“I think that there is too much emphasis placed on max drawdown without considering the recovery time. To me, it is more important how long it takes to recover from the drawdown. If you invested in an S&P 500 index fund in 2000, you still won’t have recovered from the 40% losses of 2001 & 2002 even 5 years later. With the second Sim, in less about 5 months, you are back above the previous high that occurred before the max drawdown. This is a risk/reward game, and in this case, at least in my eyes, the second Sim has a better risk/reward ratio.”
Idealistically, I agree with you completely. I think a person can analyze risk/reward more accurately by looking at Sharpe Ratio, Standard Deviation, Alpha, etc. However, max drawdown is very real, and every trader has to be honest with himself on how a drawdown is going to affect him. 25% drawdowns are very common in our models. But I can honestly say that if my portfolio was up to 100k, and then suddenly dropped down to 80k, I would be affected mentally, even though I knew this was normal for the model.
These days I am looking for models with a max drawdown less than 25%, annual returns greater than 100%, and standard deviation less than 45%. I split my money between 4 models that are not highly correlated. In theory my current mix should give me overall max drawdowns in the range of 10-14%. Recent history has shown me that this is about how much I can handle before I start reaching for the Maalox. If your stomach is stronger than mine, then you should reach $1,000,000 before me. :~)
Let me end by saying your analysis of gain/stock/day is very good. I’m going to start including this measure in my analyses. Thanks.
I definitely understand the ratio of risk to Maalox!
I lived through the losses of the 1973-74 oil embargo, the 1987 September market collapse, And the 2000-02 Tech crater. Each time I lost nearly 1/2 of my equity investments. Those were hard times to recover from.
However, the compounding effect of annual gains needs to have a higher priority than drawdown. None of the standard risk/reward ratios address the value of taking the higher risk in order to achieve the compounding effect of the higher rewards. I would rather invest in a 40 % annual gain Sim that had a 30% drawdown every year, than a 30% annual gain Sim with a 20% yearly drawdown.
The difference is that after only 7 years the 40% Portfolio will increase a factor of 10, and the 30% Portfolio will increase a factor of 6.3. Starting with $100,000, that’s your million dollars in 7 years compared to ‘only’ $630,000. A difference of $370,000 or 50% more in only 7 years, or an additional gain of 3.7 times your initial investment, is worth a lot of bottles of Maalox!
How many years of your salary at 2000 hours of work/year does it take to make $370,000? Just imagine what 20 years compounding would do. As long as you can develope a faith in the model (which I admit is hard to do), a little more work at achieving a higher annual gain, compounded, is worth striving for and accepting the higher risk.
While risk and volatility in gerla is bad, long or longer term performance of a strategy is a very important thing to me.
I do not care if a system had a large drawdown in the summer of 2002 if it recovered quickly out of it and by the end of the year or in 2003 made riple the money a lame mutual fund did that did not experience that big of a drawdown.
Higher perceived risk will always be associated with a better performing strategy. That’s is the name of the game. You accept more risk in trade of better expected returns.
There will always be the personal balance of preference to less risk or more rewards. And it is a fine balance. It is without a question though that if you shy away from risk, especially what is indicated as risk by different mathematical indicators you will be staying away from opportunities.
Long terms returns and quick recoveries from drawdowns is what ultimately should be the best characteristic of a winning strategy. Everything else is the investors personal tollerance to temporary (or not that temporary) drawdowns.
I wonder wether a formula like [anual return]^N/[standard deviation], where ^ is the symbol for power, N is a number greater than 1.0, let’s say 1.5 or 2.0 will give a better result that the Sharpe (N=1). I am thinking such a formula will give a better preference to strategies that had their high Sharpe because of high returns not because of low volatiltiy.
I should run an experiement like that to see what sims will come up in the list.
Good points. But since one of my criteria is annual return > 100%, it’s not like I am shying away from risk. I simply want to have my cake and eat it, too! If I end up living 15 additional years due to lower stress levels, it’s worth waiting an additional 3 years to achive total financial independence. :~)
To me > 100% anual is a bit extreme. Though, as I said, I do not mind volatility for returns.
I am personally satisfied with returns in a lower range. Anything above 30% or let’s say 40% is a good return for me. So from there on if I can find a good return graph, maybe lower risk or maybe even higher return (and not necessarlily >100%), I am OK.
What some may find a bit extreme with me is I do not mind having debts as long as my investments return more than what banks are charging me for my debts.
A lot avdice is coming these days on “no debts” at all including advice on paying off your house before starting to invest.
You see - I look at it as a business - income and expenses. See the recent post from Marco on Robert Kiyosaki’s article. I like Robert Kiyosaki’s thinking on this.
This is why something of the range of 30%-40% and above satisfies me. I know I can find 30%-40% return strategies with less risk than the risk in >100% return strategies. As long as I am well above what banks charge for loans am OK.
By the way, needless to say, I do not advise anybody to consider investing if they have unpaid debt.
I should clarify. The 100+% return is what I am looking for from short-term trading sims. I also invest in high-yield dividend stocks that earn about 20-30% annually (total return). My ultimate goal is to convert short-term capital gains into long-term dividends. At the same time, I am reducing my monthly expenses. When passive income (dividends) exceeds monthly expenses, I will be financially free.
As far as debt goes, I agree with you. There’s good debt and bad debt. Debt used to buy assets is good debt.
The actual dividend yields are typically 10% or more. The total return (yield + price appreciation) has been 20-30%.
I am using a P123 Port right now, and it has been working pretty well so far. The only thing I caution on is, when the port recommends a buy, I always check other sources to confirm the dividend yield. Different sources quote different yields, so I always make sure. I also get the top 3 recommendations, do my research, and pick the one I like best. That seems to work much better than just taking the port’s first recommendation.
I have made the sim available for public use. It’s called “BJS High Yield with ATR Stop”. The sim shows 16.91% annual return, but that does not include the dividends.
I double-checked. “BJS High Yield with ATR Stop” should be available in community simulations. I also double-checked to make sure the ranking system was accessible to the public. Let me know if you are still having problems.
I have another sim called “Simple High Yield” that I am not using in a port any more. It has greater annual gains and lower drawdowns, but it has higher turnover, and I didn’t like the stocks it was picking (lower dividend yield, for one thing). It’s also available in the community simulations.
Searching for BJS High Yield with ATR Stop produces nothing. If I search on your name I get:
Last Updated No of Ports/Sims Type
1 BJS Copy of IReturn09 traderblues 13 11/14/05 24 Simple Ordering
2 BJS Copy of TopPort traderblues 44 10/13/05 3 Simple Ordering
3 BJS Copy of VladsNewGrowth traderblues 56 11/15/05 8 Simple Ordering
4 BJS High Yield traderblues 5 10/15/05 6 Simple Ordering
5 BJS Mo Value traderblues 36 04/16/05 22 Simple Ordering
6 BJS Small Cap GARP traderblues 26 11/22/05 13 Simple Ordering
7 BJS Small Cap GARP DHP traderblues 26 08/12/05 15 Simple Ordering
8 IReturn09 w OTC Reduction traderblues 14 11/23/05 3 Simple Ordering
I’m probably doing something wrong…
but don’t know what??
It sounds like you’re looking in the Ranking area. Go to the Simulations page, and search in Community Simulations. Either enter “BJS High Yield with ATR Stop” in the simulation name field, or enter “BJS High Yield” in the ranking system field.
