FORUMS BROKERS SOFTWARE
Home
 
    Forums > Trading for a Living > Trade Management > Rule of thumb for position sizing


Reply
 
Thread Tools
Old Dec 16th, 2011, 06:12 AM   #31
Daal
 
 
Join Date: Oct 2002
Location: Brazil
Posts: 9,192
Quote:
Quote from Ghost of Cutten:

No they don't - what makes you think Kelly or Optimal F are the only valid ways to calculate risk? Kelly is notorious for causing blowups and massively underestimating drawdown risk. In any case, both my rule of thumb and Kelly pay attention to win rate, and I explained why I ignored the payout ratio (because in a bad drawdown of consecutive losers, you get NO payout because you have no winners during the losing streak).

Your 2nd claim is silly because I explained the reasoning in this thread - they are based on likely maximum drawdowns, which is the main measure of risk, is it not? If your fund is down 35%, no investor or creditor will care if you were using Kelly or Optimal F for your position sizing - but they will care that you just vapourised over 1/3 of their capital.

Ironically your own post, complaining of a lack of reasoning, has no reasoning behind it - it's pure assertion!
I trade to make as much money as possible with safety, not to protect myself from investors that don't exist
    Quote
Old Dec 16th, 2011, 06:37 AM   #32
Ghost of Cutten
 
 
Join Date: Aug 2009
Posts: 2,869
Quote:
Quote from Daal:

I trade to make as much money as possible with safety, not to protect myself from investors that don't exist
And for safety you need to keep drawdowns within acceptable limits. I.e. the whole subject I was discussing and basing my rules of thumb on.
    Quote
Old Dec 16th, 2011, 06:45 AM   #33
Daal
 
 
Join Date: Oct 2002
Location: Brazil
Posts: 9,192
Quote:
Quote from Ghost of Cutten:

And for safety you need to keep drawdowns within acceptable limits. I.e. the whole subject I was discussing and basing my rules of thumb on.
I'm willing to tolerate a bigger drawdown for excellent trades that have great payouts(3-1, 4-1) but might not be for others without(1-1) even though they both have the same win rate. So I would have to incorporate that factor in
    Quote
Old Dec 16th, 2011, 06:46 AM   #34
Ghost of Cutten
 
 
Join Date: Aug 2009
Posts: 2,869
The fact is that a DD in excess of your risk tolerance is a hard limit. Whereas being somewhat less profitable than you hoped is not. You will not drop out of the game or go bust because you only made 15% instead of 25%. You will drop out or go bust because you dropped 50% on some aggressively sized leverage bet, or a particularly long losing streak, instead of the 20% you had bargained for as a worst case.

Do some simulated trading runs with sizing based on Kelly or Optimal F, you will see that the recommended sizes result in catastrophic drawdowns eventually. Big losing streaks are out there. Furthermore, neither formula accounts for system degradation, or (for discretionary traders) misjudgement of the trade odds. So, in the real world where systems degrade, and discretion is sometimes totally wrong, even a sensibly sized bet which assumes accurate assessment of the trade odds, is too big. You have to trade even smaller than what the odds would suggest is a conservative size. And this is before we even account for the psychological impact of serious drawdowns, which are demonstrable and very real for all traders (albeit to varying degrees).

It's no coincidence that traders with experience, who've witnessed several market cycles, trade smaller than those with less time in the trenches. Similarly for those who have looked into the odds of losing streaks, who've studied the blowups of professional traders, who've debated with systems traders about the degradation of returns. Very few traders overestimate risk, almost all of them underestimate it.
    Quote
Old Dec 16th, 2011, 06:59 AM   #35
Daal
 
 
Join Date: Oct 2002
Location: Brazil
Posts: 9,192
Quote:
Quote from Ghost of Cutten:

The fact is that a DD in excess of your risk tolerance is a hard limit. Whereas being somewhat less profitable than you hoped is not. You will not drop out of the game or go bust because you only made 15% instead of 25%. You will drop out or go bust because you dropped 50% on some aggressively sized leverage bet, or a particularly long losing streak, instead of the 20% you had bargained for as a worst case.

Do some simulated trading runs with sizing based on Kelly or Optimal F, you will see that the recommended sizes result in catastrophic drawdowns eventually. Big losing streaks are out there. Furthermore, neither formula accounts for system degradation, or (for discretionary traders) misjudgement of the trade odds. So, in the real world where systems degrade, and discretion is sometimes totally wrong, even a sensibly sized bet which assumes accurate assessment of the trade odds, is too big. You have to trade even smaller than what the odds would suggest is a conservative size. And this is before we even account for the psychological impact of serious drawdowns, which are demonstrable and very real for all traders (albeit to varying degrees).

It's no coincidence that traders with experience, who've witnessed several market cycles, trade smaller than those with less time in the trenches. Similarly for those who have looked into the odds of losing streaks, who've studied the blowups of professional traders, who've debated with systems traders about the degradation of returns. Very few traders overestimate risk, almost all of them underestimate it.
There is no disagreement here, those formulas are not adequate due the ludic fallacy. They are just the baseline. I'd note that the large drawdowns they produce are not relevant in terms of expectation, people like to say 'if you lose 50% you must make 100% to get back to even' but it doesn't matter because those formulas will get the growth of the capital to be the fastest as possible so its pretty like you WILL get back to even, even though its a big return

I simple refuse to have the same drawdown tolerance in a 30-1 bet(say HKD revaluation) as in a 1-1 bet(say shorting US stocks with stops and profit taking exits). I don't think its mathematically sound
    Quote
Old Dec 16th, 2011, 07:50 AM   #36
Ghost of Cutten
 
 
Join Date: Aug 2009
Posts: 2,869
Quote:
Quote from Daal:

I'm willing to tolerate a bigger drawdown for excellent trades that have great payouts(3-1, 4-1) but might not be for others without(1-1) even though they both have the same win rate. So I would have to incorporate that factor in
I incorporated that by having size bounds (e.g. 0.5-1% for low win rates, or 2-4% for high win rate strategies). Doubling size from 0.5% to 1%, or from 2% to 4%, will double your drawdown, so that is a pretty healthy level of leeway on bet size.
    Quote
 
Reply
Thread Tools

Forum Jump



   Conduct Rules   Privacy Policy   Sitemap Copyright © 2014, Elite Trader. All rights reserved.   

WHILE YOU'RE HERE, TAKE A MINUTE TO VISIT SOME OF OUR SPONSORS:
Advantage Futures
Futures Trading & Clearing
AMP Global Clearing
Futures and FX Trading
Collective2
Automated Trading Services
CTS
Futures Trading Software
dom993trading.com
NinjaTrader Consulting
eSignal
Trading Software Provider
FXCM
Forex Trading Services
Global Futures
Futures, Options & FX Trading
Interactive Brokers
Pro Gateway to World Markets
JC Trading Group
Direct Access Trading
MB Trading
Direct Access Trading
NinjaTrader
Trading Software Provider
optionshouse
Option Trading & Education
Rithmic
Futures Trade Execution Platform
SpeedTrader
Direct Access Trading
SpreadProfessor
Spread Trading Instruction
thenut-trader.com
FX, Gold, & Stock Signals
TradersStudio
System Building & Backtesting
Tradier
Equity and Options Trading
Trading Technologies
Trading Software Provider