It starts with the white bull flags. The 72 degree trend angle (yellow) is nearly identical to the current run's trend angle of 70 degrees.
There were two distinct peaks created during the bull run followed by a minor correction then a continuation candle (circled in white and red). Once again there's been a identical pattern in our current run.
In 2017 there were two bearish engulfing candle formations (circled in pink), one above and one below the -81 degree trend angle (orange). The same has occurred during this bull run.
Both runs have a hammer candle reversal under the same down trend angle (circled in teal).
The current baseline support in green has a 14 degree angle. The down trend resistance line from the 2017 peak till now is -16 degrees, well within the symmetrical structure which indicates that it's going to take strong buying to break above it if it even gets retested which won't happen anytime soon if this run follows the same pattern in 2017/2018.
One thing is obvious: a correction is coming.
The question: how much?
I linked my chart on Bulkowski's Bump and Run Reversal as reference for the criteria he uses for the support/reversal zones. I moved those zones over to this chart as that pattern is still valid since the recent failed breakout above $12.5k. In 2018 the lowest low from peak bottomed out around the red circle. If the same holds true here, BTC would drop into the $7,500 range which is in the 14% chance reversal zone and right at the 0.382 Fib support.
*The percentage chance of reversals are valid. Just because BTC bottomed out in a certain way then doesn't mean it will now!!!*
It's important to note the key trend line (dashed purple). It acted as a long standing support during the 2017/2018 bull run and subsequent down hill run. Will it do the same here? Time will tell.
Next question: when?
My best estimate is within the next 30 days BTC if bottoms out similar to how it did in 2017/2018. Given the compressed nature of this pattern in comparison to that in 2017/2018, 2 weeks is a more approximate estimate.