Top 5 R2G Models - 1Wk / 4Wks

When you are shopping for an R2G model, don’t look at long term performance such as CAGR or annual returns, simply because there is no long term performance data. Back test data, while interesting from a “potential model performance” viewpoint, is not relevant from an actual performance perspective. There are many flavors of optimization that go into back test which may or may not prove advantageous going forward.

The following are the top five models for 1 week and four weeks. Models with insufficient performance data are crossed off the list.

Steve



Hi:

Although I agree that back test statistics may not be good guides for how well a R2G port will do in the future, I think 4 week results for actual performance are not any better. A month is just too short a time in my opinion.

I personally would like to see at a minimum 6 months of live results.

Regards,
Brian

Brian - this is true. But because the in-sample and out of sample performance is not kept separate, you would need to wait several years for out of sample to have a significant impact on the CAGR, profit per trade, etc.

Steve

Steve:

After six months I would be looking to see that the port is outperforming the benchmark by about the same ratio as it did for the 3 year stats for the port. As you say, it would be even better if in and out of sample results were automatically kept separate.

Congratulations on the good start of some of your ports.

Brian

So comparing 6 months / benchmark versus 13 years / benchmark is a fairly laborious process. Think about it, you would have to open up 52 separate models, determine the launch date (is it 6 months old?) and then do your 6 month versus 13 year thing.

So what will really happen is people will sort on the highest CAGR and then go through the first 10 models, probably never look at the rest.

My systems that have performed well are defensive sectors. It is probably the place to be this summer.

Steve

It would be helpful if the launch date were posted on the performance page, or maybe better yet if P123 were to highlight the numbers that are in sample on the performance page.

Don

I think people need to compare against a more relevant benchmark than the SP500 before buying. The utilities model shown topping the list has performed pretty much in line with the SP500 utilities index. It may have some merit, but there isn’t enough real world data to make sensible decision at the moment. Similarly, people should compare with the small cap indices when examining a small cap model.

Roger - that is an excellent point. Unfortunately I didn’t find an SP500 utilities index or consumer staples index for that matter.

Comparing Stitts Utilities LC to RYU (Guggenheim SP500 equal weighted Utilities), it looks like Stitts comes out ahead by a few percent since launch.

Steve

Utilities and Staples are GICS 10 Sectors. Their BB codes are:

S5UTIL Index
S5CONS Index

If you don’t have BB, alternatively, use the most liquid ETF as the benchmark.

For utilities, it would be XLU.
For staples, it would be XLP.

These are not options that P123 provides for benchmark.
Steve