Kelly Criterion & Risk Of Ruin As Risk Management Tool

[nonlinear5]

So, Kelly does not minimize "ruin" and is good for someone with deep pocket. In general, should one "bet" using fraction Kelly, like 1/2 Kelly 1/4 Kelly?
Regards,

The depth of the pockets is not part of the Kelly equation. If you bet twice Kelly (even with a profitable system), you are mathematically guaranteed a ruin, no matter how much money you have initially. That's because the Kelly bet is always proportional to the size of the account. Quarter Kelly is a reasonable leverage.

Can one numerically, i.e., using Monte Carlo to simulate risk of "ruin" using various fractional Kelly?

Yes, you can. I believe these figures are invariant to the system. For example, with full Kelly, you have a known 20% probability of losing 80% of the account.

 

[murray t turtle]

Anyone uses Kelly Criterion?

Yes - I use my version of it modified for trading equities & ETFs as opposed to gambling which it was intended for. Max position size is 20% of capital (no margin loans!). I break each position into 4 separate trades of 5% of capital - layering in - adding only to winners. This keeps my loses much smaller than my winners and still allows for the equity curve to have stellar winning years and small draw downs on losing ones (<7%). My risk per trade is just under 1% of capital.

%%
In other words; less is more.
I like the martingale dog collar. Meaning what?? As an example ,its a dog collar that gives the dog or profit a bit more slack, so as not to choke the dog or profit.

As far as math , i like IBD[ Investors Business Daily] founder math; risk $3 to make $8, or more. But that is in context of a cash market+ 250-450 page system.

 

[ironchef]

Thank you all. This is really eye opening for me.

It is actually counter intuitive to me: For the longest time, I had always been "all-in", i.e., 100% of my investable assets were in equity, zero diversification, wrongly belief that to maximize profit or outcome I need to go all in. As a result my net has always been extremely lumpy, huge ups in bull markets but also big draw downs in bear markets like 2002-3 and 2008-9. Finally, this year I started to think about risk management and came across Kelly.

I now appreciate why the experience traders at ET all telling the new kids to not risk more than 1-2% on each trade. The "risk of ruin" is very real.

 

[ironchef]

Another couple of questions for you:

1. I keep a record of all the outcome of my options trading, can I then use the record to do a Kelly sizing simulation (I can do that using Excel) and find the best Kelly size for my system? And can the Kelly size tell me my system's edge, i.e., if it is .2 it is better than a .1 system?

2. Can one determine trading edges by calculating the Kelly of one's system when backtesting? If so, what should one look for in a Kelly for developing a system?

Thanks.

 

[ironchef]

The depth of the pockets is not part of the Kelly equation. If you bet twice Kelly (even with a profitable system), you are mathematically guaranteed a ruin, no matter how much money you have initially. That's because the Kelly bet is always proportional to the size of the account. Quarter Kelly is a reasonable leverage.



Yes, you can. I believe these figures are invariant to the system. For example, with full Kelly, you have a known 20% probability of losing 80% of the account.

Thank you again sir.

 

[ironchef]

The depth of the pockets is not part of the Kelly equation. If you bet twice Kelly (even with a profitable system), you are mathematically guaranteed a ruin, no matter how much money you have initially.

But many great investors, like Soros, Druckenmiller, and the fellows in the book "Big Short" advocated and went "all in" on their trades? Survivor bias?

 

[Martinghoul]

But many great investors, like Soros, Druckenmiller, and the fellows in the book "Big Short" advocated and went "all in" on their trades? Survivor bias?

Yes, as well as other things...

 

[HobbyTrading]

I now appreciate why the experience traders at ET all telling the new kids to not risk more than 1-2% on each trade. The "risk of ruin" is very real.

Although well-intended, I believe that this advice is even half-correct. You don't want to base an investment size on its cost price but on the volatility of that cost price. You want to size the volatility on the size of your account. So a better way of saying it, at least in my view, is that a position size should be such that its value volatility is 1%~2% of the account value.

 

[MrScalper]

Another couple of questions for you:

1. I keep a record of all the outcome of my options trading, can I then use the record to do a Kelly sizing simulation (I can do that using Excel) and find the best Kelly size for my system? And can the Kelly size tell me my system's edge, i.e., if it is .2 it is better than a .1 system?

2. Can one determine trading edges by calculating the Kelly of one's system when backtesting? If so, what should one look for in a Kelly for developing a system?

Thanks.

Total waste of time and effort!

Your risk should be determined by your market of choice..your strategy..your investment horizon..and last but not least..by your available capital!

Kelly..and such..are of no real value to the small retail trader. I will be very surprised if you get anywhere with this..but I am often wrong

 

[nonlinear5]

I keep a record of all the outcome of my options trading, can I then use the record to do a Kelly sizing simulation (I can do that using Excel) and find the best Kelly size for my system?

Yes.

And can the Kelly size tell me my system's edge

No, it works the other way around. Your system edge (as determined by the distribution of trade returns) is the input into the Kelly equation, and Kelly fraction is the output.