The history of Bitcoin difficulty adjustment (in percent)

[tranthidung]

Here you go with my new thread. If you feel my threads are boring, please ignore me or this thread. No need to be serious, I am data-freak. If you read it, I thank you for your attention and your time. 


Today, bitcoin network has its second biggest difficulty decreasing (negatively) adjustment, at -16.04992%. The biggest one in negative value is at -18.03, 9 years ago, in 2011.

See details:

• Block 655199: 19997335994446
• Block 655200: 16787779609933

The network difficulty was adjusted ~ 7 hours ago with the fall of -16.04992% in difficulty between block 655199 and 655200. It is a big adjustment. The second biggest decreasing adjustment in history of bitcoin network.

The biggest one is 9 years ago (in 2011) between 2 following blocks and percent of decreasing adjustment is -18.03. Impressive for today adjustment.

• Block 151199: 1468195.4272208
• Block 151200: 1203461.92638

• Over the history, there are 123 adjustments have values > 10% and 6 adjustments have values < -10% (include the one for today).
• The biggest positive adjustment is 300% on 16jul2010.
• It was adjusted 309 times (the dataset is for 308 only). I will update with dataset for today later depends on loyceclub
• Median (interquartile range): 7.09% (1.67% - 16.42%).
• Min - max: -18.03 - 300
• Ideas: from plot, you can see some spikes and you can dig to find out what happened in the past

Code:

    variable |         N      mean        sd       p50       p25       p75       min       max
-------------+--------------------------------------------------------------------------------
     diff_pc |       309  11.75175  22.34688      7.08      1.59      16.2    -18.03       300
----------------------------------------------------------------------------------------------

Positive changes > 10%
123 recorded days.

Code:

      +------------------------+
      |      date   diff_pcpos |
      |------------------------|
 357. | 30dec2009        18.29 |
 369. | 11jan2010        10.33 |
 393. | 04feb2010        35.29 |
 403. | 14feb2010        38.99 |
 413. | 24feb2010        49.59 |
      |------------------------|
 425. | 08mar2010        19.83 |
 449. | 01apr2010         33.3 |
 460. | 12apr2010         28.5 |
 469. | 21apr2010        46.61 |
 482. | 04may2010        12.08 |
      |------------------------|
 507. | 29may2010        40.29 |
 533. | 24jun2010        11.64 |
 545. | 06jul2010        21.11 |
 552. | 13jul2010        93.12 |
 555. | 16jul2010          300 |
      |------------------------|
 566. | 27jul2010        34.52 |
 575. | 05aug2010         44.2 |
 585. | 15aug2010        45.32 |
 596. | 26aug2010        21.81 |
 609. | 08sep2010        14.36 |
      |------------------------|
 620. | 19sep2010        28.75 |
 629. | 28sep2010        43.67 |
 652. | 21oct2010        55.95 |
 661. | 30oct2010        43.87 |
 671. | 09nov2010        46.73 |
      |------------------------|
 680. | 18nov2010        51.37 |
 692. | 30nov2010        17.64 |
 701. | 09dec2010        51.67 |
 713. | 21dec2010        18.22 |
 726. | 03jan2011        12.59 |
      |------------------------|
 738. | 15jan2011        13.06 |
 750. | 27jan2011        19.39 |
 762. | 08feb2011        18.11 |
 772. | 18feb2011        40.24 |
 781. | 27feb2011        52.47 |
      |------------------------|
 791. | 09mar2011        37.06 |
 818. | 05apr2011        19.38 |
 831. | 18apr2011        12.15 |
 843. | 30apr2011        18.76 |
 852. | 09may2011        43.54 |
      |------------------------|
 861. | 18may2011        55.07 |
 869. | 26may2011        78.15 |
 880. | 06jun2011        30.44 |
 889. | 15jun2011        54.59 |
 898. | 24jun2011        57.27 |
      |------------------------|
 910. | 06jul2011        13.33 |
 936. | 01aug2011         11.7 |
1218. | 09may2012        14.89 |
1326. | 25aug2012         11.4 |
1338. | 06sep2012        10.38 |
      |------------------------|
1490. | 05feb2013        10.33 |
1503. | 18feb2013        11.47 |
1514. | 01mar2013        19.63 |
1527. | 14mar2013        10.98 |
1537. | 24mar2013        38.13 |
      |------------------------|
1549. | 05apr2013        14.59 |
1561. | 17apr2013        16.96 |
1574. | 30apr2013        12.28 |
1586. | 12may2013        11.03 |
1610. | 05jun2013        28.41 |
      |------------------------|
1621. | 16jun2013        23.92 |
1634. | 29jun2013        10.32 |
1646. | 11jul2013        22.63 |
1657. | 22jul2013        19.47 |
1669. | 03aug2013        19.63 |
      |------------------------|
1679. | 13aug2013        35.88 |
1690. | 24aug2013         29.4 |
1701. | 04sep2013        32.22 |
1711. | 14sep2013        29.56 |
1722. | 25sep2013        32.13 |
      |------------------------|
1733. | 06oct2013        27.19 |
1743. | 16oct2013        41.45 |
1753. | 26oct2013        46.02 |
1763. | 05nov2013         30.7 |
1775. | 17nov2013        19.29 |
      |------------------------|
1787. | 29nov2013        16.07 |
1798. | 10dec2013        28.41 |
1809. | 21dec2013        30.01 |
1821. | 02jan2014        20.12 |
1832. | 13jan2014        26.16 |
      |------------------------|
1843. | 24jan2014        22.59 |
1855. | 05feb2014        19.49 |
1867. | 17feb2014        19.39 |
1878. | 28feb2014        21.92 |
1891. | 13mar2014        11.39 |
      |------------------------|
1902. | 24mar2014         17.8 |
1914. | 05apr2014        22.23 |
1926. | 17apr2014        14.04 |
1938. | 29apr2014        14.64 |
1951. | 12may2014        10.66 |
      |------------------------|
1963. | 24may2014         18.1 |
1975. | 05jun2014        12.44 |
1988. | 18jun2014        14.51 |
1999. | 29jun2014        24.93 |
2050. | 19aug2014        20.86 |
      |------------------------|
2062. | 31aug2014        15.03 |
2087. | 25sep2014         16.2 |
2128. | 05nov2014        10.05 |
2512. | 24nov2015        10.44 |
2536. | 18dec2015        18.14 |
      |------------------------|
2549. | 31dec2015        11.16 |
2587. | 07feb2016        20.06 |
2599. | 19feb2016        13.44 |
2872. | 18nov2016        10.68 |
2937. | 22jan2017        16.64 |
      |------------------------|
3070. | 04jun2017         13.9 |
3110. | 14jul2017        13.53 |
3176. | 18sep2017        19.58 |
3214. | 26oct2017        21.39 |
3255. | 06dec2017        18.11 |
      |------------------------|
3267. | 18dec2017        17.74 |
3293. | 13jan2018        15.36 |
3305. | 25jan2018        16.84 |
3317. | 06feb2018        10.43 |
3436. | 05jun2018        14.71 |
      |------------------------|
3490. | 29jul2018        14.88 |
3645. | 31dec2018        10.03 |
3795. | 30may2019        11.26 |
3835. | 09jul2019        14.23 |
3862. | 05aug2019        10.78 |
      |------------------------|
3901. | 13sep2019        10.38 |
4178. | 16jun2020        14.95 |
4274. | 20sep2020        11.35 |
      +------------------------+

Negative changes < - 10%
5 days in history, and the one for today is the six record.

Code:

      +------------------------+
      |      date   diff_pcneg |
      |------------------------|
1010. | 14oct2011       -13.09 |
1027. | 31oct2011       -18.03 |
1449. | 26dec2012       -11.59 |
3617. | 03dec2018       -15.13 |
4096. | 26mar2020       -15.95 |
      +------------------------+

Updated plots

[1714184029]

1714184029

[1714184029]

1714184029

[1714184029]

1714184029

[1714184029]

1714184029

[DougM]

Here you go with my new thread. If you feel my threads are boring, please ignore me or this thread. No need to be serious, I am data-freak. If you read it, I thank you for your attention and your time.  

  Good stuff...I like data too. Keep it coming    Can you post your difficulty adjustment graph suppressing the ~300 % outliner so I could get a better look at the chart's pattern? too small to make it out for my eyes   

I am not sure you saw the articles recently posted in the press on the recent Bitcoin mining difficulty drop? If not check out this thread:

https://bitcointalk.org/index.php?topic=5286529.msg55513356#msg55513356

[tranthidung]

Can you post your difficulty adjustment graph suppressing the ~300 % outliner so I could get a better look at the chart's pattern? too small to make it out for my eyes    

What did you mean? I can give you specific-year plots for better visualization.

Is it what you suppose to have?

I am not sure you saw the articles recently posted in the press on the recent Bitcoin mining difficulty drop? If not check out this thread:

https://bitcointalk.org/index.php?topic=5286529.msg55513356#msg55513356

Thanks for the link. The article has little information, except a few additional information that what happened around the days big difficulty changes (not all history, only in 2020).

[LoyceV]

If not check out this thread:

https://bitcointalk.org/index.php?topic=5286529.msg55513356#msg55513356

I was wondering why the difficulty would go down when Bitcoin prices go up. It doesn't make much sense, but it turns out it has to do with seasonal hydro power in China.

[DougM]

Can you post your difficulty adjustment graph suppressing the ~300 % outliner so I could get a better look at the chart's pattern? too small to make it out for my eyes    

What did you mean? I can give you specific-year plots for better visualization.

Is it what you suppose to have?

I am not sure you saw the articles recently posted in the press on the recent Bitcoin mining difficulty drop? If not check out this thread:

https://bitcointalk.org/index.php?topic=5286529.msg55513356#msg55513356

Thanks for the link. The article has little information, except a few additional information that what happened around the days big difficulty changes (not all history, only in 2020).

Thank you for the new chart.  However it would be good to see it from the beginning like your original chart however the 300% max outliner forces the scale to be small so it is hard to see the pattern over time. If you set the Y axis max to 80% it will clip the 300% outliner, but the rest of the data will be easier to see. 

The past historical significant changes should be documented in this forum since they were living through it all. 
For example why did it jump 300% mid July 2010?

...Biggest change in difficult was probably around 10 july 2010 when bitcoin was announced on slashdot... resulting in the slashdot effect, a jump in difficulty of 300%.

Plenty of good related historical threads if you are interested. I grabbed a few you might like:

Price vs Difficulty Charts - indicators for buying or mining  May 06, 2011, 06:00:38 PM https://bitcointalk.org/index.php?topic=7427.0
Price drives difficulty November 08, 2012, 02:42:12 PM https://bitcointalk.org/index.php?topic=123238.0
Will the difficulty go down again ? December 13, 2014, 11:36:13 PM https://bitcointalk.org/index.php?topic=892774.0

[tranthidung]

I was wondering why the difficulty would go down when Bitcoin prices go up. It doesn't make much sense, but it turns out it has to do with seasonal hydro power in China.

It happens repeatedly many times and there are many factors are related or correlated each other:

• Seasonal changes in supplies of hydro power in China
• FOMO of newbie miners (they join when price rises up, buy rigs, set up and mine but they are unstable in emotions, don't understand how things work, unbale to survive along with short term unprofitable minings, etc.). They turn off rigs when price hits peak but mining rewards they get can not enough for power bills.
• Manipulations (I doubt about it for years and believe it is).
• When I have time, I will combine data and make charts for total hashrate (or hashrate changes in %), difficulty changes, and price. Difficulty has lag-effects from hashrate changes.

Thank you for the new chart.  However it would be good to see it from the beginning like your original chart however the 300% max outliner forces the scale to be small so it is hard to see the pattern over time. If you set the Y axis max to 80% it will clip the 300% outliner, but the rest of the data will be easier to see.  

Got it. To be honest, hours after I made that post, I seems to understand your idea about 300% outlier on the chart. Hang on a few minutes, then I will give you update. My solution is replace that outlier to the p75 value at 16.42% but not exclude the outlier from the chart.

I forgot to run sum stats after the replacement. The outlier should be replace at the max value below 93.12%. Anyway, it is not a serious issue with only one data point.

Here you go:

Most of difficulty adjustment / change (%) are positive. There are very few negative changes and big negative values tend to occur in early or last months of each year

After the replacement, stats will be like these. Max value now is 93.12% (close to what you supposed to have 80%)

Code:

    variable |         N      mean        sd       p50       p25       p75       min       max
-------------+--------------------------------------------------------------------------------
     diff_pc |       309   10.8333  15.12739      7.08      1.59      16.2    -18.03     93.12
----------------------------------------------------------------------------------------------

The past historical significant changes should be documented in this forum since they were living through it all.

For example why did it jump 300% mid July 2010?

...Biggest change in difficult was probably around 10 july 2010 when bitcoin was announced on slashdot... resulting in the slashdot effect, a jump in difficulty of 300%.

Yeap, it is the beauty of data and statistics. From raw analysis, we can dig deeper, find out what happened in the past (from news or deeper analyses later). I thank you for the information.

Plenty of good related historical threads if you are interested. I grabbed a few you might like:

Price vs Difficulty Charts - indicators for buying or mining  May 06, 2011, 06:00:38 PM https://bitcointalk.org/index.php?topic=7427.0
Price drives difficulty November 08, 2012, 02:42:12 PM https://bitcointalk.org/index.php?topic=123238.0
Will the difficulty go down again ? December 13, 2014, 11:36:13 PM https://bitcointalk.org/index.php?topic=892774.0

When I have time, we can make a chronological compilation for it: changes from difficulty and likely real things happened behind.  

[100bitcoin]

Fantastic work. Is it possible to prepare a difficulty vs. price graph, where difficulty will be in X-axis and price will be in Y-axis?

[tranthidung]

Bump it after many months of silence.

Again, new all time high of total hashrate kicks off sudden drop of hashrate. Reasons, this time, are blamed on flood, rain, coal plants, etc. but do you think any of them are real problems?

Without flood, rains and any visible reasons you can think of, there are scheduled plans for instant shutdowns of big numbers of mining rigs.

It happens and repeats many times. Hashrates top up, price tops up, then accidents come out.

• How Accidents in Chinese Coal Mines Are Affecting Bitcoin Mining
• Story from News
  
• More things behind such as TA indicators
• Time to read, filter and digest news

[tranthidung]

Update

• Current difficulty: 19932791027263
• According to diff.cryptothis.com, the next difficulty adjustment would be -28%. If the difficulty adjusts like the estimation, it would be the biggest negative adjustment in history.

• The biggest (-) adjustment happens 10 years ago, on 31 Oct 2011: -18%
• In 2020 and 2021, there are a few big negative adjustments from -12% to -16%, two in 2020 and two in 2021 (see details below)

Code:

      +------------------------+
        |      date   diff_pcneg |
        |------------------------|
     1. | 31oct2011       -18.03 |
     2. | 03nov2020       -16.05 |
     3. | 30may2021       -15.97 |
     4. | 26mar2020       -15.95 |
     5. | 03dec2018       -15.13 |
     6. | 14oct2011       -13.09 |
     7. | 01may2021       -12.61 |
     8. | 26dec2012       -11.59 |
        +------------------------+

To make stats more visual friendly, I make two bar charts for you.

See possible lag effects between price spread and difficulty adjustment with that post

• Newest update on difficulty adjustment
• Price spread is the difference between highest and lowest daily price
• Lag effects: Possibly, significant increase of difficulty (relates to hashrate) results in price topping days later. Horrific drop of difficulty causes price dumping days later
• It's neither financial advice nor prediction

[kxwhalexk]

I like to use data to comfort others， Most of the time, I also have the patience to collect and organize data
But I am still a novice and I am still learning. Do you have your own website? Or can you share your social media？
I can learn so much from you, thank you very much!

[Pmalek]

Do you have your own website? Or can you share your social media

You can bookmark tranthidung's profile and visit it whenever you want to see what posts or threads he has written recently. Bookmark this > https://bitcointalk.org/index.php?action=profile;u=1292764.

If you want to talk to him, you'll be able to find his Telegram ID at the bottom of his services thread. It's @tranthidung1292764

[tranthidung]

Update

• The latest difficulty adjustment is -27.94%, at the block #689472
• Daily price spread between
  
  • High and Low price
  • Open and Close price

• Interpretation
  
  • For price (Orange) Positive: good days, Negative ~ bad days.
  • Green spikes (+) usually have lag effects and you can see some other good spikes for price a few days later.
  • Red spikes (-) usually have inverse-lagged from price in previous days. It usually triggers by significant price falls days before. It is logical if considering about the capitulation of weak miners. Fortunately, after such negative spikes, price rallies.

I like to use data to comfort others

"In God we trust, for the others please bring data". Data is not to comfort others but to convince others what you are talking about. Obviously, it depends on quality of data and analysis method, etc. It is not true that all data reflects the truth.

Most of the time, I also have the patience to collect and organize data
But I am still a novice and I am still learning.

Practice makes perfect. So you can and I am not too good with data.

Do you have your own website? Or can you share your social media？
I can learn so much from you, thank you very much!

Thanks for your interest but unfortunately, I don't have a personal website. You can follow me on Twitter as well (it is a newbie account, LOL)

• TranDzung on Twitter