If you have other screen names in here, writing styles, typos, and particularly some ways to explain technical words could be a give away. Is the profit factor on this one 95:1?
What's The Best Way To Convert This Edge Into Money?
- Thread starter SplawnDarts
- Start date
If you have other screen names in here, writing styles, typos, and particularly some ways to explain technical words could be a give away. Is the profit factor on this one 95:1?
The issue is not the edge. Other things: people freeze, their greed gets them, they fear kicks them out, ... The list is long, but one get the idea.
When did you last take a class on stats? Maybe your professor was not good. Moreover, stats is to rule things out, not to rule things in. In other words, stats is to disprove/reject theories not to confirm them. It could also lead one to find truths, but it does not find truths.
There are no "truths" in trading. What people do with statistical analysis in regards to trading is to try to figure out how much of their result is just plain luck. Or put another way, how much of that "profit factor" is due strictly to variance.
In trading you cannot simply take 53 trades and call them independent. They are anything but. Almost all his trades are correlated which means he probably has maybe 3 or 4 data points. It's the reason why when they take polls to see who everyone is going to vote for President they don't just take 50 people who all live in hippyville and ask them. Or 50 people who only watch Fox news. They make sure their sample is legitimately random. You need to have men, women, republicans, democrats, independents, gays, war hawks, tree huggers and people who can't spell their own last name. You need to have a sample that actually resembles the voting public. Same with trading. The closest thing you can get to that in trading is a LOT of time. You need to trades across every type of market, different trade durations, different products and different correlations. And that ain't going to happen in a week or a month or even a year. Personally I think you need decades of data to "truly" get an idea. And that obviously is not feasible. So while there is no "correct" answer, I am pretty certain 53 trades is the wrong answer.
1. Your above post is then in agreement with my post on stats.
2. With respect to the text quoted below, there is another way. The models/tools should be tested against the market trend. In such case, the random walk is not only neutral, but is playing against the tools. That is why I have made posts in a journal when I tested them against the market. The typical comment: why are you against the market, etc.
I believe there is value for tools that go against the market beyond the testing: it is easier to rule out the bad, and the good could be used to reduce variance even if it is to make little returns.
Since you are in the business, I wonder if you have ever looked, or known someone who looked, at the problem from the angle described here.
3. There is yet another way to test trading models that should cost a lot less in time. I will explain it in another post.
2. With respect to the text quoted below, there is another way. The models/tools should be tested against the market trend. In such case, the random walk is not only neutral, but is playing against the tools. That is why I have made posts in a journal when I tested them against the market. The typical comment: why are you against the market, etc.
I believe there is value for tools that go against the market beyond the testing: it is easier to rule out the bad, and the good could be used to reduce variance even if it is to make little returns.
Since you are in the business, I wonder if you have ever looked, or known someone who looked, at the problem from the angle described here.
3. There is yet another way to test trading models that should cost a lot less in time. I will explain it in another post.
OP don't listen to the naysayers if you have an edge rock that s**t if its bogus you'll eventually bust out, fair is fair.
How are you earning 200k with free time leftover in the morning to trade? At that income its almost like.. why bother gambling in futures?
How are you earning 200k with free time leftover in the morning to trade? At that income its almost like.. why bother gambling in futures?
+1.
The sample size isn't the number of trades, but the window of time over which those trades happend.
I live in MST as well, and up till 11:00 EST is best times for me to trade as contract reduction of 80-85% is done by 11. If you can't make money up to 11:00EST, you need to reconsider trading at all, as especially in Indexes go through "Lunch Chop" for 3 hours. You can start earlier in GC, CL and currencies you can scalp till Indexes open. I really think you have it made to be able to trade these hours and have a decent job besides. I would not advise any lower than three years of min of triple your gross salary before considering trading on your own. And you really need to build up savings to be able to cover your expenses for three years.
Did you mention what kind of drawdown you back tested over last ten years? And if you break down individual instruments, does losing percentages get smaller more in one than others? What would it take to get losing percentages way down and still have signals, this would be the market you could consider to ave down so that breakeven trades on original entry would become winning trades overall.
This is simple.. Keep doing what you are already doing and trade your own money.. 13k is more than enough to make 2k-4k a week.. Forget about finding more trades just focus on being consistent and the money will come (i promise).. Just add more contracts if you are not satisfied with your profits.. Plain and simple.. Do this for a year straight and then you can think about quitting your job..
Ok lets assume you have an edge. You are asking here how you can exploit that hedge, perhaps what avenues you can use to raise capital or maybe which prop trading group you could join.
However, if you're going to get outside capital, your possible paths are highly dependent on proving how good and sustainable your edge is. You have not nearly provided enough data for anyone to confirm that you even have any sort of edge to begin with, so how can you expect them to judge what your options are?