hsi - study of bearish cycles, compare with times of dot-com bubble

For a 'normal' bearish move, it takes about 12 months to complete. An exceptional case within the last 20-30 years is the down move from March 2000, the burst of dot-com bubble. It took 37 months to complete and more precisely it took 18 months to a half-way (in terms of duration) bottom very close to the destination and then took another 19 months to consolidate in a range (8000-12000) before everything headed north.

My concern is:

If we follow a 12-month bearish cycle, the impoverishing game is over and we can sleep well from now on.

However, if the current bearish down move imitates that of the dot-com bubble, we will still have 3 months (Oct-Nov 2016) to see a bottom close enough to the end of bearish market.

Take a closer look at the index from March 2000 to April 2003 versus the current condition (from May 2015 '大時代'). From a naked eye, there are quite many similarities in the two different periods:

1/ a Zig-zac wave A:
2000 - hsi dropped 4232 pts from July 22 to Dec. 2, 2000 which is almost the same as the down move from Feb. 3 to Apr. 14 2001, 4213 pts. Also a AB=CD;

2015 - hsi dropped 8220 pts as wave 'a' and 5145 pts in wave 'c' almost 0.618 of wave 'a'.

2/ wave 'B' about 1/3 of wave 'A'
2000 - wave 'B' (1928 pts) is about 1/3 of wave 'A' (6064 pts)
2015 - wave 'B' (3375 pts) is almost 1/3 of wave 'A' (10310 pts)

3/ wave 'C' about 2.618 times of wave 'B'
2000 - wave 'C' (5095 pts) is about 2.618 times of wave 'B' (1928 pts)
2015 - can we expect 12818 (21654 - 3375 x 2.618)? if it is 1.618, it would be 16193; 2 times, 14904

Lastly, I hope the historical supporting trend line (the same blue line in both charts) are still valid (provided an extension less than or equal to 2 times of point 3/), so my job in future is easier just to buy near the line and hold for many years....

(July 26, 2016: some violation of Elliot wave count principles ... anyway, just for reference)