IBDs’s Mike Webster has piqued my curiosity for use of relative strength by his system of “Quick, Quicksand, and the Grateful Dead." Webster employs a combination of 21-day and 50-day simple moving averages of relative strength (RS) to trigger buy and sell signals. Remember, this is relative strength (Ticker/Index*100). This approach provides a mechanical approach to managing positions, which is particularly useful for traders such as I, who find selling to be a challenging aspect of trading. By relying on these moving averages, Webster can systematically add to or trim his positions based on the relative strength of the stock.
To discuss RS in general for a minute, my main use to date has been a takeoff from Tom Bowley using relative strength charting in a layout that uses the relative strength of a ticker compared to major indices (such as the S&P 500), its sector group, industrial group and also the industry group to the broader industry group compared to the S&P 500. Below is an example of such. Note, I use a 5-day simple moving average (SMA) to reduce noise and smooth it out. It’s a very visual approach to help confirm trends. Grayson Roze also uses a similar approach to relative strength but interestingly uses the Vanguard Total Market ETF as a broader measuring stick.
Stan Weinstein uses another relative strength approach as a key component to his stock selection. He likes to use the Mansfield relative strength in weekly charts only and using a 52 period. His methodology requires a breakout from a base to be accompanied by rising relative strength in the 52 period weekly chart. Mansfield RS is a twist on the standard RS calculation of: Mansfields RS starts with calculating the regular RS, adds a moving average and manipulates the output to create a chart that is easy to immediately see changes.
Mansfield RS = ((RS/maRS)-1)*100
In the chart below, there are three lower indicators, the top is a standard RS line with 3 moving averages (21, 34, 55). The middle is a Mansfield RS (period is 55). The zero line is actually the level where the RS crosses the maRS (moving average RS), in this case 55. Painting the plot with clouds makes it clear to see the action.
The lowest indicator is a Mansfield RS showing three different moving averages, 21, 34 and 55. Threw that in just to give you a comparison of the moving average variation.
So, back to Mike Webster. His approach is to use two moving averages of the RS, all of them simple MA’s. For daily charts, he likes 21 and 50, for weekly, 8 and 21, and for monthly, 10 and 24. Just like in regular charts, in an uptrend, the RS runs above the moving averages in an uptrend. His approach is to watch for when the RS drops and crosses the first 21 sma of the RS. Not the price chart sma, but the relative strength graph average. When it hits he sells some of his position. He refers to that drawing a line in the sand (hence the first part of the name “Sand, Quicksand and Grateful Dead”). I have listened to multiple discussions of his on this and the recommended amount is variable from 1/10th to 1/3. If it drops lower (not a specific level, just moving below the 21 and above the 50 RS lines) he recommends “cutting more.” This is the shakyarea or quicksand. Once it hits and crosses the 50, he advises selling more to selling all. He refers to O’Neils saying there’s the “quick and the dead” in the market. The reverse is also true, he buys back as the RS recovers and advances above it’s moving averages. How he integrates this with Market School and counting days, I don’t have a clue. But it certainly is easier. Here’s one of the better available discussions about his process.
It’s not sliced bread, but it’s intriguing. I think it could be a very helpful tool, but as always, you have to consider the broader market context and the tickers price action. To better tease it apart, I have been attempting to run backtesting in Excel. I will be laying out several studies but need to be clear that I am no genius with Excel and “errors” certainly can plague me, so take it all with your own due diligence. Also, please, please let me know if something looks off to you or especially if you have a better, more efficient way to approach it. I know there are folks who read this who are excel geniuses, please chime in. I’m here to learn. Any other comments in regards to how you use relative strength would be wonderful also.
INITIAL BACKTESTING
While I have created some spreadsheets looking at the full triphase use of the pattern, I think it’s helpful to first look a bit more simplistically at does using this relationship of moving averages within RS yield monetary benefit. There’s a lot of “rules” out there that folks use that provide a lot of emotional support and simplify decision making but don’t necessarily gain a lot of financial traction. So the first worksheet is simply looking at each step as a full trading signal to sell or buy. So, when testing the 21ma, when the RS line crossed from above to then equal or go lower than the 21, the entire position is closed. When the RS line equals or crosses above the 21ma, it is rebought. It’s run in the equivalent of a single share. Profit and loss being simply the difference of the closing prices on those days.
I’ve manipulated Webby’s “rules” a bit to fit an excel worksheet. Webster uses the area between the 21 and 50 curves as his middle signal. I decided to make it a hard signal using a third moving average and chose 34. I used the Fibonacci series of 21, 34 and 55 to calculate the moving averages.
Here’s a taste of the worksheet using QQQ as the test ticker and SPX as the index (chosen for no particular reason):
Running the test from January 2015 to the first part of January this year yielded these core findings:
QQQ was worth 101.62 on 1/20/2015 and was worth 516.7 on 1/15/2025. Hence the comparison increase of 415.08.
Transactions for purchase or selling QQQ were based on the closing price the day of crossover RS line and it’s corresponding moving average. Yes, that may not be always possible to do in the real world, but I was mainly looking at if this approach is worth pursuing. There is no perfect backtesting system out there. They all provide estimates and probabilities, not absolutes.
I can draw several conclusions from this study.
- Yes, there is merit and you do save (and therefore earn) monies otherwise lost to drawdowns.
- As expected, more frequent trades occur with the lower moving average.
- A little surprised that whipsaws were the same in percentage but obviously, more in the more frequent trading lower moving average. But that translates into quick turn-arounds in all trades.
- Whipsaws were pretty equal in both directions.
FULL TRIPHASE BACKTESTING
So clearly this was worth attempting some backtesting on the full methodology of three-steps of buy-sell signals. Had to again make some rules that Webster doesn’t use specifically but needed to for formula creation.
- Using the Fibonacci series of 21, 34 and 55 for generating the moving averages.
- Sell amounts at each level were set at:
a. 21ma-buy and sell 15% of position
b. 34ma-buy and sell 35% of position
c. 55ma-buy and sell 50% of position
To run this required creation of a more dynamic worksheet that functions off of an initial starting cash position of $10000 and a running total of shares owned and cash balance is used. For me, not a simplistic process and I used ChatGPT AI to help create. It’s suggestions were not always correct. Can’t say it was always a friend to me (just as AI has not been a friend to the market today……).
I am still trying to double check and verify the worksheet, although, preliminarily it seems to function correctly. I plan to share it for comments and thoughts in the next couple days. But spoiler alert, I can tell you that using the triphase system does indeed provide added earnings over ignoring the signal. The level, however, is only around 40% gain, less than the individual full signals each. Plus the number of signals and actions is a bit dizzying. Suspect I can vary that especially by flipping the amounts for the transactions with the 21ma having the largest percentage action. But only time will tell, so long as I stay interested and grind through things.
For now, I am setting up an indicator and alert system in TOS that will let me know if positions are approaching RS crossovers. Will likely use the 21 as a signal and react to a 34 move.
Would love any comments, suggestions especially if there is something out there that you all are doing along this vein.
Happy hunting,
Lakedog
