Trading eminis with cycles

[theplumber]

An overly simplistic daytrade system that I use is volume bars and money flow with either directional movement color bars (Esignal) or trends setting color bars (Ensigns) . This is maybe the only consistent indicators I've seen for daytrading besides bid/ask trade volume analysis.

The arrows point to money flow diverging with price and waiting for the color bar change or not (depending on how aggressive you are) to enter the trade.

[theplumber]

Re: Questions for the plumber..

Originally posted by Doji3333
Hey Plumber...

Do you day trade or position trade? It looks like a possible combination of both?

So you're saying we should see a pretty huge sell-off over the next x months?? Any idea of the length of this sell-off in time - or can you just gauge by the price target?

It also looks like you are deploying many different strategies in your analysis.. I've been working on developing a system that is fully automated for my needs. I would like something that does not trade alot and something that tends to stay with the direction of the market.. I have something that looks interesting and backtests well. I've recently built it into an automated trading system and am testing it...

Enough about my objectives.. I'm kinda interested in how you use these unique indicators to trade?? Obviously, you appear to be trading the longer term trends - no?

Just to clarify things, A better calculation would be 1050 S&P between this fall and summer 07. I don't really have time to post a chart but will say this will be nothing more than a correction of the move since fall 2003 to May 2006 and will be followed by 1700 S&P before the next major correction. A time frame that large is only one variable, meaning a panic that brings price down to 1050 will also make a good bottom as well. FWIW I'll have a position long July 21 and will maybe look to get longer Monday. Maybe stopped out again who knows.

[Doji3333]

Update..

Plumber,

Thanks for the update. My system went long this AM and picked up $237.50 per contract.. My system is not longterm and does not hold positions overnight.. It trades ONCE per day and does a pretty good job..

It also applies to the YM and NQ as well..

YM picked up $175 today
NQ is still trading -$15 (oh well).

It's interesting that you are able to predict that far in advance. My system is nothing more than "pick a direction and GO WITH IT".. I guess the trick to my system is HOW I pick the direction.

Will post more soon.

BTW, last Friday was flat (actually lost $25)
Last Thursday I picked up 6.25 pts

[Doji3333]

EOD Update..

Plumber,

My systems ended up +$572.50 for the day trading one contract. Here are the results..

ES = +$237.50 (one trade)
YM = +$175.00 (one trade)
NQ = +$160.00 (one trade)

(not counting commissions)..

I'm hoping to continue development of the systems to make improvements. Any thoughts from the expert??

B

[theplumber]

Re: EOD Update..

Originally posted by Doji3333
Plumber,

My systems ended up +$572.50 for the day trading one contract. Here are the results..

ES = +$237.50 (one trade)
YM = +$175.00 (one trade)
NQ = +$160.00 (one trade)

(not counting commissions)..

I'm hoping to continue development of the systems to make improvements. Any thoughts from the expert??

B

Slow down there cowboy, an expert I'm absolutely not. Never have been and never will be, just profitable.

OK with that said is this real money on the line? It looks as if not with the commish not included. 1 trade a day? What if the market trades in a tight range? Still trade? My experience with systems outside of cyclical analysis (and risk coefficients as a filter) was not pretty. First was a basic moving average crossover. Think how many times it can cross as the market fluxuates at the very cross. I traded about 5 times, and this was based on 30 minute charts and a 1 point stop. Put me in 1 trade a month, waste of time. Then further added more indicators and shrunk the time frame down. Backtested results were unbelievable and VERY different when I let it loose to trade in real time. Again commish was about $50 a day. I guess to make a long story short real world trades (and that person sweating in the trade) is the only way to measure a system. If your right you should buy that island Mel Gibson is selling, or just print the money and go anonymous like J Simons. All the best.

Almost forgot, sold half my position and put stop at break even for the rest. Maybe add based on market conditions but probably not who knows.

[Doji3333]

OK - not an expert..

Plumber,

I'm sorry if I offended you. I was just asking for your opinion and given the length of this thread and your obvious skill, I thought you would be a good person to ask. Again, I'm sorry if I said anything to offend you.

Your answer, although, was clear.. "buy an island". lol

I'm still in the testing phase (I think I've mentioned that earlier). so I'm still paper trading with an automated order execution tool. I have my esignal code developed and just need to put the finishing touches on it. I'm not a big player, so don't expect me to be moving to an island yet (besides I kinda like living in the mountains).

Again, I was just curious as to your thoughts as we've been discussing theories... My view is I would rather take a single "opportunistic trade" each day than chase 1 pt trades all day. I also believe that a totally mechanical system takes all (or most) of the pressure off "sweating a trade".

Now, my system is still very much in the testing phase. I believe I might be onto something, but have not been able to verify much other than "so far so good".

My system does have a "cost averaging trade" feature that *sometimes* comes into play. This is again totally mechanical.

From my testing, the biggest issues I currently have are..

a. containing run-away losses
b. improving the system's mechanics
c. maintaining the overall approach to secure a long-term future

I'll post more as it gets closer to DONE. In the mean time, unless you object, I would like to continue conversing with you?

[Doji3333]

Rough day..

Today was a rough day for my little test system. Lost 6.5 pts in the ES today. Everything was looking good till that afternoon rally.. Oh well, it is still pretty consistent and I'm working on ways to try to contain things like this..

One thing I was thinking of is a time based loss containment system. Something possibly like..

allow 1~1.5 hours for a trade to move into profits if the trade executed before 10am (PST) - otherwise, maybe only 40 minutes.

If the trade has not achieved at least n pts profit, reduce the risk considerably and remove the second "cost averaging" trade from the desk.

I'll have to see how something like this works by visually looking thru the last 40+ days of data. Actually, all I have to do is look for the BIG LOSS days and see what happened.

Anyway, we're still hanging in there for the month - only down 2.5 pts for the entire month. Yeah, I know that is NOT GREAT.. BUT -

It's still up over 280 pts since March of this year.. Averaging 12.73 pts per week. Still very impressive in my mind..

I'll bet the Plumber kicked some a$$ today??

[theplumber]

I'm supposed to come up with these unique charts but not today, just one showing the S&P/10 year note (in blue). Structural problems are starting to show and are now being priced in. Like 2000? No but problems just the same. I'll stay long until stopped @ 1247 for the reason of thinking a flat to slightly up into year end. 2007 will be the be year. I don't think the housing market crashes but speculation will be taken out, which will hurt but housing is a function of liquidity and not rates . Just a rant . Demand is slowing at a fast pace as shown by the transports, so when the transports turn with the industrial look out above.

[theplumber]

Sold another 1/4 of the overall position today so only have 250 left. Not a bad 50 points I must say but the 60 minute chart shows weakness coming in the short term. Other than that the rally could pull back and rally to 1310-1325 or so , depends on how many stupid shorts there are. I moved the stop for the rest to 1267 to lock in profits. No one ever went broke taking a profit.

[theplumber]

Just doodling. Lets see how this works out for today, I took a small short for a trade with a trigger happy stop in case I'm wrong. I only like to daytrade with price and volume so moneyflow and directional movement on moneyflow is a good start, which is on top and bottom respectively. They are diverging with price . Just an observation and not trade recommendation.

[theplumber]

Originally posted by theplumber
Just doodling. Lets see how this works out for today, I took a small short for a trade with a trigger happy stop in case I'm wrong. I only like to daytrade with price and volume so moneyflow and directional movement on moneyflow is a good start, which is on top and bottom respectively. They are diverging with price . Just an observation and not trade recommendation.

didn't go as planned and was stopped out for a small loss. My day trading only accounts for 35% of my total take for 1 year as my swing trades are taking over the profit ratio. Still fun to daytrade though

[theplumber]

Stopped myself out @ 1304 today of my remaining long swing position.

[theplumber]

Should have stayed long but that's trading. Short 1360 ES . As always not advice.

[theplumber]

Originally posted by theplumber
Should have stayed long but that's trading. Short 1360 ES . As always not advice.

Got more short today with a stop 1385 ES. Hate being long in Sept and short in Oct but so be it.

[theplumber]

The top model I posted here sometime this year and below is the actual data for this year so far. The dip is there but not exact in date but that's how using cyclical models are forecast for a whole year. I loosened my stops but will be very quick to pull the trigger because I don't think there will be much year end weakness. I think 1400 may be possible for year end and a little above into next year. But ask yourself this, would you go long or short here? My answer is neither so I "think" this is a huge topping process for all of 2007.