Questions: managing other people's money?

[nitro]

Originally posted by praetorian2
Who cares, how much he charges, if you deduct all fees, he's in the top 5% of performance in the world, and that's after a 50% haircut. I wouldn't complain if I were in his fund. Would you? As per Buffett, while he did charge 25%, he didn't charge anything for the first 6%. So you were guaranteed the risk free rate of return (10 yr treasury-inflation) for free. If you then calculate it out, since the first bit is free, it isn't until you get up around 50% return that he even makes 20%.

P2,

THEORETICALLY, there is nothing "wrong" with it. But my business sense has always told me that if you start from a place to high, there is no place to go but down, with no place to negotiate or for much error.

Inspite of all that, the guy seems to have trader like returns with HUGE sums of money - this guy must rank as one of the best of all time...

nitro

 

[Aaron]

I can tell you from experience that the management fees are dwarfed by the performance fees if you can make money. The Schindler Fund charges 2% of assets and 20% of profits (with a highwater mark). On a minimal $20k account, the management fees have been about $500 over the past year (with the growth in the account the 1/12 of the fee charged every month went up). With a 50.1% net return over the last 12 months, that same account generated about $2760 in performance fees.

And lest you think hedge fund managers are overpaid...The investor made $10,020 dollars and their account is now over $30k -- even after all fees. Had they invested in the S&P500 their $20k would be down to $16,840. So we are definitely adding value -- or at least Schindler Trading is.

And, of course, past performance is not necessarily indicative of futures returns and it is tough for me to plan my family budget for next year because of it! Let me also add that Schindler Trading takes on a heck of a lot more risk than your average equity mutual fund, so absolute returns aren't a fair comparison.

 

[nitro]

Originally posted by Aaron
I can tell you from experience that the management fees are dwarfed by the performance fees if you can make money. The Schindler Fund charges 2% of assets and 20% of profits (with a highwater mark). On a minimal $20k account, the management fees have been about $500 over the past year (with the growth in the account the 1/12 of the fee charged every month went up). With a 50.1% net return over the last 12 months, that same account generated about $2760 in performance fees.

And lest you think hedge fund managers are overpaid...The investor made $10,020 dollars and their account is now over $30k -- even after all fees. Had they invested in the S&P500 their $20k would be down to $16,840. So we are definitely adding value -- or at least Schindler Trading is.

And, of course, past performance is not necessarily indicative of futures returns and it is tough for me to plan my family budget for next year because of it!

Aaron,

I found you up there too!

Geesus, +234.33% return in 2000? Is that a typo? But _ONLY_ +83.30% in '01 and +32.94% '02 - what's happening dude, you slacking off?

nitro

 

[Aaron]

Originally posted by nitro

Geesus, +234.33% return in 2000? Is that a typo? But _ONLY_ +83.30% in '01 and +32.94% '02 - what's happening dude, you slacking off?

Every trade, every monthly return, and every yearly return can be modeled as independent draws from distributions of possible outcomes. The distributions have positive means or "expectations", but they have sizable tails falling off in both the negative and positive directions. Because we use anti-martingale position sizing (position size proportional to account size) the monthly and yearly curves are positively skewed -- so big years can happen.

Although I must say we were playing somewhat larger relative position sizes in 2000, so the chance of repeating such a good year again are remote. However, we did return 26.1% in April of this year and 31.9% in July and it only takes a few months like that to make a big year. So let's hope for the best.

The 32.9% for 2002 was through September. As of the end of October we are up to 47.0% for the year.

But when throwing around these big return numbers, keep in mind the risk level. We have been down 30% in a couple months. Put a couple months like this back to back and that's a 50% drawdown. Ouch! It could happen and I try to make sure all the investors know it.

 

[darkhorse]

hey Aaron, what was it like working for Trout?

any insights on how a guy manages to swing trade with billions?

 

[Aaron]

Originally posted by darkhorse
hey Aaron, what was it like working for Trout?

any insights on how a guy manages to swing trade with billions?

Very educational. I've seen other posts on Elite Trader where new traders are recommended to work for a hedge fund or money management firm to learn the ropes -- it's good advice.

The most important thing I learned at Trout Trading was how to methodically and rigorously backtest and implement strategies. It takes a lot of analytical skills and is time consuming, but I couldn't imagine trusting a strategy without it.

 

[darkhorse]

Originally posted by Aaron


The 32.9% for 2002 was through September. As of the end of October we are up to 47.0% for the year.

But when throwing around these big return numbers, keep in mind the risk level. We have been down 30% in a couple months. Put a couple months like this back to back and that's a 50% drawdown. Ouch! It could happen and I try to make sure all the investors know it.

One other question

If you are willing to incur a 50% drawdown, would you consider 47% gains subpar this far into the year (as that's a risk adjusted return of less than 2:1 based on 30% dd)?

i.e. what is your desired (expected) risk adjusted return for this system over the long run?

2:1, less, more?

 

[Aaron]

Originally posted by darkhorse


One other question

If you are willing to incur a 50% drawdown, would you consider 47% gains subpar this far into the year (as that's a risk adjusted return of less than 2:1 based on 30% dd)?

i.e. what is your desired (expected) risk adjusted return for this system over the long run?

2:1, less, more?

Every annual return can be modeled as a draw from a distribution that has a positive expectancy and long negative and positive tails. Even if all goes well and the strategies perform as expected, eventually we're going to have a losing year -- just because trading returns are stochastic. Our backtesting would indicate that in the long run we can expect about one out of every seven years to be negative. We haven't had a losing year yet and I'm thankful for it.

You are very wise to look at the Sharpe ratio (i.e. risk adjusted returns) rather than absolute returns. But I suggest you would be even more wise to look at the Treynor ratio rather than the Sharpe ratio. The Treynor ratio is a measure of the return per unit of risk also, but uses systematic risk rather than total risk. With a negligible 0.025 correlation to the S&P500, a small (5-10%)investment in the Schindler Fund can actually reduce the overall volatility of an equity portfolio while improving returns.

Sharpe's ratio is (return - risk free rate) / volatility. Treynor's ratio is (return - risk free rate) / beta. Sharpe's ratio is a measure of reward to risk and Treynor's ratio is a measure of reward to the additional risk an investment adds to a portfolio. With a beta of close to zero, our Treynor ratio is huge and that's a rare and beautiful thing.

 

[dottom]

Traders looking at performance of specific trading method should look at Sharpe. Investors looking at portfolio selection should look at Treynor.

In other words, if you care about all risk (not just systematic risk), and you are comparing one comprehensive strategy to another, Sharpe's measure is the appropriate one.

However, say you are comparing investments that are not comprehensive. That is, you are looking to invest some of your money in them, but you would (naturally) use other investments to eliminate any fund-specific risk. In this case, you are only interested in the systematic risk of those investments (since the unsystematic risk will be diversified away). In this case, the Treynor's measure would be more appropriate

 

[nitro]

Originally posted by dottom
Traders looking at performance of specific trading method should look at Sharpe. Investors looking at portfolio selection should look at Treynor.

In other words, if you care about all risk (not just systematic risk), and you are comparing one comprehensive strategy to another, Sharpe's measure is the appropriate one.

However, say you are comparing investments that are not comprehensive. That is, you are looking to invest some of your money in them, but you would (naturally) use other investments to eliminate any fund-specific risk. In this case, you are only interested in the systematic risk of those investments (since the unsystematic risk will be diversified away). In this case, the Treynor's measure would be more appropriate

Hmmm,

Interesting - thx dottom.

nitro