Sat 8 Jul 2006

This will be the most juicy post thus far regarding PowerRatings. I signed up for the free trial (required a credit card, so I used one that’s set to expire on 7/31/2006–ha ha). I went straight to see the top 25 stocks with PowerRatings. They are posted below:

There were no stocks with a PowerRating of 10.

Stocks with a PowerRating of 9:

FMD– current price of 47.65

COGO– current price of 10.49

ASPV– current price of 25.82

Several stocks with a PowerRating of 8 showed up as well, but there were too many to comfortably list here and after going through all the S&P 500 data, I’m not in the mood to type all those out.

For reference, here’s what TradingMarkets website says about the PowerRatings stocks: *From 1995-2005, stocks rated 7 have outperformed the S&P 500 by a better than 4-1 margin over a five-day period. Stocks rated 8 have outperformed the S&P 500 by a better than 8-1 margin; stocks rated 9 have outperformed the S&P 500 by a 13-1 margin, and those stocks that have achieved a 10 PowerRating have outperformed the S&P 500 by a 16.9-1 margin.*

Stocks with a PowerRating of 1 (meaning that from 1995-2005, stocks with this rating underperformed the S&P 500 by a factor of 4 over the next 5 days):

MDTH– current price of 21.36

RSAS– current price of 27.28

WGR– current price of 60.01

As with the 8’s, there were multiple stocks with a PowerRating of 2.

They have a section on the site that shows stocks that scored 9’s or 10’s in the past 2 days and you can view their charts. Only one was listed: **AXR**. AXR had a PR of 9 on 7/6, a PR of 8 on 7/7, and it’s currently marked with a PR of 6 for Monday (7/10). Assuming you had this data on the evening of 7/5 (*the new PR ratings come out at 7:00 pm EST, so you would have been able to setup a trade for AXR that was set to buy at the opening of 7/6*), you would have bought AXR most likely at 50.5 on at 9:30 am on 7/6. At the close of 7/7, AXR was at 53.27 a 5.5% 2-day profit, minus commissions, assuming you sold right at the close. Since the PR rating has fallen to 6, I’m interested to see if you should cash in and take your 5.5% profit immediately or hold on for another 3 days.

Another interesting part of the site is their list of stocks with the largest change in their PR either up or down. CLRK just changed from a 2 yesterday to a 6 today. LBIX, ENS, FSII, and SERO each just changed from a 4 yesterday to a 7 today.

There’s plenty to discuss and I’m looking forward to an interesting week looking at PowerRatings. If PowerRatings can quickly provide a small list of interesting stocks that have a high probability to move in an expected direction quickly, it’s worth $50/month without a doubt. Here’s a list of the followup items I’d like to focus on regarding PowerRatings:

- Finding Optimal Trading Strategies using PowerRatings. The current assumption is to just look for 9’s or 10’s and trade them as long as they remain above 7. An alternate assumption is to just look for 9’s or 10’s and trade them for 5 days and take profit sometime on the 5th trading day.
- Finding Optimal Money Management Strategies using PowerRatings. I’m assuming that these volatile stocks likely will need somewhat wide stop-loss orders on the order of 5% or 8% per trade. With a stop-loss of 5%, can an overall profit be made or will too many trades be stopped out before fruition? How many times does a 9 or 10 bust (draw down 10% or more) within a 5 day period?
- What option is best for trading multiple 9s and 10s? On any given trading day there could be 3 or more stocks with 9s or 10s. Let’s assume that you had $50,000 allocated to trade PowerRating stocks. On Monday, 3 stocks with 9’s show up, so you would invest $3,333 in each of these stocks. On Tues, 2 new stocks show up, one 9 and one 10–I’m assuming that you would invest $5,000 in each of these stocks. And so on throughout the week evenly dividing your $50,000 amongst these stocks. If there were no 9’s or 10’s one day, consider looking at the 1’s and shorting those with the $10,000 allocated for that trading day. The basic idea would be to not trade any more than 1/5th of the total investment amount on any given trading day.
- How long do stocks stay as 9s or 10s? If a stock can remain a 9 or 10 for more than one day in a row, do stocks that have been 9s for two or more days out perform stocks that are classified as 9s or 10s for only one day? Is it better to purchase a stock on the second day it’s been a 9 or 10? Or is it better to purchase all stocks on their first day as a 9 or 10, but double the position if they remain a 9 or 10 on the second day?

July 9th, 2006 at 11:40 am

On point #3…

I’d recommend position sizing based on the risk per trade rather than on an absolute value per trade.

For example, suppose you have $50k in your account. You determine you don’t want to risk more than 1% on any trade, so that is a $500 nominal amount to risk. You see a stock you want to buy at $40 and figure your stop loss is a 10% decline. Since your loss is $4 per share, you would buy up to 125 shares ($500 max loss / $4 per share potential loss). How many trades you can put on becomes a matter of arithmetic.

I would probably recommend trading even smaller until you know the performance behavior of power rating stocks and your own ability to trade them profitably (What happens if you’re not able to open positions at 9:30 when the signals are made? If a position goes against you, does it do it so violently that you can’t maintain your stop losses and truly limit your risk?).

Also, don’t forget to factor in the $50/month for the service in your profitability calculations. If you have a $50k account, the $600 per year is 1.2% of your account size… if you have a $5k account you have to clear more than 12% return per year just to break even.

July 9th, 2006 at 1:02 pm

To add to the position sizing discussion, you really need a sample of past trades including commission and slippage to get an idea of what % of equity to use in deciding to trade. % of Equity is the stat to always refer to and is what Jason is saying. For any given trade, decide to risk a certain % of equity and then find out how many shares get you that % based on your stop loss. This way, things are self-adjusting. As you account grows, so will the shares you will by based on that % number. So compounding will happen naturally.

But to get that % number is the rub. The best approach I’ve found is once you have some reliable trading record, run those numbers through a monte carlo simulator (plenty of free ones are available for Excel). This will simulate what trading might be like in the future and can give you an estimate of drawdowns. Adjust your % of equity to give you a drawdown you are comfortable with and viola. Until then though, you should trade the minimum and not worry about optimal returns. Once a record is established, a tool like Monte Carlo will help you squeeze out more returns.

As to PowerRatings and the cost of them, that’s a very good point. Small accounts will have a hard time overcoming the overhead. Is there any clue as to how these ratings are derived and could you simulate them for free. I mean most people pay for the luxury but with a little extra time you could save yourself the dough.