Blay Timing Indicator
Two environments are constructive to higher stock prices and can be objectively measured, thereby lending themselves to be integrated into a timing indicator. The first one is the trend of margin debt. The second one is the relative strength of “aggressive” sectors (i.e., the S&P500) v. defensive sectors (i.e., Utilities). There is a high probability of higher prices when either is present.
The trend of margin debt was documented (“Stock Market Logic”, Fosback 1991) to be an excellent timing indicator. Margin debt has not lost its mojo, and my updated calculations show that, to this day, it continues to be an effective tool for forecasting future stock prices. Margin debt (MD) is the ultimate measure of investor participation. Cash in brokers’ accounts may be bullish (money ready to be invested) or bearish (money hiding from stocks). However, MD can only increase if investors are bullish and willing to leverage their existing holdings to buy more.
Conversely, if MD shrinks, it shows that investors are pulling back or, worse, that margin calls are forcing liquidation. My studies show that an expanding MD is associated with higher stock prices in the future. A dwindling MD results in lower prices. The spreadsheet below, which updates Forsback findings, shows what happens when one buys the S&P500 when MD is above its 12-month moving average (MA) and sells it when MD is below its 12-month MA. Conversely, it shows the results of buying when the MD is below its MA and selling when it is above. The test starts on 1/1/78 and finishes on 10/18/22.
The Table below summarises the key performance figures:
Indeed, a rising trend for MD is a tailwind for stock prices. You may access the full performance report for both tests HERE.
From the above example, it is clear that MD can be a good timing indicator. Our BTI does not use a MA to appraise the trend of MD. It uses similar patterns as those applied by the Dow Theory to pinpoint the relevant highs and lows to be pierced to signal a change of trend. As with the Dow Theory, there is a minimum extent requirement for any movement to be meaningful. The spreadsheet below shows the results of buying the S&P500 when the trend of MD is bullish and selling when bearish. It results in a higher net profit and average trade than the 12 months MA. If we go against the trend and buy when MD is bearish and sell when in turns bullish, we hardly make a profit. Once again, there are not many occasions to make money when MD is shrinking.
Interestingly, my Dow Theory-based trend determination of margin debt makes even less money than the 12-month MA when we fight the trend, which shows that it is an even more accurate way of determining the trend (you make more if you go with it, and make less if you go against it). Look also at the lower Profit Factor (P.F.): 1.24 for MD trend based on the DT v. 1.71 when based on a 12-month MA. The lower the P.F. when we go against the trend, the better.
If MD is such an effective timing device, shouldn’t we stop here? No, for two reasons:
1) While favorable to higher stock prices, an upward trend for MD is not a guarantee against sudden downturns. The BTI protects against such drops in a two-fold manner: (a) Firstly, by having trailing stops based on confirmation which change depending on the market condition; (b) secondly, by rejecting an entry when, despite a positive trend for MD, the internals of the market are weak.
2) TIME is essential to making money in the markets. Roughly 34.5% of the time, MD is in a bearish mode. However, during such instances of bearishness, there are spells of market bullishness. The critical issue is to detect when an investable bullish episode is likely to occur within a bearish MD to increase the bottom line. Here comes into play the second element of the BTI.
So, let’s take a look at the second component of the BTI. Relative strength between market sectors can forecast the most probable price action. The less the market favors defensive sectors (i.e., utilities), the more likely higher prices are ahead. When defensive sectors are the strongest, the implications for the whole market are bearish. This fact has been thoroughly researched (“An Intermarket approach to Beta Rotation”, Michael Gayed, 2020). So, even when the trend of MD is bearish, if aggressive sectors are stronger than defensive ones, we’ll get a bullish reading, and it pays off to return to the market.
Relative strength between sectors also helps us to time our exits. When defensive sectors outperform aggressive ones, it is time to tighten our stops.
To calculate signals, the BTI uses the S&P500 (most aggressive), The Dow Industrials, The Dow Transportation, and the Dow Utilities (less aggressive). I could have replaced the S&P500 with the Nasdaq100 but given that the Nasdaq 100’s started on January 31st, 1985, I preferred to start its record at an earlier date. The BTI begins in 1978 because my reliable Dow Utilities data starts that year.
While the BTI’s makeup is proprietary, it is not subject to individual interpretation. As with all good timing indicators (more about good “timing indicators HERE”), the BTI may give a signal every day, as it is based on daily bars. However, also as with all good timing indicators, it triggers seldom, given that the average duration of a BTI trade is 511 days.
Bottom Line: The BTI is a successful addition to the Dow Theory for the 21st Century. While both indicators usually result in close signal dates, they are constructed differently, providing diversification. The BTI is constructed through mathematical calculations of relative strength, momentum, and margin debt trend, whereas the external chart patterns determine the Dow Theory. In my humble opinion, these are the two premier stock market major trend timing indicators with documented long-term records that set the standard for market timing. While we certainly endorse this excellent indicator on a stand-alone basis, we also include it in our COMPOSITE Timing Indicator, which is the synergistic combination of the Dow Theory and the Blay Timing Indicator. Synergy: “The interaction of two or more agents or forces so that their combined effect is greater than the sum of their individual effects”.
Below you have the comparison on an annual basis between the BTI applied to the Dow Industrials v. Buy and Hold (Total return figures). A start equity of $ 10,000 on 12/29/1978 would have become $4,115,580.96 versus only $1,470,741.52 for B&H on 12/31/2022. In that period, the BTI had a CAGR of 14.66%, whereas Buy and Hold merely reached 12.01% with more negative years (8 for B&H v. 3 for the BTI) and deeper drawdowns.
The results are hypothetical and are NOT an indicator of future results and do NOT represent returns that any investor attained
The Table below summarises the key performance figures on the S&P500 (no total return from 1978 to 2021):
You can see that the BTI beats B&H on all counts: much higher Profit Factor, total percentage won/lost, Maximum Losing Year, number of negative years, CAGR, etc.
The spreadsheet below shows all trades and performance of the BTI applied on the Dow Industrials. I thank Subscriber Tom Halgren for producing the total return figures:
| DATE | Signal | Invested | Dow Ind. | BTI return |
|---|---|---|---|---|
| 7/6/1978 | BUY | 100 | 807.19 | 10000 |
| 11/24/1978 | SELL | 0 | 810.12 | 10240 |
| 12/11/1978 | BUY | 100 | 817.65 | 10278 |
| 12/29/1978 | 100 | 805.01 | 10147 | |
| 12/31/1979 | 100 | 838.74 | 11168 | |
| 3/18/1980 | SELL | 0 | 801.62 | 10811 |
| 3/19/1980 | BUY | 100 | 800.94 | 10814 |
| 12/31/1980 | 100 | 963.99 | 13582 | |
| 2/25/1981 | SELL | 0 | 954.4 | 13558 |
| 9/10/1981 | BUY | 100 | 862.44 | 14547 |
| 9/23/1981 | SELL | 0 | 840.94 | 14215 |
| 11/24/1981 | BUY | 100 | 870.24 | 14477 |
| 12/31/1981 | 100 | 874.99 | 14646 | |
| 2/23/1982 | SELL | 0 | 812.98 | 13738 |
| 8/30/1982 | BUY | 100 | 893.3 | 14487 |
| 12/31/1982 | 100 | 1046.54 | 17312 | |
| 12/30/1983 | 100 | 1258.64 | 21762 | |
| 2/23/1984 | SELL | 0 | 1134.63 | 19753 |
| 3/9/1984 | BUY | 100 | 1139.76 | 19817 |
| 12/31/1984 | 100 | 1211.57 | 21858 | |
| 12/31/1985 | 100 | 1546.67 | 29113 | |
| 12/31/1986 | 100 | 1895.95 | 36944 | |
| 10/16/1987 | SELL | 0 | 2246.74 | 44755 |
| 10/19/1987 | BUY (*) | 100 | 1738.74 | 44776 |
| 11/24/1987 | SELL | 0 | 1963.53 | 50733 |
| 12/3/1987 | BUY (1) | 50 | 1776.53 | 50803 |
| 12/31/1987 | 50 | 1938.83 | 53306 | |
| 3/10/1988 | BUY | 100 | 2026.03 | 54959 |
| 12/30/1988 | 100 | 2168.57 | 60520 | |
| 12/29/1989 | 100 | 2753.2 | 79502 | |
| 8/20/1990 | SELL | 0 | 2656.44 | 78645 |
| 8/23/1990 | BUY (1) | 50 | 2483.42 | 78691 |
| 12/31/1990 | 50 | 2633.66 | 82586 | |
| 2/7/1991 | BUY | 100 | 2810.64 | 85761 |
| 12/31/1991 | 100 | 3168.83 | 99390 | |
| 12/31/1992 | 100 | 3301.11 | 106563 | |
| 12/31/1993 | 100 | 3754.09 | 124596 | |
| 12/30/1994 | 100 | 3834.44 | 130578 | |
| 12/29/1995 | 100 | 5117.12 | 178366 | |
| 12/31/1996 | 100 | 6448.27 | 229456 | |
| 12/31/1997 | 100 | 7908.24 | 286121 | |
| 8/28/1998 | SELL | 0 | 8051.68 | 294461 |
| 8/31/1998 | BUY (1) | 50 | 7539.07 | 294573 |
| 9/1/1998 | BUY | 100 | 7827.43 | 300233 |
| 12/31/1998 | 100 | 9181.43 | 354128 | |
| 9/23/1999 | SELL | 0 | 10318.59 | 402544 |
| 10/8/1999 | BUY | 100 | 10649.76 | 403263 |
| 12/31/1999 | 100 | 11497.12 | 436780 | |
| 10/13/2000 | SELL | 0 | 10192.18 | 391676 |
| 12/29/2000 | 0 | 10786.85 | 396428 | |
| 9/20/2001 | BUY (1) | 50 | 8376.21 | 407318 |
| 12/31/2001 | 50 | 10021.5 | 449670 | |
| 7/10/2002 | SELL | 0 | 8813.5 | 426467 |
| 7/19/2002 | BUY (1) | 50 | 8019.26 | 426638 |
| 10/9/2002 | BUY (2) | 75 | 7286.27 | 408846 |
| 12/31/2002 | 75 | 8341.63 | 455254 | |
| 2/7/2003 | BUY | 100 | 7864.23 | 436604 |
| 12/31/2003 | 100 | 10453.92 | 591485 | |
| 12/31/2004 | 100 | 10783.01 | 622482 | |
| 12/30/2005 | 100 | 10717.5 | 633228 | |
| 12/29/2006 | 100 | 12463.15 | 752360 | |
| 11/27/2007 | SELL | 0 | 12958.44 | 797363 |
| 12/31/2007 | 0 | 13264.82 | 799798 | |
| 3/25/2008 | BUY | 100 | 12532.6 | 803282 |
| 5/12/2008 | SELL | 0 | 12876.05 | 827779 |
| 10/7/2008 | BUY (1) | 50 | 9447.11 | 832569 |
| 11/12/2008 | BUY (2) | 75 | 8282.66 | 782486 |
| 12/31/2008 | 75 | 8776.39 | 819965 | |
| 2/19/2009 | SELL | 0 | 7465.95 | 731228 |
| 2/23/2009 | BUY (1) | 50 | 7114.78 | 731244 |
| 3/27/2009 | BUY | 100 | 7776.18 | 766628 |
| 12/31/2009 | 100 | 10428.05 | 1053024 | |
| 6/8/2010 | SELL | 0 | 9939.98 | 1015789 |
| 6/9/2010 | BUY | 100 | 9899.25 | 1015793 |
| 6/23/2010 | SELL | 0 | 10298.44 | 1057822 |
| 7/9/2010 | BUY (3) | 100 | 10198.03 | 1057888 |
| 12/31/2010 | 100 | 11577.51 | 1216585 | |
| 8/9/2011 | SELL (3) | 50 | 11239.77 | 1200002 |
| 12/7/2011 | BUY (3) | 100 | 12196.37 | 1257036 |
| 12/30/2011 | 100 | 12217.56 | 1261473 | |
| 12/31/2012 | 100 | 13104.14 | 1390624 | |
| 12/31/2013 | 100 | 16576.66 | 1802429 | |
| 12/31/2014 | 100 | 17823.07 | 1982591 | |
| 3/4/2015 | SELL (3) | 0 | 18096.9 | 2021449 |
| 3/24/2015 | BUY | 100 | 18011.14 | 2021515 |
| 8/25/2015 | SELL | 0 | 15666.44 | 1777162 |
| 12/31/2015 | 0 | 17425.03 | 1777533 | |
| 8/23/2016 | BUY | 100 | 18547.3 | 1778211 |
| 12/30/2016 | 100 | 19762.6 | 1912831 | |
| 12/29/2017 | 100 | 24719.22 | 2450548 | |
| 10/23/2018 | SELL | 0 | 25191.43 | 2541723 |
| 12/24/2018 | BUY (1) | 50 | 21792.2 | 2551983 |
| 12/31/2018 | 50 | 23327.46 | 2643089 | |
| 4/2/2019 | BUY | 100 | 26179.13 | 2820193 |
| 12/31/2019 | 100 | 28538.44 | 3134895 | |
| 2/28/2020 | SELL | 0 | 25409.36 | 2804515 |
| 3/9/2020 | BUY (1) | 50 | 23851.02 | 2805203 |
| 4/7/2020 | BUY | 100 | 22653.86 | 2738534 |
| 12/31/2020 | 100 | 30606.48 | 3760080 | |
| 12/31/2021 | 100 | 36338.3 | 4547817 | |
| 3/8/2022 | SELL | 0 | 32632.64 | 4097960 |
| 12/30/2022 | 0 | 33,147.25 | 4176080 | |
| 4/20/2023 | BUY | 100 | 33,786.62 | 4234140 |
| 12/29/2023 | 100 | 37,689.62 | 4794584 | |
| 1/4/2024 | SELL | 0 | 37440.34 | 4764392 |
| 1/17/2024 | BUY | 100 | 37266.67 | 4773357 |
| (*) Simultaneous Capitulation and BTI buy signal. Go to 100% invested | (2) Second Capitulation. Bought only 25%. | |||
| (1) First Capitulation. Go to 50% invested | (3) Buy on 7/9/10 sold only 50% on 8/9/11 due to previous Capitulation. Next Buy on 12/7/11 bought only 50% since BTI was already 50% invested. The remaning 50% bought on 7/9/10 not sold on 8/9/11 was sold together with the 50% Buy made on 12/7/11 on 3/4/15 |
NOTE: This is a 14.69% compound annual increase from 12/29/1978 to 12/31/23 v. a 12.09% gain for Buy & Hold over the same period.