Newbie, Interesting stuff. There's an old trading adage "You're worst drawdown is yet to come." Good thing to remember. I'll give Jones credit for not sugar-coating how bad a drawdown feels, plus he has a pretty good method for scaling back when you're in one. However, you can't make any sort of plan based on 3 or 4 months of trades. You need years of data.
While I like Ryan Jones' work on money management (still wouldn't buy a system from him though, lol) I prefer a simpler method. While my disdain for the vendoring business is well known here, Larry Williams (who many people dislike for his unabashed marketing efforts) is one of the very few that I learned something from that was worth more than I paid for it. His money management formula is very simple. Its based on the concept that the major downfall of the various formulas like fixed fractional or fixed ratio is the effect of the system's largest losers. They can decimate an account when they occur at your maximum position size. So, he incorporates the largest loss from his backtesting (although he admits the worst is always yet to come, so you can also throw in extra cushion if you want) into his formula. The formula is: # of contracts or shares to trade = (account balance * % of account you are willing to risk)/largest system loss per share or contract The appeal of it is that it is very simple and makes sense to me (an important requirement for a non-mathematician like myself). Kirk
I like that formula..... my only concern is.... If I'm trading with $100,000 and my biggest lost is $5,000 and I'm not willing to risk more than 5% on any one trade.... the most I can trade is one contract. I will need to wiat until the account is double at $200,000 before I can increase to two contracts correct? Am I understanding that formula correctly or am I lost? I want to investigate this method....... does Williams have any books for sale on this method? Does he have a website? Thanks for the help.
newbie, You are correct in your example. But Williams risks more (he says he uses 10-15% in his own trading) on the premise that it is unlikely that you will actually lose the amount of the largest loser on each of a cluster of trades to knock you out completely. At 10%, you would need to lose the largest historical amount on 10 consecutive trades (or a cluster of them with smaller winners in between). Not just 10 losers in a row. The odds are probably against that if you have properly analyzed your system over an acceptable timeframe (although not completely impossible). Its a risk he is willing to take. I prefer a lower percentage, and thus have to trade fewer contracts. I would say, that if you have a max loss of $5000 in a system (on only one contract) that you should not trade more than 1 or 2 contracts per $100,000. If you had 5 contracts on when a loss like that happened, 25% of your account is gone in one trade. Most of Williams' published systems (I'm not endorsing them) have max losses of $3000 or less trading the full S&P contract...that's only $600 on the mini. With a system like that, you could trade a higher number of contracts. There are two books (at least) that have this formula. "Long Term Secrets of Short Term Trading" and "Day Trade Futures Online". In fact, they are virtually the same book (one of my gripes with Williams). The second one is cheaper and usually on the shelf at Barnes and Noble. The part with this formula is only one chapter, but like I said, its simple. Spend your money on a latte instead and read it while your at the store. Kirk
One way to reduce portfolio risk is to diversify into several markets (futures), or stocks at the same time, rather than all your eggs in one basket. This obviously only works on a longer time frame (probably not scalping), but it allows you to control much more size with less risk. I'm up to holding 3,000 (ave.) shares now at one time, a lot of times overnight. But I do this with 10 stocks, not 1. I also cover many industries with those ten. If I have one gap down for a 20% loss, its really only a 2% portfolio loss. The 10 stocks also tend to offset each other when the market is wandering aimlessly, allowing me to hold for a big move. But the funny thing is when the market rallies, all my stocks tend to go with the rally, so its the best of both worlds. My losing days are when the whole market turns against most of my positions simultaneously, but I have stops to halt the bleeding. You might consider trading mini-contracts in several markets instead of 1 full contract in one market as one way to smooth the volatility of your portfolio. Especially if your using all your money on 1 trade.
This is awsome! In other words, this guy says, "show me a winning system, and I will show you how you would make 10 times as much using my money management system." Hmmm, I like that. But the problem is the same as always. First, we need a winning system. Then, we need the sytem's losing period to fall under a certain category, so you don't blow your account. The problems with using math to define risk managemet is that it relies on expectancy. It does not take into account volatility and liquidity. And it certainly discounts disasters. Example: If your system tells you to be long x number of contracts and a major event takes place. You can say good bye to your money! What did you do wrong? You kept increasing your bets and ... shit happens. Oh shoot, did I forget to mention that stops are no good on limit down (or up) days? And if you have three of them in a row ... Can you spell C-H-A-P-T-E-R E-L-E-V-E-N? Bill
OK. But it's not realistic. With the capital markets behaviour you should assume, that you lose 1 every time heads landed down, and win 1 ( not 2 ) every time heads landed up. And then: let's risk 25% of our capital on each flip. After 100 flips, 50 ups and 50 downs ( zero-sum game ) we'll turn a 100 into 3,96. Why? because of compounding. When you lose 10% and then win 10% you won't get your 100 back. Example: 100 x 0,9 = 90 90 x 1,1 = only 99. Even if you have 56 wins and 44 losses in 100 flips, with 25% of capital risked in every flip, the result will be -15 pts from 100 at the start. Let's try another "money management" strategy. If the price goes up by some x%, we open long position worth x dollars. If it goes up by another x% we open another long position worth the same x dollars. And so on. Than: If the price falls by x% from the last time we open long, we close all long positions, and open 1 ( worth the same x dollars ) short position. We'll win where there are long strikes of x% price moves, and we lose on "choppy" market, when there are mixed x% moves ( upside, downside, no direction ). What will be the number of trades closed after one x% move (in our direction ), after 2 x% moves in our dir, after 3 x% moves...? Remember, we close all positions after single x% move in wrong direction. If the market prices change in random ( or very close to random )way the results will be: 50% of trades will be closed after 1 x% move in our dir 25% of trades will be closed after 2 x% moves.... 12,5% of trades .... after 3 x% moves 6,25% of trades .... after 4 x% moves ...and so on We lose [ 1x (x% move) ] when we close after 1 x% move in our dir. For example: prices go: 100.00 , 100.25 ( open long ) 100.00 ( move in wrong dir, we close ). result : -0.25 We lose [ 1x (x% move) ] when we close after 2x% moves in our dir. For example: 100.00 , 100.25 ( open long ) , 100.50 ( open 2nd long ) , 100.25 ( move in wrong dir, close ). result: 100.25+100.50 - (100.25x2) = -0.25 We go flat when we close after 3 x% moves. We earn [ 2x (x% move ) ] when we close after 4 x% moves. Further calculations: check it in excel. The result will be ZERO. I think the key things in trading are entry end exit points. When you should enter and exit - it should be based on individual intuition. The market behaviour is very complicated, and it's impossible to describe it with few simple parameters. It can only be done by human brain. What do you think about it? DT-waw
Same with "cut your losses, let your profits grow". If your losses can be max 1 point, and you take every profit at 3 points, you will likely to have 3 times more losing than winning trades. Very important issue is: does stock/futures/options/currency prices move in a 100% random way or not? And if yes, is it possible to win? Make a 1000 or more random prices chain in excel, create a chart, and you'll see that they will look like stock or futures charts... You'll see TA patterns, you can draw on them support and resistance lines... DT-waw
WarEagle ," Larry Williams (who many people dislike for his unabashed marketing efforts) is one of the very few that I learned something from that was worth more than I paid for it. " I've noticed that he gets bashed a lot. I think its just the fashionable thing to do. I like him too. He's a funny guy.