See? Just pay 3 satoshis per byte and wait a day, rather than a hundred+ for half a day...
You always can get cheaper transaction fees if you can wait for 6 hours. Instant transactions with first confirmation within next 1 or 2 blocks eat up your capital because fee rates to use are always at tip of mempool (never be cheap). 
Topic: Bitcoin transaction fees (feer per KB - in USD) with median, outliers, plots (Read 285 times)
October 29, 2020, 10:05:42 PM
October 29, 2020, 08:38:36 PM
Well done. Its about time to debunk this myth of "high fees", the current storm is nothing whatsoever like late 2017 or early 2018.
And the wallets are racing for the top on their own when everything below is empty. This is what happens when people don't pay attention, don't let a wallet ever decide a fee.
Tx feerate 10 min: 280.9 sat/byte
Tx feerate 1 hrs: 266.5 sat/byte
Tx feerate 6 hrs: 164.0 sat/byte
Tx feerate 12 hrs: 105.9 sat/byte
Tx feerate 1 day: 3.0 sat/byte
Tx feerate 3 days: 1.1 sat/byte
See? Just pay 3 satoshis per byte and wait a day, rather than a hundred+ for half a day...
And the wallets are racing for the top on their own when everything below is empty. This is what happens when people don't pay attention, don't let a wallet ever decide a fee.
Tx feerate 10 min: 280.9 sat/byte
Tx feerate 1 hrs: 266.5 sat/byte
Tx feerate 6 hrs: 164.0 sat/byte
Tx feerate 12 hrs: 105.9 sat/byte
Tx feerate 1 day: 3.0 sat/byte
Tx feerate 3 days: 1.1 sat/byte
See? Just pay 3 satoshis per byte and wait a day, rather than a hundred+ for half a day...
October 15, 2020, 01:24:45 AM
Reserved
October 15, 2020, 01:24:25 AM
Reserved
October 14, 2020, 01:52:17 AM
ABSTRACT
Data sources
Plots
For transactions
For median of daily transactions
Last 4 years only (2017 - 2020)
Categorisation of transactions with 2 groups of fee
5 blocks with fee per kB >= $ 200
In 2020
Raw results
Median
Fee_dif
In percent
Bitcoin price and fee difference (in %)
It is overview and next steps, I will move further by looking at smaller rolling windows (weekly median, not periodic median like 1-year or 2-year). Smaller rolling windows can help to see better effects of bitcoin price on median of daily fees and their correlations.
- Make a plan to move your bitcoin when fee is cheap!
- It seems we are in the uptrend with fee in last 3 years (2018 - 2020).
- 2017 is the year with too many high-fee transactions (look at box plots with outliers are displayed in red circles and time series plot for daily median fee)
- Two highest transactions is in 2013 and 2017 with fee per kB in USD at $296.6 and $230, respectively
- 99436 blocks or 15.24 % of total blocks have fee per kB at 2+ USD
- 5 blocks with fee per kB (in USD) >= 200 (1 in 2013 and 4 in 2017)
- In 2020
- Median of transaction is $2.3 per kb, with the interquartile range is $1.2 to $4.9.
- In October: median is $5.1 per kb, with the interquartile range is $3.7 to $7.0. It means most of impatient people are paying more fees (more than 179%) in this month. It is the third highest month in 2020, just behind May and August.
- Fee (reflects in fee difference) usually increases with bitcoin rises. I guess the reasons can be selling forces are dominantly from people who have available BTC on exchanges with BTC falls (investors hold in this case). In contrast, more investors will accept high fee and move their BTC from personal wallets to exchange wallets to cash out with BTC rises.
Data sources
Plots
For transactions
Code:
Summary for variables: feeperkbusd
by categories of: year
year | N mean sd p50 p25 p75 min max
----------+--------------------------------------------------------------------------------
2009 | 32490.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2010 | 67920.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2011 | 59627.0 0.0 0.2 0.0 0.0 0.0 0.0 36.6
2012 | 54526.0 0.0 0.3 0.0 0.0 0.0 0.0 49.1
2013 | 63433.0 0.2 1.6 0.1 0.1 0.2 0.0 296.6
2014 | 58865.0 0.2 0.8 0.1 0.1 0.2 0.0 190.3
2015 | 54321.0 0.1 0.3 0.1 0.1 0.1 0.0 47.6
2016 | 54851.0 0.3 0.6 0.3 0.2 0.4 0.0 134.6
2017 | 55928.0 9.7 19.3 3.7 1.6 8.7 0.0 230.0
2018 | 54498.0 5.2 13.1 1.4 0.7 2.7 0.0 134.0
2019 | 54232.0 2.5 2.6 1.5 0.7 3.1 0.0 44.0
2020 | 41933.0 3.3 3.3 2.0 1.0 4.5 0.0 23.6
----------+--------------------------------------------------------------------------------
Total | 652624.0 1.8 7.5 0.1 0.0 0.8 0.0 296.6
-------------------------------------------------------------------------------------------
For median of daily transactions
- I calculated median fees of all transactions for each day
- Then get summary stats over years
Last 4 years only (2017 - 2020)
Code:
Summary for variables: median
by categories of: year
year | N mean sd p50 p25 p75 min max
---------+--------------------------------------------------------------------------------
2009 | 358.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2010 | 365.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2011 | 365.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2
2012 | 366.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2013 | 365.0 0.2 0.1 0.1 0.1 0.2 0.0 0.6
2014 | 365.0 0.2 0.1 0.2 0.1 0.2 0.1 0.4
2015 | 365.0 0.1 0.0 0.1 0.1 0.1 0.0 0.2
2016 | 366.0 0.3 0.2 0.3 0.2 0.4 0.1 0.7
2017 | 365.0 9.9 18.6 4.2 1.8 9.0 0.6 153.6
2018 | 365.0 4.8 12.2 1.2 0.9 2.3 0.3 71.4
2019 | 365.0 2.4 2.3 1.7 0.8 3.0 0.3 14.2
2020 | 287.0 3.4 2.8 2.3 1.2 4.9 0.6 12.9
---------+--------------------------------------------------------------------------------
Total | 4297.0 1.7 7.2 0.1 0.0 1.0 0.0 153.6
------------------------------------------------------------------------------------------
Categorisation of transactions with 2 groups of fee
- Cutoff: 2 USD/ kB
- I chose the cutoff based on median of all transactions is at 0.1, p25 = 0, p75 = 0.8. Potential high outliers will be 1.5*0.8+0.8 = 2 USD
Code:
Fee per KB | year
(in USD) | 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 | Total
------------+------------------------------------------------------------------------------------------------------------------------------------+----------
< 2 USD/ kb | 32,490 67,920 59,618 54,506 63,150 58,771 54,313 54,763 18,577 35,375 32,462 21,243 | 553,188
| 100.00 100.00 99.98 99.96 99.55 99.84 99.99 99.84 33.22 64.91 59.86 50.66 | 84.76
------------+------------------------------------------------------------------------------------------------------------------------------------+----------
2+ USD/ kb | 0 0 9 20 283 94 8 88 37,351 19,123 21,770 20,690 | 99,436
| 0.00 0.00 0.02 0.04 0.45 0.16 0.01 0.16 66.78 35.09 40.14 49.34 | 15.24
------------+------------------------------------------------------------------------------------------------------------------------------------+----------
Total | 32,490 67,920 59,627 54,526 63,433 58,865 54,321 54,851 55,928 54,498 54,232 41,933 | 652,624
| 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 | 100.00
5 blocks with fee per kB >= $ 200
Code:
+------------------------------+
| blockid feeperkbusd year |
|------------------------------|
273641. | 273640 296.6493 2013 |
|------------------------------|
470190. | 470189 225.7396 2017 |
500440. | 500439 201.0803 2017 |
500522. | 500521 206.068 2017 |
500547. | 500546 229.978 2017 |
+------------------------------+
In 2020
- median: Median of fee per kb (in USD) by each date
- median_y: Median of fee per kb (in USD) for the whole year 2020 so far
- fee_dif = median - median_y.
- fee_difpc = (median - median_y)/median*100.
Raw results
Median
Code:
Summary for variables: median
by categories of: month
month | N mean sd p50 p25 p75 min max
----------+--------------------------------------------------------------------------------
2020m1 | 31.0 1.0 0.2 1.0 0.7 1.2 0.6 1.4
2020m2 | 29.0 1.5 0.3 1.4 1.2 1.7 1.0 2.2
2020m3 | 31.0 1.7 0.7 1.5 1.1 2.2 0.8 3.8
2020m4 | 30.0 1.1 0.8 0.9 0.8 1.1 0.6 5.4
2020m5 | 31.0 5.8 3.1 5.2 3.4 6.9 2.5 12.9
2020m6 | 30.0 2.3 1.1 1.9 1.6 2.8 1.1 6.3
2020m7 | 31.0 4.5 3.1 3.2 2.2 6.1 1.4 12.7
2020m8 | 31.0 6.8 2.6 6.8 4.6 8.6 2.9 12.5
2020m9 | 30.0 4.8 2.1 4.0 3.3 5.6 2.2 9.8
2020m10 | 13.0 5.4 1.8 5.1 3.7 7.0 2.7 8.0
----------+--------------------------------------------------------------------------------
Total | 287.0 3.4 2.8 2.3 1.2 4.9 0.6 12.9
-------------------------------------------------------------------------------------------
Fee_dif
Code:
Summary for variables: fee_dif
by categories of: month
month | N mean sd p50 p25 p75 min max
----------+--------------------------------------------------------------------------------
2020m1 | 31.0 -1.0 0.2 -1.0 -1.2 -0.8 -1.4 -0.6
2020m2 | 29.0 -0.5 0.3 -0.6 -0.7 -0.3 -0.9 0.2
2020m3 | 31.0 -0.3 0.7 -0.5 -0.9 0.2 -1.2 1.8
2020m4 | 30.0 -0.8 0.8 -1.0 -1.2 -0.9 -1.4 3.4
2020m5 | 31.0 3.8 3.1 3.2 1.4 5.0 0.5 11.0
2020m6 | 30.0 0.3 1.1 -0.0 -0.4 0.8 -0.9 4.3
2020m7 | 31.0 2.6 3.1 1.3 0.3 4.1 -0.5 10.8
2020m8 | 31.0 4.8 2.6 4.9 2.6 6.6 0.9 10.5
2020m9 | 30.0 2.8 2.1 2.1 1.3 3.7 0.2 7.8
2020m10 | 13.0 3.5 1.8 3.1 1.8 5.1 0.7 6.0
----------+--------------------------------------------------------------------------------
Total | 287.0 1.4 2.8 0.3 -0.8 2.9 -1.4 11.0
-------------------------------------------------------------------------------------------
In percent
Code:
+---------------------+
| month fee_difpc |
|---------------------|
1. | 2020m1 -50.73672 |
2. | 2020m2 -25.70358 |
3. | 2020m3 -13.5845 |
4. | 2020m4 -42.70243 |
5. | 2020m5 195.7841 |
|---------------------|
6. | 2020m6 15.97533 |
7. | 2020m7 129.9602 |
8. | 2020m8 244.3547 |
9. | 2020m9 142.0839 |
10. | 2020m10 175.995 |
+---------------------+
Bitcoin price and fee difference (in %)
It is overview and next steps, I will move further by looking at smaller rolling windows (weekly median, not periodic median like 1-year or 2-year). Smaller rolling windows can help to see better effects of bitcoin price on median of daily fees and their correlations.
- In plots, I used multipliers for fee_difpc variables simply to make more visually comparative.
- Median: calculated as same as above
- Median_y: calculated for specific period: 2020 only or from 2019 to 2020.
Jump to: