What if

What if?

http://www.hquotes.com/tradehard/simulator.html



It’s not exactly a joke. You can have a draft opinion for your lovely system when you have to leave the ship :wink:



Eu



Thanks for this web site Eu.

You’re welcome. I can very easy to observe Tango in terms of the simulation. It’s not big deal for me. I just like to hear some responses before I’ll start analyzing my system on the simulation :wink:

Eu

It’s a funny tool, but it doesn’t capture the full distribution and ordering of the profits and losses. Don’t use it for anything serious.

i.e. you can have 2 systems with very different Sharpe (or Sortino) ratios that would have the same inputs in this tool and hence would produce the same equity curves.



The tool is actually helpful in understanding that using the profit factor all by itself is of limited use in picking a trading system.

It’s a funny tool

Of course it’s a toy. Let’s put “Monaco” in the baby’s sandbox. :wink:

Win/loss: 1.41

Win prob.: 0.636



Wow! You already have all money of the world lol Unfortunately, “Monaco” is statistically unrepresentable ;(



However, any toy has some meaning :wink:



Eu

Unfortunately, it IS a toy. More precisely, it’s a perfect illustration of the GIGO principle (“garbage in, garbage out”).



What’s missing is two absolutely essential elements, to make this useful.



I. A crucial 3rd variable, ideally to be user-defined: % of account to risk on each of the 453 trades per simulation path. It is 1) undisclosed and not even mentioned… despite the author obviously being aware of Kelly optimal f; 2) extremely low, sometimes << 1%; 3) perhaps worst of all, not constant.



II. An indication of how compounding, as well as position sizing, is handled after each trade.



This could’ve been actually useful to show to clients. Good idea, terrible execution. What a shame.

2M West Have you ever mentioned amount of C2’s system that slip on the toy? :wink:

I would agree it’s half of a joke, but some systems cannot overcome the joke :frowning:



It’s sad story. :frowning:



Eu



P.S. I don’t think that you’re in position for speaking about GIGO principle.

Huh?.. Does one need to possess special qualifications to be “in a position” to bring up the widely-known GIGO principle? Your statement doesn’t make any sense.

Huh?.. Does one need to possess special qualifications to be “in a position” to bring up the widely-known GIGO principle?

lol Not of course. Nothing personal only my piglish. Translation is "We’re speaking from different positions where nobody agrees with an opponent"

Your point of view as I understand it.

I. A crucial 3rd variable…

It doesn’t make any sense, because we’re speaking about draft opinion that can be made on the toy. Of course you have to play with Monte Carlo for better opinion.

II. An indication of how compounding

Simplified compounding. Kelly is calculated after all simulations.



My point. It’s a toy. However:

- Correlation of Win/Loss and Win Prob is simplified version of an edge of a system,

- Random feed of data is ideal check for having ideal equity curve for a system.



Simply, if a system doesn’t correlate with the toy on statistically significant data there isn’t any system :wink:



Eu