Author

Topic: Are logarithmic scales useless in deciding what bitcoins price will do? (Read 283 times)

full member
Activity: 322
Merit: 114
Log scales definitely aren't useless and should be considered. It helps to give a clearer long term picture and remove some of the stochasticity. But as others have said, all charts will eventually take a different route. Any linear or non-linear progression will be wrong in time because markets are not predictable. Common sense also needs to prevail. That chart would have bitcoin at over $1m in around 10 years time and much higher inside the next 20 years

Before getting to the point of logarithm, should be worth considering first if TA is worthless or not. If you're like me and enjoy TA purely for entertainment value, then for sure, a lot of arguments for, and even I'd say for certain linear methods. But if you're seriously into TA and use it to make decisions, then be wary of Bitcoin's caveat: all moves can be explained, but only after the moves happen. All bets are off from this point forward.

@jackg: somewhat agree, but there are many exceptions, perhaps not among news writers but commentators. Lifelong Bitcoin educator Antonoupolos for example never had Bitcoin until he was donated some but has never been a pessimist.

True, TA should only ever be used as an indicator. It's best used in conjunction with a multitude of other things, the main one being common sense and understanding. The crypto market is a wild and unpredictable place and over time will make a mockery of any TA.
legendary
Activity: 2968
Merit: 3684
Join the world-leading crypto sportsbook NOW!
Log scales definitely aren't useless and should be considered. It helps to give a clearer long term picture and remove some of the stochasticity. But as others have said, all charts will eventually take a different route. Any linear or non-linear progression will be wrong in time because markets are not predictable. Common sense also needs to prevail. That chart would have bitcoin at over $1m in around 10 years time and much higher inside the next 20 years

Before getting to the point of logarithm, should be worth considering first if TA is worthless or not. If you're like me and enjoy TA purely for entertainment value, then for sure, a lot of arguments for, and even I'd say for certain linear methods. But if you're seriously into TA and use it to make decisions, then be wary of Bitcoin's caveat: all moves can be explained, but only after the moves happen. All bets are off from this point forward.

@jackg: somewhat agree, but there are many exceptions, perhaps not among news writers but commentators. Lifelong Bitcoin educator Antonoupolos for example never had Bitcoin until he was donated some but has never been a pessimist.
full member
Activity: 322
Merit: 114
Log scales definitely aren't useless and should be considered. It helps to give a clearer long term picture and remove some of the stochasticity. But as others have said, all charts will eventually take a different route. Any linear or non-linear progression will be wrong in time because markets are not predictable. Common sense also needs to prevail. That chart would have bitcoin at over $1m in around 10 years time and much higher inside the next 20 years
legendary
Activity: 1372
Merit: 1252
@cellard that is true there are two types of people,who write news articles:
1. Ones with bitcoin
2. Ones who don’t have any

The first are biased and optimistic in bitcoin,
The second are biased and pessimistic as they want bitcoin to fall so they can buy it on the cheap or so they can make themselves happy they never bought any to begin with.

News articles are bad if you’re a long term investor. Do like the people who bury their private keys for long term holdings, so you won’t be tempted.

There is also a third option: People that are getting paid to either shit on bitcoin and crypto in general or to pump it, so they aren't even having an opinion based on what they are holding.
Im sure there is people on the press getting paid in fiat to talk trash on bitcoin and call it a tool for criminals while they secretly are holding coins.

The point is, news are useful, for trading, but not that relevant for value investing. Also if you do go the "bury the private keys" route, be sure to remember where you store them as well as any passwords. I lost the passwords of encrypted hard drives myself and now they are useless.
copper member
Activity: 2856
Merit: 3071
https://bit.ly/387FXHi lightning theory
@cellard that is true there are two types of people,who write news articles:
1. Ones with bitcoin
2. Ones who don’t have any

The first are biased and optimistic in bitcoin,
The second are biased and pessimistic as they want bitcoin to fall so they can buy it on the cheap or so they can make themselves happy they never bought any to begin with.

News articles are bad if you’re a long term investor. Do like the people who bury their private keys for long term holdings, so you won’t be tempted.
full member
Activity: 715
Merit: 220
It depends on which time frame you are trading.
If you trade intraday the log scale is not that useful.
But the two scales(linear and log) remain valid because traders use both of them to perform there analysis and therefore the two scales will allow you to see different pattern formation.
But if you are a long-term trader, the log scale is mandatory because it will allow you to see old btc price mouvements when bitcoin was less than 1000 $ and to define long-term trends. If you use linear scale btc mouvement price below 1000 $ will be on the noise.
legendary
Activity: 2898
Merit: 1823
OP, logarithmic charts are good for "smoothing out" the scale and to do away with some of the outliers in the data that make the chart look unnecessarily volatile for viewing the long term.

But in predicting what the price will do, any chart can't help you if you're not a good trader, while a good trader can use any chart to make good calculated estimates.
legendary
Activity: 1372
Merit: 1252
I think linear scales are useful to get a feel for the activities of the whales and the pumpers. Wicks combined with volume charts are another useful indicator. For long term investment, I think you can forget the charts, look at the news, and try to judge investment opinion.

I wouldn't even look at the news for long term investment, specially on an asset like bitcoin. There's endless bullshit on the press, and increasing at exponential rate now that bitcoin has grown. We have different media releasing different news about the same thing just like in the old world.

I would just focus on the tech and economic reality of bitcoin. My conclusion is that bitcoin is hard money, a better gold ready for the information age that has survived for 10 years with no downtime or failed tx's. The fundamentals are good and therefore there's no reason to not hold irrespective of what press is saying (including governments).
STT
legendary
Activity: 4102
Merit: 1454
LOG scales to me is about the very long term charts which for Bitcoin is only a few years or more.     I have to admit I really dont use that type at all, my long term is at most about 18 months and I dont look further back then that for a trend.   

All my charts are linear and since it seems to match price finding some ceiling or pause in its movement, I presume its got some accuracy.   Also I think the general population would be using linear charts, the more people follow an indicator or trend the more significance it can have short term as people place orders based on that perception.   Same for Simple moving average over Exponential moving average, I think more people are tracking 200 day simple MA and 50 day over a more complicated indicator and it can become self confirming which is a part of the short to medium term.
   Extrapolating from the 2013 peak to now I find hard to do with confidence, so log scale might make more sense
copper member
Activity: 2856
Merit: 3071
https://bit.ly/387FXHi lightning theory
Yes, I guess whatever happens from now the line will become useless.

The line predicts that bitcoin will continue at its current rate as far as I could see. With a currency so weird, I wouldn’t be surprised if it goes to a filament lamp shape. Probably correlating to f(£e^|x|] although it’s been a while since I’ve done transformations so...
legendary
Activity: 1806
Merit: 1521
Using logarithmic scale is about viewing price movements in percentage terms. It means displaying the distance between price movements (the y-axis) in percentage terms by scaling the y-axis in base-10.

And there’s always one..
Not in my system, I’m talking about logging the price not using logged axis.

I guess I misinterpreted your intent. Still, I don't see how this answers the question of whether "logarithmic scales" are useless.

My mere question is one you repeated. Why? It gives a nice curve. The question now is what does it show (kinda) and is there antguing that can come from that result. I think personally that the curve is useless.

Yes, I assume the curve will just fail eventually when the trajectory of average price growth changes. Every trend eventually fails on a long enough timeline. It's the same with drawing a simple trend line or doing any kind of linear regression.
copper member
Activity: 2856
Merit: 3071
https://bit.ly/387FXHi lightning theory
If you log the bitcoin price 3 times (at base 10) you end up with a very nice curve (using daily market data I got from xe which covered 2011 until 2018). I’m thinking therefore the log function might be useless in predicting bitcoin’s price because of this

The premise doesn't lead to your conclusion. Why would you apply a log function to the price? You're supposed to apply it to the price scale.

Using logarithmic scale is about viewing price movements in percentage terms. It means displaying the distance between price movements (the y-axis) in percentage terms by scaling the y-axis in base-10.

And there’s always one..
Not in my system, I’m talking about logging the price not using logged axis.
My mere question is one you repeated. Why? It gives a nice curve. The question now is what does it show (kinda) and is there antguing that can come from that result. I think personally that the curve is useless.
legendary
Activity: 1806
Merit: 1521
If you log the bitcoin price 3 times (at base 10) you end up with a very nice curve (using daily market data I got from xe which covered 2011 until 2018). I’m thinking therefore the log function might be useless in predicting bitcoin’s price because of this

The premise doesn't lead to your conclusion. Why would you apply a log function to the price? You're supposed to apply it to the price scale.

Using logarithmic scale is about viewing price movements in percentage terms. It means displaying the distance between price movements (the y-axis) in percentage terms by scaling the y-axis in base-10.
legendary
Activity: 2814
Merit: 2472
https://JetCash.com
I think linear scales are useful to get a feel for the activities of the whales and the pumpers. Wicks combined with volume charts are another useful indicator. For long term investment, I think you can forget the charts, look at the news, and try to judge investment opinion.
legendary
Activity: 1372
Merit: 1252
All competent technical analyists are using logarithmic charts these days. It's simply impossible to keep monitoring Bitcoin with a linear scale chart, you lose the sense of movement because of the huge ups and downs, your perspective is tricked like that. I still use the linear chart sometimes because it gives me a nice nostalgic feel of when the marketcap was smaller. I sometimes study the charts like that and look back at the first peaks and try to imagine how the people that was around back then would have reacted.

The strongest TA indicator nowadays for me is this channel (log chart of course)

copper member
Activity: 2856
Merit: 3071
https://bit.ly/387FXHi lightning theory
I’m wondering how useful/useless logarithmic scales actually are to predict the price of bitcoin.

If you log the bitcoin price 3 times (at base 10) you end up with a very nice curve (using daily market data I got from xe which covered 2011 until 2018). I’m thinking therefore the log function might be useless in predicting bitcoin’s price because of this and was wondering of other people’s opinions.(I don’t think I’ve made a thread on this before but I might be wrong).
Jump to: