Question on optimal f* and TWR

[Straddle8]

I've got a question regarding optimal f* and the Terminal Wealth Relative. I understand how to calculate both values but I'm a bit confused on the concept of TWR. Is this the multiplicator by which you multiplicate your:

a) starting equity ?
b) worst case scenario (maximum DD)?

For example, I have a starting equity of $100.000 and I make 10 trades with $-10.000 as worst case scenario. I use the table in Thomas Stridsman his book "Trading Systems That Work" on page 276.

In this table I end up with a TWR of 1,0965. Does this mean I made 1,0965 times my initial equity or 1,0965 times my worst drawdown? The author in the book multiplies this by the equity, but I'm not sure if this is the right method to apply optimal f*.

 

[Adamus]

Hi Straddle8, TWR is (Final Stake after compounding / Starting Stake) for your system.

So the book is right.

Have a look at the first chapter in Mathematics of Money Management.

 

[Straddle8]

Thanks for your answer. I was confusing f* with the '% Equity' to invest in your trades. Now I understand why everyone says this is such a volatile method

 

[Adamus]

Yes, optimal f means high leverage and in theory for the trade history you plug into it, it's perfect, but I think it's wrong to use it in real trading unless you know with absolute certainty what your greatest loss is going to be. The formula assumes that you are going to plug in the greatest loss from your trading history, but in reality, your greatest loss is always at some point in the future, so to use this safely you should make an estimate of how bad you think it might be.

but if you are trading options for instance, and you know that the most you can lose is the option premium that you paid, then this is really excellent.

 

[Straddle8]

Yes, optimal f means high leverage and in theory for the trade history you plug into it, it's perfect, but I think it's wrong to use it in real trading unless you know with absolute certainty what your greatest loss is going to be. The formula assumes that you are going to plug in the greatest loss from your trading history, but in reality, your greatest loss is always at some point in the future, so to use this safely you should make an estimate of how bad you think it might be.

but if you are trading options for instance, and you know that the most you can lose is the option premium that you paid, then this is really excellent.

The author also mentions this in one of the chapters. I thought I could minimize this effect by doubling my biggest loss of the backtest, but the danger of going broke still stays big after doing this.

 

[Adamus]

The problem with options though is the transaction costs are so high.

re biggest loss, I don't think it's so difficult to make a reasonable estimate of something you're comfortable with. technically your biggest loss if you're trading short futures is infinity, so in that case you just give up and go home.

i just looked at the markets and checked out what the biggest gaps were on their charts, and extrapolated from that.

isn't it MMM where he also says you should have a career target and get out after x years - because of the risk of a black swan getting you and wiping you out?

on the other hand, if you find an edge and successfully trade with optimal f for a while, start taking cash out and putting it in gold or whatever.

 

[JDW1411]

Ed Seykota has done some great work in this field.

 

[zishor]

I know I am digging out a very old thread here but I couldn't find anything related so far..
Regarding optimal F and the trade size: I have a system that I have thouroughly back and forward tested. I have calculated Optimal f according to Ralph Vince's approach with the P&Ls of the individual trades. The result is 0,341 for f. The biggest loss was $1,069. Tests were all done with one standard FX lot (100,000). If I now divide biggest loss and f the "increment units" results in $3,134. How would this now actually work? It sounds a bit dangerous to me to trade one more lot for every $3,134 profit.
Would be interesting to see what I am missing here... Thanks in advance for your help. Frank

 

[Shazam75]

you need to divide the 3134 value with your total "stake" - so if your account size is $50000, then divide 50,000 / 3134 = 15.95 units. If it was futures I was trading, that would be 15 futures contracts.

Regards