Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Nope no slippage. When I write strategies I always use market orders to get around all the Ninja bugs, but when I run them live I always use GetCurrentAsk() or GetCurrentBid() instead. It will take me a while of real live testing to see how it really performs.
I've written 100 strategies that look just as amazing but turn out to be curve fitted. However, I will admit this is the first strategy that has not been based on a traditional type of indicator. This is also the only strategy that I've written (that I recall) that has only used one single indicator. And it's the only strategy I've written that uses any type of pivot or S/R lines, as I mainly don't trade them.
So it's new ground in that way. But I'll let you know if I'm rich in a couple weeks after it forward-tests some more. I really want more historical tick data to test against, there is a thread in VIP where I suggested a group-purchase of historical data and we could all split the costs...
I'll run a new optimization that includes a tick of slippage each way and see what happens. I'm using my own optimizer type that weighs the # of trades per day, SQN, expectancy and profit all in one, so it likes to gravitate towards higher frequency trading that results in large winners, we'll see what the slippage does to it.
Quick question Mike. If you ran the strategy whilst trading CL, that would give you an indication if it was successful or not?
obviously you're not going to really buy/sell real money. The reason I ask is, running the strategy against market replay isnt really going to give you the correct information, where as running it during market open will accommodate slippage?
Market replay is pretty worthless in my opinion for testing the real-world performance of a strategy. There are just too many bugs and limitations. Market replay is useful for trying to solve a problem with code, that's about it.
The only test that matters is trading for real. But before I trade for real with any strategy, I first do my absolute best to not curve fit the strategy. Then I do as much testing as I can stand (patience) in sim, and then finally cash if all goes well. Sim is pretty accurate for most of what I do because I really don't care about that extra tick and place all my orders @ the ask, so it's rare that sim is a real problem, but the sim engine is not the best...
Specifically, slippage should work in theory on both sim and market replay. But real world, they both suck. Cash is king.
For more strategy-specific discussion let's continue this in this thread:
If there are several of us who are interested in purchasing tick data from a commercial provider (HISTORICAL TICK DATA ? Historical tick data for stocks, equities, futures, options, indexes and indicators! ( for instance) then maybe …
Ok last post about the strategy... you can read more on strategies on the other link I posted two posts ago.
This is CL, I excluded pre-market and after hours, I filtered for news (bars that come too fast) and I included 1 tick slippage each way. Commission is already included on above results as well.
Now you you have done it. You have got me extremely PO'ed. Seeing those results I was thinking "Way to good to be true". Now I am thinking "What if". Was going to take a break from serious NT coding until the first revision of NT 7. Not any more. So now I will attempt to duplicate what you have done. I am sure our logic will be different which is a good thing because we might want to compare results. Nothing wrong with taking a little from this and a little from that in an attempt to create an AutoTrader that might actually work. Chances are slim but like a moth to a flame I have to go for it.
With MML's, the example on post 1 is great, my question is, what happens if a trend never touches a line? lets say it goes between 2 lines i.e.
At what point would you sell/buy? Also, on the purple line, according to post 1, we should of bought there, in which case, we would of made a huge lose because it went down big time.
I don't think a post ever says what MML line to buy or sell at and that is the end of it. MML's are meant to act as support and resistance areas, but it does not mean automatic buy or sell at each and every MML. You need to look at how price is interacting with the MML.
In your screen shot, I would not have taken the long that did not come down to the MML at the end of the shot. But, it was clear it was struggling to move higher and then looks like another opportunity to short once it failed to do so. Plus you need to look at the overall trades not one individual trade, if you look at the three prior MML's they were all nice shorts.
It doesn't matter where you plot the lines. Price only moves up and down.
It doesn't matter what chart period you use. Price is the same on all chart periods.
Do you really believe that just because you give a line a name like ULTIMATE SUPPORT that it actually effects what price is going to do?
Don't you realize with all the traders in the world that just about every price on the chart has a "name" associated with it? Think about all the traders, chart periods, and indicators.
DO NOT BE FOOLED or else the rat will beat you.