Fri 14 Jul 2006
Ray Johns has been publishing his model portfolio on Daytraders.com since 1997. For $39.95 per month, stock traders can follow along. His live performance record is reported on the site but what can be made from it? I decided to find out so I compiled them all into this Excel spreadsheet which assumes a $1000 investment in each trade call made since 1997. Now when you look at this your probably going to have two reactions, amazement and disbelief. So let me help you manage your feelings. First, I did subtract commissions and used IB commission rates assuming that any trader in their right mind would only use IB for short-term trading. But the chart as seen when you first open the file assumes no slippage. Now this is very unrealistic because no system is without slippage but also because Ray’s calls are not exact like “Buy X at 9am with a stop loss of 5%” or “Exit Y at the close”. Instead they are more like “Yesterday I bought XYZ and closed out ABC”. So subscribers will always experience some lag and this can’t ever be reflected in the historical data. He makes is live calls in a chat room service.? But since his trades are held for at least a few days to weeks, you should be able to get some of the move he gets.? The question is how much slippage will a subscriber have? Some lag might lead to actually outperforming Ray and other times it will lead to lesser performance. Is it 1%? Less than 1%? Greater? There is a cell in the spreadsheet that will allow you to see the effect of slippage. AH, the power of slippage. Note what happens at 1%. Not much. Now 2%. Hmm, it’s clear now that since the end of the bubble, things haven’t come as easily and that’s not too big of a surprise. At least it didn’t lose money. But past 2%, it all falls apart.
Only time and forward testing will tell what slippage rate one can actually expect. If it is low, then this is a pretty great track record. Otherwise, it’s a pipe dream. So the next step is to perhaps organize a little test using the 2 week free trial offer. Anybody up for a little tag team trail where we each take consecutive trail periods and see how we do relative to Ray Ray? That would give us 6 weeks of data which would probably provide a good clue into how close we can get to his performance record. We can establish some sort of non-discretionary entry or exit rule like “Trade on the open (or close) the day after he mentions a change to the portfolio” or something like that. We could even use a simulated trading site to make it more realistic.
July 14th, 2006 at 2:32 pm
I already tried out the free trial, so I can’t contribute to the trial periods… though sites like this usually key off the email address, in which case I have plenty of email addresses. 😉
I think the lag reduces its usefulness as a trading system and instead you’d just be watching someone else trade and trying to follow along (most likely with limited success).
July 14th, 2006 at 9:10 pm
Yeah, that’s the rub. But it is true that there are some guidelines provided with each trade like “attempt 2.5 profit target and cost average at -10%” so one way to test it out would be to make it as mechanical as possible and see how different it is from the posted results then extrapolate that back over the whole history.
I’ve some extra emails as well so I might get this thing started next week.