StockFetcher Forums · General Discussion · New Top Ranking G5-SCORE<< 1 2 3 4 >>Post Follow-up
jim_c_hill
61 posts
msg #28735
Ignore jim_c_hill
9/20/2003 4:53:35 PM

Geodepe
Thanks for adding the average number of matches data record. I note many of the top filters are now becoming penny stock trades at very low prices. As you develop your beta site I suggest you consider the following:
1. Sort feature for average number of trades, percentage for one week, etc.
2. A universal filter add-on that would take a filter and add a person's trading style: for example, average volume and price range, and add it to a published filter for testing with the universal add on.
3. A button that when pushed would take a given filter and automatically place it in the create box on stockfetcher, if a person has an account so stockfetcher would open and accept it.
4. And down the road...being able to specify the 120 day test period for dates.

You have done a great job and really helps in checking new filters. I know it is saving me many hours of checking, day by day.

Many thanks
Jim



jim_c_hill
61 posts
msg #28737
Ignore jim_c_hill
9/20/2003 6:09:26 PM

Also
at top of list of filters and results, the percentage change for a couple of the indexes over the time span of the filter analysis. If a filter can not beat the indexes, then maybe we should just invest in an index.

Also, perhaps for the same time space, the percentage change in the stockfetcher collection of stocks that span a given price range with no other limitation, say 0 to 1, 1 to 5, etc. Again, if a filter can not beat a collection of stocks in the price range of interest, then it may not be a sufficient filter to trade.

Jim



TheRumpledOne
6,358 posts
msg #28738
Ignore TheRumpledOne
9/20/2003 6:12:16 PM

Adrianf...

You missed the point entirely!

The point was that using RSI(2) can improve just about any filter.

All I did was take the #1 filter at the time and tweak it to make it better and better.

Look at all of the filters on the site that I tweaked with RSI(2) below X ... once the approriate value was found, it improved the filters results.

Keep your mind open. You can ignore me but for your portfolio's sake, don't ignore what RSI(2) can do for you.

For the record, HolyGrail and I trade OTCBB stocks all the time. Look at what NEOM, CYPT and TTXI have done recently.

MAY YOUR FILLS BE COMPLETE.



JoeGrossinger
165 posts
msg #28741
Ignore JoeGrossinger
9/20/2003 6:31:17 PM

adrianf, I would appreciate it if you would leave me out of your flame posts.

I looked at the standings and had to scratch my head a bit because I run the HG filters on Koliga.com in very realistic trading scenarios where stocks are bought and sold instead of just running them through a back tester or whatever.
On Koliga.com the HG filters like the one that is sitting in the number one postion on G5 and is quoted here are underperforming money losers.

price is below .05 and 60 day slope of the close is above 0 and rsi(2) below 1

So this had me wondering as to how they could look so good on G5 and lose money on Koliga.
I ran the above quoted filter through the SF back test and found the entire premise of the back test to be bogus and totally unreal.

I only checked from day 1 to day 5 and got these interesting results, which anyone can duplicate.

Date Offset 5 Date Offset 4 Date Offset 3 Offset 2 Offset 1
VPTI 0% ASPXQ 0% ASPXQ 0% CNCW 100% TALL 0%
ASPXQ 0% FCRZ 0% SPNT 0% ATDU 0% APPS 0%
LMRI 25% LMRI 25% VERI 100% VPTI 0% FCRZ 0%
EPAR 0% AUTC 0% MAHT 25% ASPXQ 0% FWGO 0%
APPS 0% APPS 0% HOMI 0% APPS 0% HOMI 0%

Now, please note the number of repeated hits, all of which need to be tossed, and then you have all those 0% performers.
This is junk.
On Koliga.com I run realistic trading models that actually buy and sell stocks.
If somebody just stumbled on G5 and grabbed the top filter, which is a natural enough reaction, then they would be badly burned using that piece of trash with actual money. Yet it holds the #1 position.
The figures speak for themselves.
Just grinding stuff through a computer a lot of times produces the most amazing trash.
I would not have caught it except for the fact that the HG filters do very badly on Koliga.com and that got me to wondering.

http://www.koliga.com/forum/index.php?c=5




TheRumpledOne
6,358 posts
msg #28744
Ignore TheRumpledOne
9/20/2003 7:17:43 PM

AdrianF...

Your filter:

Fetcher[Williams %R(14) is below -80 and Williams %R is rising the last 1 day and daily volume is above 100000 and price is below 4 and RSI(14) is below 40 and MFI(14) is rising the last 1 day and RSI(14) is rising the last 1 day]



had a G5 score of 4.95

Your filter with RSI(2) added:

Fetcher[Williams %R(14) is below -80 and Williams %R is rising the last 1 day and daily volume is above 100000 and price is below 4 and RSI(14) is below 40 and MFI(14) is rising the last 1 day and RSI(14) is rising the last 1 day and rsi(2) below 5]



had a G5 score of 32.36 and is in the top ten! It says #11 but the #1 & #2 and #3 and #4 are duplicates.

DOESN'T THAT PROVE ANYTHING TO YOU?

Why drive a horse and buggy when you can have a modern race car?


TheRumpledOne
6,358 posts
msg #28745
Ignore TheRumpledOne
9/20/2003 7:18:31 PM

Joe G.

The RSI(2) filter you run is not the one we use. You changed the parameters.

We are traders. We don't buy on Monday and Sell on Friday.
We take profit when it is there for the taking.

Look at TTXI, NEOM, and COPY this week.

Had I not taking profit on NEOM on Monday, I would be in a losing position.

TTXI provided a Double on Thursday if you took profit that day and a 50% gain on Friday if you entered and exited again.

COPY ran up 50% on Friday.

So either use the correct filter as posted and TRADE it properly or cut the crap.

You paper trade. HG and I have real money on the line. Anytime you want to post your confirmations against mine, you just let me know. You have my email address.

Till then... RSI(2) rules.

Admit it... RSI(2) improved your HolySmoke filter on the G5 site.

https://www.ipropel.com:8443/stocks/index.jsp

IT'S ALL THERE PEOPLE... IMPARTIAL JUDGING OF FILTERS.




JoeGrossinger
165 posts
msg #28747
Ignore JoeGrossinger
9/20/2003 8:38:05 PM

I changed nothing. I did a copy and paste from the #1 filter.
The entire backtesting notion is bogus as I have shown

Your claim that you day trade the HG filters is just as bogus because most of their day range is way too narrow.

If I want to day trade I use something designed for day trading.
show stocks where the day range is above 20% for the past 5 days

That one liner will get you stocks to day trade.
Every HG filter has been found to be an underperformer when tested in REAL LIFE and not some flawed and bogus back test that counts 0 performance as green and gives repeat after repeat hit.


TheRumpledOne
6,358 posts
msg #28751
Ignore TheRumpledOne
9/20/2003 11:31:14 PM

Joe you don't test in real life, you paper trade.

When you post real confirmations, then and only then, can you prove your claim that RSI(2) doesn't work.

Think about it. What you are saying is the computers and/or data are wrong.
In other words, had you bought the stock 2 days ago you would have lost money but the backtest says you would have made money. That just doesn't compute!

On 9/15, MRXT bottomed on RSI(2)=.065 and lower lin reg line. A buy at the open the next day at .061 would be .14 at the close on 9/18. More than a double.

On 9/15, Q bottomed on RSI(2)=1.34. A buy at the open the next day at 3.86 would have turned into 4.16 at the close 2 days later.

On 9/16, COPY was sitting on the lower lin reg line and rsi(2) was showing positive divergence. A buy at the open the next day at .55 would have turned into .77 at the close on Friday. That's a 40% gain in 2 days! And I posted that BEFORE the pop not after in this forum.

That's just 3 examples of what RSI(2) has done this week.

So your "real life" buying on Monday and selling on Friday has nothing at all to do with real life! Cause you should be looking at stocks daily and either protecting profits with stop losses or taking profits when you can.

Perhaps if you spent more time learning the LR/RSI system instead of trying to prove it to be no good, you just might make back that $100,000 you lost in the stock market.







TheRumpledOne
6,358 posts
msg #28752
Ignore TheRumpledOne
9/20/2003 11:37:56 PM

Joe G.

My claim I daytrade the filter is bogus?

NEOM on 8/25 RSI(2)= .364.

On 9/5 NEOM showed positive divergence on RSI(2) = 10.46.

I have been trading NEOM the past 6 days and will continue to do so.

What have YOU been trading, Joe G. with real money?




EWZuber
1,373 posts
msg #28759
Ignore EWZuber
9/21/2003 3:34:36 AM

Just for the heck of it I tried adding RSI(2) to this momentum filter,

RSI(15)is above 80 and RSI(15)is below 95 and price is between $2 and $15 and Average Volume(10)is above 100,000 and where Weekly Slow Stochastic(5,3)%K crossed above Weekly Slow Stochastic(5,3)%D within the last 2 days.

and got no returns at all. I suspect that, as I understand the objective of the RSI(2) parameter is for bottom feeding, it is incompatible with a momentum type of trading. With momentum you are looking for a stock where traders can't get enough of it, where with RSI(2) you are looking for a stock where sellers are exhausted, at least in the short term. This is what makes momentum trading such a powerful trading vehicle, particularly in a wildly bullish market like we have right now.

I also tried the RSI(2) parameter with this filter,

FIND STOCKS WHERE SLOW STOCHASTICS(5,3)%K IS LESS THAN 50 AND CLOSE IS NEAR 20 DAY HIGH AND CLOSE IS BELOW 20 DAY HIGH AND WEEKLY FAST %K IS INCREASING AND FAST %K IS INCREASING AND PRICE IS BETWEEN 1 AND 15 AND VOLUME(30) IS GREATER THAN 100,000.

and again got nothing. This filter is designed to find stocks that are very bullish and in great demand. It is demand that continually drives prices higher and I believe there is inherently less risk associated with high demand stocks as long as the move is not too mature.

I feel it is important to let the uninitiated know that there are few other trading styles that carry more risk than short term trading of penny stocks. You will likely get much more satisfactory results with longer term trades in quality stocks of companies that have excellent growth potential, high ROA and ROE, no or little long term debt, high fwd. P/E, insider buying, and possibly new innovative products and/or is hiring new personel.

Finding these companies checking the charts, then waiting in the tall grass for just the right moment to pounce on a technically favorable position.

Holding until a trend line that fits your timeline is violated.

I think that the arguments that I see on this thread are asking the wrong questions. Maybe it is more practical to look for ways to make good returns with the least amount of risk. I suspect one will find that there is a curve that is relative to price. Take BCON for instance, sure it made a big move but if you didn't realize that it was going to test resistance at a previous price level and likely bounce down quite violently, you could lose a ton of money. To play a stock like this you had better be on top of it every second and also be right in your entries and exits cause with these kinds of fluctuations there is a lot on the line. If you try to run several of these at once you might go insane. JMHO


StockFetcher Forums · General Discussion · New Top Ranking G5-SCORE<< 1 2 3 4 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus