Tally for March '06

[tsptalk]

By the way, I ordered a bunch of nice TSP Talk stainless steel travel mugs to give away to the winners. They should be in later this week. We'll have to come up with some kind of a plan of attack.

I hope to get more things like shirts/ hats, free month of the newsletter, etc. Ideas welcomed.

Thanks,
Tom

 

[FundSurfer]

Tom, If you do give out award for Tally, I'd suggest it be for longer term results (ie. last 6 or 12 months total). The monthly winner can occur because of luck. Nothing wrong with being lucky, but I'd rather see a reward go to long term success.

Might want to consider the best calender year return to have a prize, since it is a logical goal for anyone.

As to rewards for the Tally, remember that not everyone is out for the best return. Many are more interested in safe growth with less risky movement. You could set a category for return with minimal moves (ie. less than X moves per year).

Just my immediate reaction, for what it is worth.

 

[Dave M]

Rev

So where do we find the numbers for the reverend Mr Shark? He was not listed on your chart, F-S.

Dave

 

[tsptalk]

As to rewards for the Tally, remember that not everyone is out for the best return. Many are more interested in safe growth with less risky movement. You could set a category for return with minimal moves (ie. less than X moves per year).

Excellent idea!

 

[tsptalk]

Re: Rev

So where do we find the numbers for the reverend Mr Shark? He was not listed on your chart, F-S.

Current return for the TSP Timing Newsletter vs other funds:

TSP Timing +5.41%
G +1.07%
F -0.94%
C +2.88%
S +3.80%
I +7.34%
20% each +2.83%

 

[SystemTrader]

While it's more complicated, another nice measure for the risk-conscious would be to divide the return by the maximum drawdown (MDD).

MDD is the largest peak-to-valley drop in a given time period. For example, the S&P 500 had an MDD of almost 50% from 2000 to 2002...from the high of 1500+ in March '00 to the low of 768 in October '02.

When you divide the return over the MDD you have a very good indication of how much return you're squeezing out for your risk (drawdown). Professional money managers and traders watch MDD like a hawk.

Awhile back, I added a couple of columns to calculate the MDD for one of Tom's old spreadsheets I downloaded from here. (You need daily valuations to calculate the MDD, so the current format won't work.)

I may try to dig up that file. However, I'm afraid it's on my old PC which is now resting in peace...

ST

Tom, If you do give out award for Tally, I'd suggest it be for longer term results (ie. last 6 or 12 months total). The monthly winner can occur because of luck. Nothing wrong with being lucky, but I'd rather see a reward go to long term success.

Might want to consider the best calender year return to have a prize, since it is a logical goal for anyone.

As to rewards for the Tally, remember that not everyone is out for the best return. Many are more interested in safe growth with less risky movement. You could set a category for return with minimal moves (ie. less than X moves per year).

Just my immediate reaction, for what it is worth.

 

[Show-me]

I think a monthly award would be nice. Why should good luck not be rewarded? A good return is a good return if one is luck you will see it in the tally. You really can't eliminate some element of luck from competitions. Also a shorter term will keep us interested and on our toes..................unless the yearly award is a new car. LOL!!!

 

[tsptalk]

More on the awards - I think Fundsurfer had a good idea to have more than one category. And we can have better prizes for the best longer term returns.

The question is, what would the categories be? mlk_man, rokid, FS and some others probably know better than I what would work. If we had a category for people that make no transfers during the month, we may have the same one or two winners each month as there are few here that do that.

I'll let you come up with something that works. I'll give away 2 or 3 of the new TSP Talk travel mugs per month. You just let me know who won and I'll ship them.

Then I'll look for some nicer items, like shirts etc., for the longer term prizes. I'll make sure that the folks working on this get something for their trouble.

Thanks!
Tom

 

[oldschool]

I like system trader's idea of finding some way to recognize low levels of Maximum Draw Downs...

Reveals how truely risky a trader's performance is over time, no?

 

[mlk_man]

I manage a couple friends 401k's that are limited to one move per month like we use to be so I'd recommend a conservative (<12 moves a year or once a month) versus an aggressive (>12 moves a year) investor.

I don't think a monthly prize is necessary for the conversative investor since we'd probably have a lot of ties. Perhaps a 6 and 12 month prize.

For the aggressive investor, I think monthly, quarterly, and yearly would suffice. I'd even throw in a 6 monther, but that's just me.


M_M

 

[tsptalk]

Good idea mlk - inactive accounts - 6 month awards, or even quarterly. But not monthly.

I'm not so sure about the MDD. A good idea, but as ST says, we'd have to track returns every day. Sounds like a lot of work.

 

[FundSurfer]

I agree that the MDD indicator would be a good measure of risk. I'm not so sure it will be easy to implement. I'll give the spreadsheet some thought, perhaps I can write a macro that could make the calculations but I'm not promising anything. I'd be against anything that increases the amount of time spent entering data. I've done it for a while with about half as many people as milk_man currently enters data for and it takes a lot of time. I've automated the Tally to a point but there is only so much you can do with a spreadsheet.

 

[tsptalk]

I agree. Too much work. Plus the spreadsheets would get huge.

It wouldn't be a bad idea for people to do that individually for their own info. But not for one person to tally everyone's.

 

[oldschool]

Maybe simply looking at monthly divergence from, say, the C fund might be a meaningful but cheap proxy for MDD? If so, that's pretty much in current tracker data. I agree MDD way to much to do as a daily matter.

 

[SystemTrader]

Though I was the one who brought up the MDD, I agree about the workload. Unless someone could write a macro and automate the process, as FS mentionied, it would be too much. Milk and FS have more than enough on their plates already with the ever-growing number of particants.

Maybe simply looking at monthly divergence from, say, the C fund might be a meaningful but cheap proxy for MDD? If so, that's pretty much in current tracker data. I agree MDD way to much to do as a daily matter.

 

[Wheels]

Though I was the one who brought up the MDD, I agree about the workload. Unless someone could write a macro and automate the process, as FS mentionied, it would be too much. Milk and FS have more than enough on their plates already with the ever-growing number of particants.

Furthermore I'm not convinced that weighing risk is appropriate for this particular "contest" (or whatever we are calling this). I think highest returns, pure and simple, is the way to go.

Just my 2 cents.

Dave
<><

 

[Birchtree]

IMHO - 14% of no money is still no money. Often investing behavior is dictated by your balance. Pistol shooting is trying to make a quick hit - where as buy and hold strategy may mean previous arrival. Percentages can certainly be misleading as an indicator of success. Simple though they are.

 

[Wheels]

IMHO - 14% of no money is still no money. Often investing behavior is dictated by your balance. Pistol shooting is trying to make a quick hit - where as buy and hold strategy may mean previous arrival. Percentages can certainly be misleading as an indicator of success. Simple though they are.

Are you suggesting that this contest should be based on dollars instead of percentages? That's real fair. Don't be so sure you are the biggest dog in this pound.

Dave
<><

 

[Birchtree]

I will be if I live long enough. There are so many bright days ahead to invest and make green dollars. Most participants with power accounts have a tendency to hesitate - I'm full speed ahead in my own little way. Waiting for the economic slow down and holding a pot full of wall flowers that will spring to life with the right application of superlative bull manure. Wish many times that I were back in New Hampshire - picking raspberries. Take care.

 

[rokid]

Furthermore I'm not convinced that weighing risk is appropriate for this particular "contest" (or whatever we are calling this). I think highest returns, pure and simple, is the way to go.

Just my 2 cents.

Dave
<><

Even though it means I'm out of the game before it begins, I agree with Dave.

However, how do market timers gauge risk prior to the outcome? Someone suggested number of moves. However, looking at the tracker data, I don't see a correlation (and I really tried to find one ) between number of moves and risk (as measured by standard deviation).

Just curious.