Out of the many attempts to create a bitcoin price signal the “NVT Signal” is the most accurate one with 15+ correct data points. It is not leading, rather a coincident indicator, but still very useful for the medium term market swings.
Look at this chart and look up the signal after 2012 (it needed a few years for calibration).
Every time the signal value broke 140, there was a significant correction in the price, in the short to medium term. There were 6 correct data points — Apr’13; Dec ’13; Jun ’14; Dec ’17; May ’18; Jul-Nov ’18 and now we have a 7th “sell” signal.
Once we reach 140+ we don’t know when and how much but we know it is going to fall.
Similar pattern is found near medium term bottoms. Any time the signal goes below 55 it seems like a great opportunity to buy, though it may continue dropping lower in the very short term. Here are 8 points with a signal value lower than 55 which were great buying opportunities.
If we reach signal <55 today (16 months in the bear market) it would very likely mean a long term bottom.
Currently, the signal shows levels similar to Dec 17, which does not justify buying at this point. Even though we broke the 200 days moving average, there is still not enough evidence for a long term trend reversal.
Entering a short position now could be a great speculative call (not investment advice).
BTC targets* for a potential bottom if the signal moves to a “buy” zone.
Based on the 2015 bottom pattern an increase of 1.5x or lower is the most likely scenario.
* targets calculated based on the 90 day average blockchain.info’s estimation for USD onchain volume on the BTC blockchain and NVT signal in the range 40–55. It is $557M as of Apr 10th 2019.
The big question is can the signal be wrong?
Too easy to be true? Here are some potential problems.
Some people claim BTC has become more of a store of value than a medium of exchange, so transaction volume is naturally reduced. Funds could be investing in BTC for the potential of price increase and future usage and hodl without doing any transactions.
However, we had 3 correct NVT signal calls in 2018 so it’s still working despite the above concerns.
Can the volume data be wrong so that the signal is misleading? Onchain volume could be gamed to the upside (potentially) but not to the downside. Someone can make fake large transactions but cannot stop others from transacting. So no one could manipulate the NVT Signal to show a false “sell” alert (level 140+).
Blockchain.info may have changed their algorithms to show artificially lower volumes than before. However, Coinmetrics data shows similar volume pattern here, here and maybe here (NVT signal with “brand new” volume data). So this is also not a good argument.
I have not yet seen a reasonable argument that the NVT Signal is broken. So if it’s not, we better buckle up for a ride down…
Arguments against this thesis are welcome!
Also look at the how permanent onchain volume increase after coming out of the bear in 2015. We are not there yet…