Funny - that topic came up in the conversation. To some extent I agree with what Mr Hare claims.
Get in on a coin toss & manage it.
Now - we didn't discuss tossing but we did discuss the way your focus entirely changes once you are in a trade.
In short "I wish I could read the action when I'm not in a trade as well as I can when I'm in a trade"
There is something about being in a trade that gives you a very accurate feel for whether it will work out or not. Now, whether you act on that is a different matter. I just can't get the same feel for the market without having any money on.
Yeah that makes sense, I'd say the same thing applies with an algo to a certain extent.
By that I mean unrealised profit / loss provides an easier framework to base trade management programming on.
Its a solid reference point.
Trying it to base trade management and exit off price alone is harder to program
(and makes no sense anyway), thats also why the entry has a random element.
I don't really think for a minute that anyone who trades randomly allows every
parameter to be truly random.
Time is a prime example, a random entry at 0430 GMT for instance doesn't make much sense.
My entries are really pseudo random, some parameters are totally devoid of randomness.
As for ninja, been a while since I looked at the 3rd party add ons...grown a fair bit:
NinjaTrader stock, futures and forex charting software and online trading platform. NinjaTrader Partners.
To say that the entry is not important is pure bull, imo. Proper entry reduces risk.
It's once you've entered that the exit matters. Good entry = stay the course,
bad entry = bail asap. But a good entry will always provide a better opportunity to minimize loss, and maximize profits.
Completely agree, as mike said, entry, trade management and exit are all equally
important in short term trades or scalping.
Random entries can still work, obviously they will not be as efficient and are more suited to longer trade duration.
That also means the focus is firmly on managing the trade.
Thats the real purpose of the coin flip experiment in general,
and why its mentioned in that prop trader exercise list.
A coin flip entry is basically a trade management exercise.
It can also be the basis in part for an algo for the reasons I mentioned above.
When you consider typical strike rates of 40-60% (i.e. coinflip territory) you have
to wonder if that is any better than random anyway...(hardcore tape scalpers aside before anyone mentions it
![Stick out tongue :p :p](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)
)
The real reason for discretionary trade being more efficient is not strike rate,
its trade management.