System with 100% Accuracy Over 55 Years: "Bear Market Has Started"

This is a cross-post from another thread that was getting very long, with 80 replies. I feel it deserves its own thread. Here is more than a plate of food for thought:

You and I are going to have to disagree fundamentally on whether market timing is possible. Not only is it possible, but it is easy if you take the time to learn the techniques. I am not expressing an opinion; in this post I’ll provide objective, empirical evidence, backed by an academic study to support this statement.

The vast majority of fundamentally oriented analysts have the same attitude as you; I know I also did for my first 25 years. I just sucked up the losses sticking with fundamentals from the time I started in this business as an equity analyst with Drexel Burnham Lambert in California in the early 1980s.

However, when a major health problem limited me to working from home 12 years ago, I launched my value-oriented newsletter that included model portfolios. It was very successful from the start, but I soon discovered that my individual-investor subscribers would cancel en-masse’ whenever there was a downturn. To deal with this challenge and at the prompting of a mentor, I dedicated myself to learning technical analysis.

I implemented a market timing system, which I call the Intelligent Market Risk Analysis (IMRA) into my newsletter back in 2006, and it was a very timely decision. This rules-based system told me to to exit the market in October 2007 and then signaled to issue a significant MARKET UP signal to subscribers on March 9, 2009. The timing was perfect to catch the upturn. When combined with my P123-based quantitative value systems, my small-cap model portfolio produced a return of 1600%, and my large-cap portfolio logged a 700% return that year.

Closer to present, on June 28, 2015 I published a free newsletter titled, ‘Big Change Coming to the Market This Week ,’ and notified p123 members of its location in another post. I placed tight stops on all of my model portfolio stocks and gradually exited the market through last July.

In that newsletter, I demonstrated a simple, three-indicator analysis that is run on a monthly basis to identify major market turning points. This system is so simple anyone can use it. It uses a 10-month simple moving average, the MACD, and my proprietary settings on an indicator I call the ‘Market Risk Oscillator.’ In that newsletter, I showed that MACD had already rolled over, but the other two indicators were not there yet. However, it appeared that a MARKET DOWN signal was imminent.

That situation has changed now. All three indicators have rolled over and the system is signaling that a bear market has begun. Here is the current chart, which I published in this past Sunday’s (free) newsletter:

Here is a close-up of the indicators for the last 1.5 years:

In testing on the TradeStation platform, working with Tobias Berr, we found that when applied to the S&P 500 index starting in Jan 1960, this simple system has 100% winning trades and turns every $100 invested into $2,102. Max drawdown is less than 10%. It does not use shorting and simply goes to cash when these three fundamental rules signal a significant downturn is beginning. My Intelligent Market Risk Analysis, which I use for the newsletter’s model portfolios is more sophisticated and uses weekly signals for faster, more accurate signals. It produces even more powerful results.

Market timing IS possible, and it is actually not that difficult. The decision to use it or not makes a world of difference in both your wealth and peace of mind. I know this is easier said than done, but all it requires is a dedication to learning TA, the same as the dedication you devoted to studying fundamental analysis. Or you could just use a simple system like the one I displayed above to keep you out of the big downturns.

Just to back up my appeal with objective data, there was recently an interesting article by Mark Hulbert, titled, “Technicians vs. analysts in a CNBC stock-picking slapdown: Who wins?” published on MarketWatch.com. Hulbert summarizes an academic study of thousands of recommendations made on the CNBC TV show “Talking Numbers.” The focus of the study was a thousand pairs of recommendations made between November 2011 and December 2014. The result isn’t even close. Read the article here or the actual study on SSRN here .

For those interested, I am now publishing my Intelligent Value Alert newsletters for free. The index for 2015 editions is located at [color=blue]http://www.intelligentvalue.com/2015-value-alerts/alerts.htm[/color] . You can sign up to for notification when I publish a free newsletter at [color=blue]http://www.intelligentvalue.com/free.htm[/color] . Don’t worry, I don’t send spam or marketing. There are also links to samples of my weekly Market Risk Analysis publication and other free content on that page.

I would say, ‘best of luck out there,’ but we can take chance and randomness out of the equation by applying thoughtful technical analysis to supplement our p123 quantitative systems. As I recommended to my subscribers back in late June, the best advice is to go to cash now. However, we don’t have to sit on the sidelines in cash. I will be picking some robust ETFs for paid members in this weekend’s (Oct. 4) portfolio updates that will profit from the coming bear market.

Christopher I don’t know if you’re trying to sell memberships to your service or what but for the sake of integrity you need to give credit where its due, you are not the first to post a similar chart. I posted the same style of chart weeks ago and I’ll post it again.

This chart is easily produced on Stockcharts.com

Plus you really should make reference to the white paper published by Meb Faber in ahemm… 2006 who has conducted extensive research into the 10 month SMA.

Meb Faber “A Quantitative Approach to Tactical Asset Allocation”


Hi Brad,

I think you need to read a little closer and mellow a bit. I’m not trying to sell subscriptions at all. All the links I posted are for free content. I am just trying to advocate for market timing here on P123 so that others (such as Marc Gerstein), who are not open to the idea, might benefit. (No insult intended, Marc. I admire your work.)

Also, I did provide the settings on the chart so it can easily be reproduced on a TA site (such as StockCharts, which is what I use). You say you published your chart a couple of weeks ago. However, if you had read closer, you would see that I noted in my initial post that I previously published my chart free to the public on June 28 in my newsletter, ‘[color=blue] -value-alerts/150628.htm][color=blue]Big Change Coming to the Market This Week[/color] [/color].’ I posted a link to that free newsletter here in the p123 forums with a warning that a downturn looked imminent. This was almost two months before the August selloff.

Using the 10-month MA is an old technique. It’s everywhere, so I didn’t feel the need to look for the link to Faber’s paper. It’s not some intricate, magic formula that he created either and was being used decades before Faber wrote his study. Charles Dow was using it in the early 1900s. The Japanese were using it in the 1800s. It’s the same as a simple 200-day MA, except it avoids whipsaws better because it’s applied on a monthly basis. My post was intended to show that a very simple market timing system can help investors avoid subjective decision making, stress and losses.

I have published that same monthly chart occasionally for many years. It’s basic, but it’s easy to understand and it can aid immensely in avoiding the big losses that can destroy an investor’s returns. It tracks the market’s upside momentum, and that momentum has now dissapeared on a large-scale, monthly basis. Your chart provides similar indicators that track the market’s momentum. There’s nothing special about these charts. I just noted Marc Gerstein’s statement in another post that market timing doesn’t work and wanted to rebut it.

This selloff may be similar in intensity to 2008, as in -50%. What will be key is if/when the S&P 500 drops below its August 24 and mid-October 2014 lows around 1860. That would be the ‘point of no return’ and look out below.

My newsletters are free now for as long as I enjoy writing them. I’m just trying to make a contribution, share 30 years of experience, and give back a bit to the individual investor community, in this case via p123. I had no motives of promotion at all. The newsletters are free now. Enjoy.

PS: I noticed that we live in the same town, Brad.

Hi Christopher,

I think I missed the settings. How is the market risk oscillator calculated?

Best,

Walter

Hi Christopher

It appears even the academics are beginning to agree with you: http://www.marketwatch.com/story/technicians-vs-analysts-in-a-stock-picking-slapdown-who-wins-2015-09-04

From the SSRN paper (attached below):

Cheers, Donald


technicians vs fundamentals.pdf (1.66 MB)

Not impressed.

The big chart you posted shows your model caught 2000 and 2008. That’s no big deal. We know we have models that can do that. They were published on p123 and lots of members use them or variations. And i know nobody likes to hear this after the fact, but back at the time, before the p123 models were created, I stayed out of the market because of my brain. Ditto 1987, the first crashI experienced as an adult.

I notice that your signal missed the 2011, one which although it doesn’t look so horrible on charts was, as i and I’m sure many other remember, was pretty bad at the time . . . especially since you don;t know where the bottom is until after you’ve gotten there and turned around. But 2011 is significant beyond that. It represents a different type of decline than we’ve been experiencing in the past. It was not driven by fundamentals or even internal market supply-demand dynamics. It was determined by the news cycle and the now-extreme levels of decision-making concentration. Going forward, we have to be concerned about that – and more so as the big-money computers are now starting to adopt and trade based on text-based algorithms. Because it was only 20% or so, 2011 may not be an idea numerical test. But in terms of the broader market dynamics, it does, I believe, make for a pretty important sample . . . and one that must be properly addressed by anyone offering a forward-looking timing system in which i’d have confidence. (By the way, you also missed 1998, another period of significant financial crisis that before it was rescued by external behaviors looked to investors back then as something that was in the process of imploding the system; also, you claim 55 years but only go back to 1996. Can you show the rest.).

Good comments, Marc. I agree fully.
Market Timing can be attempted but there will never be a 100% method. The underlying dynamics can change overnight. Not even speaking of Black Swans which, by the very definition, can never be predicted. Therefore market timing will work - at times. The infamous 200-day average that is so much worshipped by millions is a crude example. At times it has also failed bitterly.

Maybe some market timing is better than no market timing though.
Geov has developed some interesting concepts and “combined timers”. Very useful in my opinion.
But just like ANY indicator(s), nothing will ALWAYS work.

To my eye, it looks like your system also gave a sell signal in 1998 and 2006. The bull markets continued a while longer on both occasions.

Hi Chris,

I concur with Marc. Your monthly MACD is a curve-fitting exercise which only catches the long-term declines. Also, because its a moving average, it has an inherent delay. For this reason your buy signal comes very late. In the 2009 example for instance the rise from oversold only occurs later in the year in your chart, when the index has risen to 1000 points (i.e. after a 50% gain from the low at 666 points).

In my opinion the are only two ways of effective market timing:

  1. a constant hedge or market neutral strategy;
  2. a high-frequency hedging that looks at short-term weakness in the market.

I tried to implement the latter in my latest R2G:
https://www.portfolio123.com/app/r2g/summary/1383819

This strategy looks at the weakness within the universe, instead of a benchmark. It has many hedge perids in uptrends, and thus also catches smaller retreats such as the ones in July 2011 and October 2014. I am in no doubt that even this approach will fail at some point, because the future might hold more dramatic swings than we can imagine.

Best regards, Florian

WTF . . . Why did you cite this paper?

It shows absolutely nothing relevant here except the usefulness of the AvgRec data-point, which is known to be weak.

it doesn’t measure fundamental analysis at all. The authors say they do, but they botched it horrendously. They define recommendations based on fundamental analysis as being explicit analyst strong buy-buy- etc ratings, or what the authors are able to infer along those lines based on how enthusiastically the analysts discuss them on TV. The stocks considered are, according to the paper, “assets that make headlines in the financial media. Examples include stocks of prominent firms that are about to post financials, hot sectors, hot markets, and general assets experiencing substantial price changes.”

That’s not fundamental analysis. That’s pure unadulterated garbage. Fundamental gurus like Buffett and Lynch are particularly clear that those stocks are poison. The big benefit of platforms like p123 is that it gives us a way to ignore crap like that and find stocks with genuine fundamental and/or technical merit. And the authors even cite research by (competent) academicians that long ago demonstrated that analyst rating are terrible as stock picking devices (changes in ratings might be better).

I didn’t see the Hulbert thing, but according to your post, he cites the p[aper as comparing technical analysis to analyst recommendations. He’s right. that’s what the paper does. And it’s conclusions are not at all surprising. Presumably systemic technical analysis outperformed seat-of-the-pants wishes by analysts who serve clients by gathering and distilling facts on the companies they cover but have long been known to be lousy stock pickers. You don;'t need technical analysis to beat them. A dartboard will do just fine.

“I did provide the settings on the chart so it can easily be reproduced on a TA site (such as StockCharts, which is what I use).”

I don’t see these settings. Would you kindly post them? Monthly settings to replicate the chart would be nice. Weekly would also be appreciated to support your assertions that it works better.

I am working to see if a reliable signal can be created and your openness and transparency would be appreciated as the oscillator looks interesting.

http://www.fool.com/investing/value/avoid-the-mistake-that-cost-buffett-8-years-of-bet.aspx

Believe, Mgerstein models are similar to Buffet and IBD ideas, 20%+ avg return long term investment.
His Chainkin market timing R2G model up for last 3 months, 6 months and 12 months.

Most of the R2G description says keep 80% investment in long term; rest of 20% in technical bets.
BRK.B was down 50% from it’s recent high in 2009. In 2014 it went from $100 to $150 (50% return).
BRK.B has return of 20% avg for last 50 years;

Both the systems always works; Need to follow with patience and discipline.

Thanks
Kumar

Chris, I would be interested to see the summary of basic P123 portfolio backtest with all available P123 data that tests the 3 indicator timing method, and invests in long SPY, cash, and long SH, according to what the timing method would prescribe.

Tom C

Wow, my post advocating using market timing here on p123, which goes against Marc Gerstein’s advice, elicited a substantial amount of interest. I’ll answer some of the questions and comments in this post.

Marc, I wasn’t trying to show a sophisticated technical analysis approach that would solve all of a P123 member’s risk problems. However, it certainly solves a lot of them (details below).

In my original post I said, “In that (June 28) newsletter, I demonstrated a simple, three-indicator analysis that is run on a monthly basis to identify major market turning points. This system is so simple anyone can use it.” and “My post was intended to show that a very simple market timing system can help investors avoid subjective decision making, stress and losses.”

In other words, it’s a very basic system that can be implemented by anyone, without feeling the need to master technical analysis. TA can be intimidating, and with standard indicator positions is set up for day-traders. Because of this, TA frequently whipsaws longer-term investors (as opposed to traders) and frequently appears to users to be useless. I often hear it said very definitively that (in your words, Marc) “I don’t think we can time the market…” What the person really means is “I can’t figure out how to time the market.” My post was simply a rebuttal to your statement and an advocation to other members that the market CAN be timed.

I offered an example of a very simple, momentum-based system for doing so. Yes, it does only catch the major turnpoints, but in its defense, for the average investor (or even us), it has a 100% winner ratio over 55 years. It’s not the market timing system that I use, but it works and is easy to demonstrate and use.

Marc, you are the ’guidance counselor’ and an instructor on what has become a very sophisticated (some would say ‘complicated’) website in Portfolio123. Most of your work is 5-star quality. Many members, especially the newbies, pay close attention to what you say. However, when you make a broad-brush statement saying that the market can’t be timed, you potentially forever taint those individual’s perspective regarding what’s possible. Words matter. I respect your extensive experience and body of work, but no one has all the answers, and perhaps you should keep an open mind.

After all, what could possibly be the reason to try to dissuade the average p123 member from using a significant capability of the website? There are already many members, including George Vrba, Tom Yani, Walter, Chaim, me and a number of others who are successfully using and profiting from sophisticated (or even simple) technical analysis algorithms created in P123.

In my original post I said, “It uses a 10-month simple moving average, the MACD, and my proprietary settings on an indicator I call the ‘Market Risk Oscillator.’” Sorry, but the settings on the Intelligent Market Risk Oscillator (IMRO) are proprietary. You can use the simple 10-month moving average along with the MACD (12,26,9) for confirmation, without my indicator and still get very good results (I show the long-term results below). My IMRO helps to establish times when the other two indicators may be giving signals that will result in an unnecessary whipsaw out of the market.

Regarding the Mark Hulbert article about TA beating fundamental analysis based on a long-running CNBC TV show… [quote]
Marc Gerstein: “WTF . . . Why did you cite this paper?”
[/quote] That’s a pretty harsh attitude, Marc. Do we have Orwelllian word police on P123 now? I see that P123 member David (motu) also made a post about the article in this thread. I simply said it was and “an interesting article.” I didn’t make any claims about its validity or authority.

Of course the stocks hyped all day on CNBC are crap. In fact, I haven’t turned on CNBC (or any other market-related media) in many years. I rely solely on P123 for stock selection for IntelligentValue.com and my personal account and have done so since 2004. However, it’s not professional to imply someone is stupid or has done something wrong simply to post a link because it is ‘interesting’ food for thought.

[quote]
Marc Gertein: “…also, you claim 55 years but only go back to 1996. Can you show the rest.”
[/quote]Yes, here are screen captures for the system going back to 1960. This simple market-timing approach was tested on TradeStation in coordination with fellow p123 member Tobais Berr. There is a link below each chart to open a full-size view.

1960-1987

Link for large version: [color=blue]http://www.intelligentvalue.com/2015-value-alerts/images/150927_1960-1987.jpg[/color]

1991-Sept 25, 2015

Link for large version: [color=blue]http://www.intelligentvalue.com/2015-value-alerts/images/150927_1991-2015.jpg[/color]

If you have a look at the charts above, you can see that the signals are pretty accurate in keeping an investor out of trouble by avoiding the heinous drawdowns that destroy confidence. A significant number of individual investors have missed the record-setting bull market of the last six-years out of fear from the last crash. A simple system like this would have avoided that crash and the lingering nervousness about the market. There are very few whipsaws in the system (I don’t even know if you would consider a period of several months a ‘whipsaw’).

This very basic, unsophisticated market timing system, using 1) the 10-month moving average, 2) a standard monthy MACD, combined with 3) my custom Risk Oscillator and run on the S&P 500 index achieves 100% winning trades.

The system had a total of 15 trades over the 55 years studied (actually it comes out to 49.59 years because the first market entry is on May 1, 1964). The winning trades lasted an average of 29.67 months and there were no losing trades. Time in the market is 69.76%, leaving about 30% of the time in cash. Of course, this excercise is for a basic, long-term analysis of a very simple timing system to determine its robustness. I would say it passes that test. Today, there are many alternatives to satying in cash all those months.

If you were to run this same simple system on the entire database WITHOUT my IMRO indicator (which has settings I am going to keep to myself for now), it is still a great winning approach. It produces 16 winning trades and 4 losers, for a winning percentage of 80%. The winning trades lasted an avg of 29.38 months and the losing trades lasted an average of only 3.25 months. In this version, the system produced a total return of 8,742% for an annual return of 8.70%.

While that may not sound like much, a single starting investment of $100k becomes $8,742,495 over the time period reviewed. This compares to a return of 7.44% for the S&P 500 over that same period. That works out to an ending balance of $3,509,217, for a difference of $5,233, 278. The 1.26% annual difference attained from using a simple market timing system results in 2.5 times the balance at the end of the period! Of course, applying sound, fundamental quantitative stock selection via p123 instead of simply running a system on a large-cap index like the S&P 500 would have produced a return that would be dramatically higher.

Florian, I hope that the 55-year charts and stats I show above will eliminate the accusation of ‘curve fitting.’

The system I used as an example is not something I use either professionally or personally. It is a very basic view of two momentum indicators accompanied by one of the technical oscillators I use on my website. It was not intended as an example of a sophisticated market-timing system. Every Saturday I publish on IntelligentValue a review of market conditions, called the Intelligent Market Risk Analysis (IMRA), which assesses the risk of market conditions causing overall portfolio losses and provides guidance on prudent portfolio exposure levels to my subscribers. The charts I publish there use a combination of six indicators, although for the sake of simplicity I only show two in my newsletter. There are some samples of recent issues of this publication in the ‘free’ materials section of my site.

I have spent the last nine years refining that system and it does not rely at all on moving averages or the MACD. You are correct that moving averages have an inherent delay. After all, they are backward-looking. However, my point in displaying the ‘Big Picture’ monthly chart with two very simple technical indicators is that the market CAN BE TIMED. And, as shown in the charts and statistics above, which cover 50 years, it is definately not ‘curve fitting.’

REAL-TIME EXAMPLE

As an example of putting market timing into live action (not backtest), after I started using technical analysis for my newletter in 2006, my new IMRA system got me out of stocks in October 2007 and back in the market precisely on March 9, 2009. I had spent the entirety of the downturn examining which factors/formulas worked coming out of other major market bottoms. I knew another bottom was approaching and wanted to be ready when my new TA skills said it was time to go ‘all in.’

When my system gave me the signal, I launched two new live portfolios on IntelligentValue, both created in P123. One targeted solid, profitable Russell 1000 stocks that were selling at a discount on multiple fundamentals to their peers. I dubbed it the ‘Relative Value’ portfolio. The other portfolio was focused on small-cap stocks with a min of $100m mkt cap and $3m ADT, which were selling at a deep discount to their liquidation value and cash. I called this one the ‘Deep Value’ portfolio. Both have live updates on my website each week and I have dozens of long-time subscribers who will attest to that fact.

I applied my newfound TA skills, combined with p123, to these portfolios as the market progressed over the last 6+ years and I am more than pleased with the result. I’ll use the example of my mid/large-cap Relative Value portfolio, since it has lower returns of the two: Total return = 6,518% (benchmark return, 192%), Annualized Return = 89.73% and Max drawdown = -23.12%. Overall winners = 78.18%, Annual Turnover = 298%, Sharpe Ratio = 1.58.

This is the Relative Value portfolio, showing the periods when market timing told me to be out of the market and in cash, along with max drawdown:

In my opinion, that is a very easy-to-live-with equity curve. Notice that other than one brief period in 2009, the max drawdown has only been -13%, and post-2009, it averaged only about -8%. As a comparison, in a backtest of the same portfolio today without using market timing, I would have only received a return of 28% and would have had a max drawdown of -32%.

Note: I do NOT use signals from the simple technical chart I used as an example to start this thread for my newsletter’s portfolio timing. I use a more sophisticated, multi-faceted approach that runs on a weekly basis. There are more frequent signals, but they are timely and do not lag the market. The latest signal was on June 28, and allowed my and subscribers to miss the entire late-summer downturn. Even with signals from market timing, annual turnover is still less than 300%.

As much as Marc Gerstein seems to wish to dissuade this, there’s nothing wrong with supplementing your fundamental work on p123 with learning TA by using a site like StockCharts or TradeStation. But you won’t find a better platform for stock analysis than p123. Back in the early 2000’s, I used to pay something like $90k/yr for a Capital IQ product and for the typical investor, P123 is far superior. While a site like TradeStation has 55 years of data and p123 has 16, TradeStation is wholly focused on technical analysis. They have very few fundamental capabilities. Newcomers to p123 should be thrilled with the capabilities and dataset on p123. Heck, when I first joined p123 in 2004, there was only 3 years of data to use. Marco and the crew are doing a fantastic job!

One last note: I am posting this simply to advocate for members to make use of the market timing capabilities within the p123 platform. I am not seeking new subscribers. However, you are welcome to register for my Value Alert newsletters which are now published for free on an ad-hoc basis for as long as I enjoy writing and wish to continue the site. You can also click the ‘Value Alert Newsletters’ link at the top of the home page to access an index of back issues.

I hope this post inspires some people who eschew technical analysis to reconsider it again. Significant market-related losses are the fear of every investor. You are torn between, 'Is this an opportunity to buy on the dip – or just the beginning of hellacious losses if I stay in?" Sound market timing techniques can answer this question for you, enormously enhance your wealth, and provide peace of mind.

I will advocate again to those that have to ability to affect new development here at P123 that one of the best possible items to work on is multiple time frame analysis. Adding the ability to run weekly and monthly analysis like StockCharts on items like MA, MACD, RSI, etc. would be quite useful.

In addition, being able to run tues-tues, wed-wed, thurs-thurs, fri-fri weekly hold times would likely affect a large number of the P123 constituency.

For both these items, the data is all there.

Best,
Ro

All:

 I have found TA very useful over the years, and managed to avoid the damage of 87, 00, 08 and most of 15.  A simple statement of fact: Ya all are quick to attack Chris.  Why not make him feel welcome? Just saying....

Bill

Bill, I agree!

I say attack away, if you want, but do it with information, facts, numbers and sim results-- we will probably learn something. However, arguments along the lines of “this is true because I’m so knowledgeable” are causing some to stop posting, I think. I’ll let them speak for themselves on this (if they post again).

Without a difference of opinion we would not have a market. Or worse: without bad opinions, we would have an efficient market: how boring would that be?. A difference of opinion is a good thing.

There is no doubt that even the most basic market timing system will provide better returns than buy-and-hold. Here is a very simple example which only required the investor to make 13 asset allocation changes over the last 15 years. That is not trading.

I employ the MAC-US system (a modified golden cross system previously discussed at length on this forum) to vary asset allocation according to market climate. Specifically my System1a uses the Vanguard’s Total Bond Market Index Fund and Total Stock Market Index Fund in combination, echoing the broad based Index approach to investing. Taking an asset allocation of 60%Stocks and 40%Bonds during up-stock-market periods and reversing the allocation during down-market periods to 40%Stocks and 60%Bonds provides much better return with lower drawdowns than a constant 60%Stocks and 40%Bonds allocation like that of the Vanguard LifeStrategy Moderate Growth Fund (VSMGX), as can be seen in the table below which also shows performance for other percentages.

What could be simpler than this system? Yet all the Fund Companies advocate that market timing does not work and advise their participants accordingly.


Chris,

I don’t know you and I don;t know your system. But I do know enough to feel a deep sense of revulsion when anybody throws around phrases like 100% accuracy and easy on anything related to the stock market, and especially in an area that has perplexed minds far greater than mine. (I’m particularly struck by the cute way you craft your statements such that “100%” appears next to “over 55 years;” a construction obviously designed as a response in case the SEC or FINRA come around challenging you for guaranteeing that past performance determines future outcomes. OK. Cool. So legally you can deny doing that.

Now let’s move to the court of public opinion, and let’s make it put up or shut up.

Let’s come up with an arrangement in which you set up a simple market timed R2G (in SPY or Cash) but one in which you specify, up front and with full legally enforceable specify, the declines your model will prevent subscribers from experiencing and your agreement to reimburse subscribers, in full, for any non-compliant losses they suffer – with, of course, your obligation secured by the posting of bond. (I’m not talking about refunding subscription prices; I’m talking about reimbursing for losses that may be incurred contrary to what you say your model can prevent.)

I’ve never made this sort of challenge before because I’ve never encountered anybody who insists with such vehemence that 100% success is so easy to achieve. But since you are insisting on this, then let’s do it. Let’s figure out the logistics of the details of your guarantee, the establishment of a bond/escrow arrangement, any potential ceilings on numbers of subscribers and amount invested per subscriber, and a method of verifying any reimbursable losses subscribers might experience.

Chris, could you please provide us the following info about your system.
What is st dev for your port? Because if ret is about 90% and sharpe 1.5 it means that st dev (on annual basis I guess) should be around 60%.

How many tickers in it and what’s the turnover (not market timing turnover)? It looks like slippage is close to zero and some kind of optimization exists (if market lets say rose 15% annually excluding drawdown periods, your system rose at least 6 times faster than market then, how is that possible). What does it show from 1999 to 2009? And what happened starting july 2014?