Mousetrap 3/24/2012 -- why market timing doesn't work

[Timothy Clontz]

Condition Bear Market Rally
S&P Target 1020
Small Portfolio IAU & XLF 15.61%
Hedge XLU -1.21%

Position Date Return Days Call
GCI 7/14/2011 17.66% 254 Hold
CSGS 10/3/2011 23.42% 173 Hold
NLY 10/25/2011 2.78% 151 Hold
DD 10/27/2011 16.09% 149 Hold
KBR 10/27/2011 26.01% 149 Hold
VG 10/27/2011 -32.52% 149 Buy
TTM 11/30/2011 59.23% 115 Hold
BT 1/4/2012 18.28% 80 Hold
PDLI 3/7/2012 4.11% 17 Hold
CLF 3/19/2012 -1.74% 5 Hold

S&P Annualized 4.73%
Small Portfolio Annualized 19.13%
Mousetrap Annualized 23.39%
Hedged Annualized 21.91%

Market timing is the holy grail of investing. If you can be long SPY when it’s going up and short SPY when it’s going down, you can make a mint.

There are a few – very few – who can do it. My friend Len can do it. EffectiveVolume.com can do it.

A lot of other folks WOULD be able to do it too, except for one problem: timing the market means that your trading costs accelerate. If you trade just under once a year, you might have trading costs of 1% or less, and 30% capital gains taxes.

At just longer than once a year the capital gains are cut to 15%.

But market timers generally trade every few weeks, with stop losses in place. Day traders are even worse. A day trader with just one trade a day and a nest egg of 2000 dollars would lose at an annualized rate of 365% a year. At 20,000 dollars that’s 36.5% a year losses on trading costs – BEFORE you even consider the wins. I mentioned last week that a person using something like my Mousetrap model with 20,000 dollars would have to hold over two months just to break even – BEFORE you even consider the wins.

And that’s why market timing doesn’t work: trading costs and taxes eat you alive, UNLESS you are that extremely rare person like Len or the folks at EffectiveVolume.

So how long should a person hold? Ideally, more than a year.

You’ll note that I haven’t been doing that, and that’s why my performance has been so bad. Yes, a 23.39% annualized rate is rather attractive. But if I had not sold any of my stocks my annualized gains would be at a 43.13% rate.

That’s a 20% performance hit because I was trying too hard.

The good news is that I’ve paid attention to the bad news, and I’m now mining data to find the ideal holding period. So far it appears to be around 558 days per trade. That will fine tune as time goes on. But it gives me the happy circumstance of making a trade every two months or so as I rotate between ten stocks – and the next trade doesn’t have to be for a long while. Granted, if there’s a drastic money flow change like there was with GTAT last year, I would make that trade. But I believe GTAT was the only stock of the first ten I should have traded by now.

This model is about to get REAL boring, then, with very few changes.

My realized return goal is 30% a year – on average. I think, by slowing down… slowing WAY down… that’s a reachable goal. Others may be able to do better, but 30% after trading costs and taxes would meet most folks needs… certainly my own.

We’ll see.

But with only a trade every two months or so, I might be able to take a vacation AND sleep at night too :-).

Tim

[Neil Stoloff]

... I’m now mining data to find the ideal holding period. So far it appears to be around 558 days per trade ...

Tim,

Is that 558 calendar days or 558 trading days?

Neil

[Timothy Clontz]

Calendar days.

[Harry]

What about frequent trading within a Roth? As I understand, as long as you don't trade the same equities with other accounts (avoid wash sales) you can trade as frequent as you want without tax implications.

[Timothy Clontz]

Yes, an IRA account avoids the tax burden, but not the frequent trading costs. It's those little 9 dollar transaction costs that kill you.

A system has to beat the market AND taxes AND transaction costs AND inflation to truly "beat the market". I think Billy and Pascal can do that, but most people can't do it on their own unless they use a fundamental approach, instead of a technical one.

[Mike]

Yes, an IRA account avoids the tax burden, but not the frequent trading costs. It's those little 9 dollar transaction costs that kill you.

A system has to beat the market AND taxes AND transaction costs AND inflation to truly "beat the market". I think Billy and Pascal can do that, but most people can't do it on their own unless they use a fundamental approach, instead of a technical one.

I don't really understand the transaction cost problem. My position sizes are usually >$100K and I wait for 20-25% profits to be had. I time the market all the time and transactions costs are nil. If I trade 100-200 times a year my costs at $7 per trade is $700-$1400, much-much less than the profit on a single trade.