This post is a follow up to the analysis of quarterly chart RSI peaks from a few weeks ago. Today I'm looking at the monthly version.

The chart below goes back to 1915. Yes I know it is hard to see the price. I'll zoom into the specific incidences soon enough. The point is that RSI is in historically rare territory and we want to examine what has happened in the past when it has been up at these levels.

1925
1928-29
1945
1955
1959
1986-87
1996-97
2017

1925-29
The first of these was a momentum high that was enough to pause the market for several months. The index went fractionally above but did not close higher; and a -16% drop to near support followed. After that, parabolic move began.

There was a cluster of months in 1928 when the monthly RSI was at today's level or higher; 3 months powered higher, then a 5 month sideways pullback, before a zoom to the final top. There was a -13% decline before the real trouble began. I think something similar 'should' be the case today. There was some divergence on the final high via Bollinger bands but RSI was not glaring and the decline was rather sudden, crashing -49% in just 3 months.

Note the later highs in 1936 reached 80 but not 82+ of today.

1945
Similar to 1925, the first high was not 'it' but a few months later a bit higher high in price with notably lower high in RSI was followed by a -25% drop over 6 months that then had to stabilize for more than 2.5 years before another rally took off.

1955 & 1959
This was the nicest RSI peaks. 1955 went steadily higher for another year before putting in a clear RSI divergence high shown at the slanted red arrows.

1959 had a high retest before a drop of -17% into 1960.

Note the high that really mattered in 1966 happened with much lower RSI.

1986-87
Like the late 20s, 1986 had a series of months above current RSI level but most of the gains were in before a sideways period & pullback, then another massive trust to the top. That red bar in 9/1986 was a -10% pullback, something we have not seen in 20 months!

1987 really did come out of the blue and while indexes faded back inside the BB I would not have guessed that a single 1 day drop of -20% was on the way.

1996-97
This was an amazing move as the first RSI above 82 was just 1 year into the run! While I doubt we are in the same scenario i could be wrong. There was a lengthy topping process with all kinds of crystal clear divergence on the 2000 top. But now that everyone is looking for a top like this it may not play out in the same way.

2017
What is the next move? While this could go higher by a few or several bars, the more likely move according to history is a slowing of the pace, some pullback, then a higher high or at least a high test with RSI divergence. No divergence means no worry and no premature top picking just yet, even though i do think YR2 is the place for some pause this quarter.

The only thing that bothers me with this idea is that now everyone is looking for a topping process like 2000 or 2007. Who knows, the market may be kind enough to give us a year of stall before a truly painful drop, but at the same time, given algos that are now in charge, we may see something rather sudden at some point down the line.

The point with the two red horizontal lines is that the post election move was not really about Trump but the blast off above the 2015 highs which had stalled the market for quite some time and had started to look like resistance. From that breakout a massive one way move has followed.

In this post I am going to explore RSI peaks on the quarterly charts for INDU, or Dow Industrials. TradingView data goes back to 1914 with RSIs kicking in 1918 on the quarterly chart. Red line is drawn at the current level.

Sum
Both 1929 and 1987 tops were followed by violent drops that came rather suddenly, with little warning or time. However, both moves had rallied much more excessively than today.

1954 RSI peak didn't have a high that really mattered until 1961, and even that after a drop was followed by another rally to higher highs in 1965. Similarly, the 1999 top gave plenty of warnings through RSI divergence and a nice long period to see distribution to get out.

Sudden top and sharp drop, or gradual slowdown in momentum and time before rolling over? So far we don't have the excesses of 1929 or 1987 in RSI terms, nor do we have any slowdown or divergence in RSI. We only have four examples but I think the conclusion is to remain bullish until we see more excess (RSI 85+) or RSI slowdown & divergence.

Quarterly RSI values have exceeded the current level in:

1927-29
1954-56
1986-87
1995-99

3 out of 4 of these are rather glaring tops at the end, so let's explore them all in more detail.

1927-29
This amazing run accelerated with the move above the 1925-26 sideways period in 1927. From 1927 Q3, 7 of 9 bars closed outside the Bollinger Band (BB) before the massive drop. RSI reached a peak of 91 on the top.

The current environment while impressive is nowhere near the excesses of 1927-29. But let's file that for other indexes - Q chart RSI 90 is a potential sell.

That said when 1929 came apart it was in a massive way - if i see a candle like 1929 Q3, i will be OUT and probably short.

1954-56
This is a lot more comforting because even though the RSI peak was 1956 Q1 that was just the first in a series of higher highs. There was a sideways period and pullback to 1957 Q4, anther rally to 1959 and 1961 highs, and sharper drop to 1962 lows, then another strong rally to 1965-66 highs. These were the highs that mattered as ~15 years of sideways followed.

If only the current environment would play out like this... we can sleep comfortably until we see a noticeable divergence top, ideally both on RSIs and Bollinger bands such as 1961 Q4 with RSI merely 71 and the highs all inside the BB compared to prior two peaks. These are highlighted by the red arrows on price and RSI in the chart below.

But it is not inconceivable that the current move could end like 1962 - a sudden drop out of the blue.

1986-87
The power move really began in 1985 Q2, with the jump above the prior 1983 highs, but RSI was still only 68 then. From 1985 Q4 6 bars out of 8 closed outside the Q BB before the massive drop.

The worrisome point about this chart is no warning like 1929 or 1961; out of the blue, higher highs in RSI, a mild wick but nothing that screamed sell.

1995-99
Q chart RSI was an amazingly above 90 from 1996 Q2 to 1998 Q2, before a sudden shakeout during 1998. Also note how many quarterly bars were closing outside the BB - a run of 10 in a row from 1995 Q2 to 1997 Q3, a pause, then one more in 1998 Q1.

I like the 2000 top, and probably everyone does. Crystal clear divergence. A nice long time to get out. Plenty of warnings and time.

2017