CEX shenanigans: Coinbase has 0.6% maker/1.2% taker fees on "Advanced", however it is IMPOSSIBLE to be a maker because there are no price spots available in the order book within a few %...i guess Coinbase traders fill them all and then say: "see, your order was a taker"..and zap you with a 1.2% fee.
If, alternatively, you try Coinbase one, which allegedly has no fees, then they claim that they still charge a spread fee, but then they say that the spread is much bigger than on advance (you can't use coinbase One on Advanced)...and your fee is STILL above 1%, effectively.
What a racket!
Disclaimer: I've never used coinbase, so I'm just speculating. As I understand it, if your order is out of the money and isn't matching a standing orderbook entry (so it doesn't get filled), it goes on top of some existing (maker) order to increase the amount available at that price. So the pre-existing amount goes first, then yours. With maker fees. Or do they have fixed slots with a maximum amount per slot? This doesn't seem reasonable. Coinbase has post-only. Select post only so your order doesn't get snatched up. Post only means maker only.
|
|
|
 I’ve posted about the blue “triangle oscillator” previously. Its highs correspond to highs on the btc price chart. For a long time I’ve also been trying to come up with an algo that identifies the transition from bear to bull. Finally did it this morning, and it was dead simple: the red trace equals -1 times the half cycle moving average of the triangle wave oscillator. In this backtest, a vertical bar marks each transition signal, which comes at the oscillator minimum. A local minimum marks the dip in the 2021 double top too. Not that I think I could have traded that signal. Clear in hindsight but it would have been confusing at the time.
|
|
|
End stage bitcoin bull market Q3 2025 In the chart above, the Triangle Oscillator is nearing a top. If past is prologue, the bull market's end is approaching. As the first chart's horizontal labeling should make clear, this algorithm indexes the price by block height, not date. The chart below re-indexes per the calendar (while rescaling the oscillator vertically to make the chart more readable).  The extrapolation (red) suggests that the present boom, such as it is, will lose impetus around the end of August. One could reasonably expect a subsequent decline in bitcoin's exchange rate. It's also possible, as a result of declining volatility, that the halvings' influence on price is now so inconsequential that bitcoin's future cyclic declines won't even cancel the natural long term monotonic growth of bitcoin's price, in which case we might simply see many months of choppy sideways price action instead of real bear markets. Nothing's for sure, as you can tell from looking at 2021. The turn signal came at the first top, then exceeded by a subsequent top, though not by much. Looking back, the way bitcoin's price action rocketed upward in Q4 2020 hinted at something unusual in the offing. This price action, as reflected in the oscillator, wasn't in keeping with expected cyclic behavior (see previous posts for why the oscillator is expected to show a triangle wave in response to regular cyclic price action). This next chart compares the vagaries of bitcoin's cyclic price action to an ideal triangle wave.  The chart below shows the phase relationship. By the third halving, bitcoin had got well out of phase and responded with some wild swings between block 650000 and 700000. Eventually she settled down. The price is now in phase with Q4 2015, almost like we've come home. 
|
|
|
 In the rainbow channel, the black oscillator is a pale blue under-layer. Where is 'black", sorry? I see red, light blue, dark blue, yellow, green and purple. Is 'purple' actually black? Must be one of those color "illusions". more importantly...where we are heading, then? It's the one you're calling purple. The oscillator at the bottom of the chart. I called it black, though it's actually sort of dark brown, at least to my eyes. Everything in my charts seems to imply that BTC will go up. But I don't know. All models are wrong, as the saying goes, but some are useful.
|
|
|
 In the rainbow channel, the black oscillator is a pale blue under-layer.
|
|
|
I know...bitcoin aims to kill us all with all this turbulence (heart attacks and and failed inheritance transfers), then there will be, like, only Saylor's coins in existence, and then, when he is gone and if he does what he heavily hinted at, there will be a couple of bitcoins for everybody, ech at , maybe $100 tril a coin, lol.
100 trillion usd... that you, PlanB?
|
|
|
[deleted old chart]
This oscillator uses trig functions and moving averages of 210,000 blocks, which makes it top once and bottom once per halving. The next top may be five months away or so unless bitcoin misbehaves, which it usually seems to do sooner or later (just look at 2021). I scaled the oscillator arbitrarily and superimposed it on the price chart.
Hey man!  Did you do do this for before 2017 too? Or does if not work there for some reason? Every point on the oscillator requires six years of data to compute (actually 315,000 blocks, but whatever). My dataset starts in July 2010 with the Mt Gox price feed. So the oscillator starts in 2016, and there's no way to chart any earlier. I could have started the chart in 2016 but there's nothing much to look at there. No top. Here's a chart I hope is less confusing than the previous one. 
|
|
|
If the image looks confusing at first glance, it simply shows bitcoin's price history in green with USD on the vertical axis. Then an oscillator is superimposed, at arbitrary scale and position so as to show its correspondence with the price to best effect, in blue. The oscillator uses trig functions and moving means of 210,000 blocks, which makes it top once and bottom once per halving. Oscillator input comprises btc daily closes and block timestamps. The oscillator has not reached the top of the current cycle. Maybe I'll get lucky and the oscillator will track the price movement like in 2017, but maybe we'll get some crazy stuff like in 2021.
|
|
|
 This oscillator uses trig functions and moving averages of 210,000 blocks, which makes it top once and bottom once per halving. The next top may be five months away or so unless bitcoin misbehaves, which it usually seems to do sooner or later (just look at 2021). I scaled the oscillator arbitrarily and superimposed it on the price chart.
|
|
|
Sine (red) with zero crossings at the halvings, superimposed on a rolling power regression (yellow)  Pretty nice, taking into account phase jitter There's gold in them thar hills signal under all the noise
|
|
|
Chart 1 shows the back-test of the Redneck Fourier indicator for cycle tops.  The log of bitcoin's daily closing price appears in blue, indexed by block height on the horizontal axis. Major grids mark the halvings, which occur every 210,000 blocks. Minor grids on quarter cycle, or about one year. Following the Redneck Fourier indicator, negative zero crossings indicate tops in bitcoin's halving price cycle. The idea in this chart is, wait for the negative zero crossing then sell or go short. But the RF can also be used to forecast. Choose a rate of increase and draw a straight line, like this  The Redneck shows about when and where the market top is likely to occur, under conditions where bitcoin's price appreciates on that slope. Mark that point where the yellow and blue lines intersect on the graph. Repeat the exercise for different slopes and join the points in a line. Redneck thinks the price is going to bounce off that ceiling.  I performed a back-test of the prognostication function at block 630,000, plugging bitcoin's price history up to that point (dark blue) into the Redneck. The black line shows the Redneck's predicted ceiling. Pale blue shows what actually happened.  I ran the test again at block 650,000 
|
|
|
 Horizontal axis: block height. Red in positive territory: bull market. Zero crossings (marked green and yellow) indicate begin and end of bull. wordpress.com/post/hardworkandlowpay.wordpress.com/852
|
|
|
The next halving (block 840,000) will happen on April 23, 2024. You read it here first.
|
|
|
Bull market when green is above red.  Death crosses where green drops below red indicate major tops. Block height on the horizontal axis.
|
|
|
If anybody's bored and wants something to play with, here's a bash script to estimate the number of days until the next halving #!/bin/bash date --iso-8601 bid=$(curl -s "https://www.bitstamp.net/api/ticker/" | jq -r ".bid") ask=$(curl -s "https://www.bitstamp.net/api/ticker/" | jq -r ".ask") #p0 is midmarket price to 2 decimals P0=$(bc -l <<< "scale=2; ($bid+$ask)/2") echo "mid-market BTCUSD: "$P0 #height=$(curl -s "https://blockchain.info/latestblock" | jq -r ".height") height=$(wget -q -O - "https://blockchain.info/latestblock" | jq -r ".height") time=$(wget -q -O - "https://blockchain.info/latestblock" | jq -r ".time") #May 11, 2020 19:23 halving time 1589225023 #seconds since last halving (sslh) = $time-1589225023 sslh=$(($time-1589225023)) #days since last halving (dsince) = seconds/86400 dsince=$(bc -l <<< "scale=6; $sslh/86400") #projected days of 2020-2024 halving (dproj) = (time-1589225023)/((height/210000-3)*86400) dproj=$(bc -l <<< "scale=6; $dsince/(($height/210000)-3)") echo "projected duration of the current halving:" echo $dproj #rounding procedure #extract decimals, double, add 1 num=$(bc -l <<< "scale=6; 2*($dproj-${dproj%.*})+1") #truncate, subtract 1, divide by 2 num=$(bc -l <<< "scale=6; (${num%.*}-1)/2") #add to original variable dproj=$(bc -l <<< "scale=6; $dproj+$num") #truncate dproj=${dproj%.*} echo $dproj "days"
|
|
|
 That green trace is the bitcoin usd moving geometric mean over 105,000 blocks, one half cycle. chart2 below better illustrates the sinusoidal nature of bitcoin's price cycle as its moving geomean swings between peak (pale blue) and trough (red).  The horizontal axis units are block height divided by 140, so the halvings occur at multiples of 1500. As you can see, the geomean had a zero crossing at 3000 (the second halving). The price action recently is getting compressed in time. The geometric mean had a zero crossing early, before the third halving. And the latest geomean peak appears to have come prematurely  . Maybe people are trying to front-run the halvings.
|
|
|
 Horizontal axis, block height divided by 140. Red, moving geometric mean over 210,000 blocks (one halving). Green, moving geometric mean over 105,000 blocks (half halving). Yellow, their ratio (times 100 to bring it into the frame). Sell signal when the rising half cycle ratio crosses 2; I guess crossing 2 in the other direction indicates buy. Which we are approaching now.
|
|
|
Answered you on the other forum where we hang out.
|
|
|
 Take the 105,000 block moving geometric mean, multiply it by the blue sinusoid which has a period of 210,000 blocks (one halving). You get the red trace. Vertical lines mark halvings.
|
|
|
|