenfitifies possible trades where your profit target is more than 3 or 4 times the stop-loss. Isn't that a definition of low risk / high return? Of course the strike rate isn't 100%, it seems to be about 1 in 3 for most users. The training manual (downloadable online for nowt) is very clear on the risk/reward methodology (you'll find Chapter 8 interesting given your thread on random stock selection and money management).

Nothing that cannot be done with a chart, a pencil and your own risk / position sizing calculations (there are plenty of good ones on the Internet)

Absolutely true. But I can scan the entire market in a few seconds for the sort of set-ups I want to trade - it would take me hours with a pencil and charts! The reason I spent my money on this is that it isn't a black box - basically it's an Elliot/Fib based screener combined with a good money management system. Is it expensive for that? Don't think so - how many successful trades do you need for it to pay for itself?

 

[*JDR*]

Don't get me wrong. I am not having a go at you for buying it. I'm going to have to buy some software eventually also.

I just took exception to their low risk comment. Especially since they cannot guarantee a 3:1 risk return.

I agree that although it is theoretically possible to do it with a pencil and some paper, in reality, doing backtesting of a system for 10 years and then applying that to the current market requires something more than my calculator and some graphing paper !

Don't know if you know yet. Does it enable you to back test a portfolio over time and apply risk & money management to that ?

 

[Jack o'Clubs]

Don't know if you know yet. Does it enable you to back test a portfolio over time and apply risk & money management to that ?

It does - see post #29 in this thread. The problem is that it's very time consuming (testing futures back to the start of this year took a few hours), because to do it properly you have to look at each day individually and choose which of the trades it highlights that you would actually follow. You don't trade all set-ups, there's a degree of subjectivity involved - if you have a look at the training manual which you can download from their website, you'll see what I mean - page 83.

 

[Tahi]

Have been following an online trading journal posted by an MTPredictor user.. may be of interest to some.. http://www.forexfactory.com/forexforum/showthread.php?t=5624

As to similarities to Dynamic Trader and Advanced Get it is hardly surprising for if you look at their home page you will see that the developer worked for Dynamic Trader and both the developer and the U.S rep both worked for Advanced Get.

 

[Pat494]

From the website given above by Tahi there is also an excellent article about MTP in pdf -
mtptrader.com/Active.pdf
The author seems to have stopped posting his journal for personal reasons. One point the article makes is that to use all the parts of the programme one needs both EOD and realtime versions - a lot of dough ! If i understand it correctly the EOD version hasn't got the advanced bit where you draw in your own lines ( pivot points ) !??

 

[Jack o'Clubs]

If anyone's interested I've started an MTP journal which you can access from the link under my details, left.

This is the chart of CRT which I mention as today's 'pick of the day'.

 

[ralpher]

If anyone's interested I've started an MTP journal which you can access from the link under my details, left.

This is the chart of CRT which I mention as today's 'pick of the day'.

Really appreciate you doing this Jack o'Clubs, I'm taking great interest.

Have you tried monitoring the major currency pairs and even some minor markets?

 

[Jack o'Clubs]

Really appreciate you doing this Jack o'Clubs, I'm taking great interest.

Have you tried monitoring the major currency pairs and even some minor markets?

The major currency pairs come up in the futures that I look through, but I don't scan the spot markets. Haven't bothered with minor markets because I'm getting enough opportunities through the US, and I'd need to source (ie pay for!) the data from somewhere. At the moment I'm generally getting more set-up signals than I would realistically trade, so don't feel there's a need for me to expand my markets further. Glad you're finding my journal of interest.

While I'm on here (and as I can't post images to the journal) - here's a shot of the trade I've taken today:

 

[Pat494]

Thanks to all those who replied to this thread. Have just finished part1 of their free tutorial which in fairness to MT is one of the best I have ever read. It is well worth reading just for article on risk/reward/money management.

My question to MT is this. Why charge $250 for testing your software. Over the years I have purchased software just as expensive but all came with a full money guarantee. All provided services such as email, videos, tutorials and phone. However nobody charged for them. If the intention of the charge is to deter those people who download your software and never bother learning to use it then in my case you have failed.

You have an impressive bit of software that used as directed would appear to have the ability to give consistent profilts. Dont go an spoil your customer relations by placing a charge to test it.

WELL did you buy it or not ??????????????