Author Topic: What's your Eddington number ?  (Read 51519 times)

CrazyEnglishTriathlete

  • Miles eaten don't satisfy hunger
  • 3x Brimstone ancien 3x Pendle/Tan Hill DNF
    • CET Ride Reports and Blogs
Re: What's your Eddington number ?
« Reply #50 on: December 30, 2012, 07:26:46 am »
Like all good numbers it invites some debate over definitions. I've worked on the view that events like Paris-Brest-Paris are single rides, even if they take multiple days,

The definition commonly attributed to Eddington is quite clear (and different to your interpretation) - "the number of days a cyclist has cycled more than E miles"

Indeed it includes no reference to "single rides" so for a ride to work in the morning and a ride home in the evening would be combined into the total for that day.

You are almost certainly more correct than me - however - I will stick with my calculation 1) as it coincides with my personal cycling log that has been going since 1997 and so would be impossible to rework, 2) it avoids the difficulty of trying to calculate (for example) where I was at midnight on a 24 hour TT, 3) it will be a conservative estimate of the number, and 4) it doesn't involve double counting   ;D
Eddington Numbers 125 (imperial), 168 (metric) 518 (furlongs)  111 (nautical miles)

Re: What's your Eddington number ?
« Reply #51 on: December 30, 2012, 04:04:55 pm »
Quote from: CrazyEnglishTriathlete
You are almost certainly more correct than me - however - I will stick with my calculation 1) as it coincides with my personal cycling log that has been going since 1997 and so would be impossible to rework, 2) it avoids the difficulty of trying to calculate (for example) where I was at midnight on a 24 hour TT, 3) it will be a conservative estimate of the number, and 4) it doesn't involve double counting   ;D

That's fine but it wouldn't be right to refer to it as your 'Eddington Number' - let's just call it your 'Crazy English Number'  ;)

CrazyEnglishTriathlete

  • Miles eaten don't satisfy hunger
  • 3x Brimstone ancien 3x Pendle/Tan Hill DNF
    • CET Ride Reports and Blogs
Re: What's your Eddington number ?
« Reply #52 on: December 30, 2012, 05:11:25 pm »
I've amended my signature... :thumbsup:
Eddington Numbers 125 (imperial), 168 (metric) 518 (furlongs)  111 (nautical miles)

CrazyEnglishTriathlete

  • Miles eaten don't satisfy hunger
  • 3x Brimstone ancien 3x Pendle/Tan Hill DNF
    • CET Ride Reports and Blogs
Re: What's your Eddington number ?
« Reply #53 on: January 04, 2013, 05:39:25 pm »
This gets a little bit brain-frying.  I've looked at the multi-day continuous rides I have done:

8 600km Audax rides, with at least 200k done on two consecutive days
2 24-hr time trials, with at least 300k done on two consecutive days
The Mille Alba - 200 miles or more on 3 consecutive days
Paris Brest Paris x2.  Each of these had 200km+ on 3 consecutive days.  Technically with the 4pm start in 2011 I did over 200km on day 1, a further 400km to Brest on day 2, 400km to Villaine on day 3 and 200km to Paris on day 4
LEL - 200km+ on 4 days (the 5th day was a little under at about 180km)
Mille Miglia - 200km+ on 4 days - the first "day" was only 3 hours and we only did 120km on the last day.

If we use an Eddington number calculated on the basis of what was done on a Calendar day - this would give me an additional 22 rides of 200km (124.5 miles or more) - which would make a dramatic improvement to my Eddington number (imperial) from 102 to 118.  Its much more difficult to work out what it would do to the metric number as several of the 400km rides I have done spanned 2 days and had between 136km and 200km on the second day.  However, at the moment it would be fairly safe to say it was in the region of 161 due to the milestone-driven cluster of rides I've done of that distance or slightly over.

I guess, however, that Sir Arthur wasn't in the business of Audax or 24hr time trials and therefore did not have this conundrum.  At midnight on a 24Hr time trial I wouldn't be able to tell you what road I was on let alone how far I had gone.

And, if we take Wowbagger's suggestion of a 24 hour period it gets even more difficult.  At least it resolves the 24-TT situation but for the long Audax rides, especially the 600s I have no idea how far I would have gone at 6am on the second day as typically I would be halfway through the first stage after the sleep stop and wouldnt have a proper time of departure from the first go.

So, I'm going to stick with the way i've calculated it and know that on a strict definition the results are higher.

By the way - a 114 mile ride today has improved it per my calculation   ;D but I guess would not under the strict defintion.   :demon:
Eddington Numbers 125 (imperial), 168 (metric) 518 (furlongs)  111 (nautical miles)

Re: What's your Eddington number ?
« Reply #54 on: January 04, 2013, 08:42:49 pm »
Once an accountant, always an accountant  ;) (takes one to know that)

Chapeaux on your true E number Sir (and for working it out)  :thumbsup: Eddington presumably would have devised a watertight definition if he'd have realised that one day some cyclists would ride such crazy distances on such crazy timescales. You don't need to worry about metric by the way, I don't think Eddington mentioned that.

CrazyEnglishTriathlete

  • Miles eaten don't satisfy hunger
  • 3x Brimstone ancien 3x Pendle/Tan Hill DNF
    • CET Ride Reports and Blogs
Re: What's your Eddington number ?
« Reply #55 on: January 05, 2013, 01:19:38 pm »
Oh the perils of doing a quick calculation!  :demon:  I simply took 23 rides out of my list in revising the Eddington calculation.  So that had 102 rides more than 118 miles!  But the Eddington number has its subtleties - 118 is 118 rides of 118 miles.  And I can't claim that.  Its more like 110.   :(

Never mind.  I will stick with 102 and have to do 5 rides of 103+ miles to increase it, so that should be by the end of May.  After that it starts to get easier as I get past the cluster of rides just over 100 miles and it will be a while before I get to 125 x 125 which is where I hit the 200km Audax distances. 

Then I might revert to a previously set target.  When I was young my ambition was to be a cricketer - around about the time that Messrs Cowdrey, Amiss, Boycott, Zaheer and others were reaching 100 first class centuries.  I passed that milestone on the bike in 2011 and so the next one is Jack Hobbs first class record of 197.

You're right.  Once an accountant.....

(although large quantities of Smirnoff are reputed to help)
Eddington Numbers 125 (imperial), 168 (metric) 518 (furlongs)  111 (nautical miles)

Re: What's your Eddington number ?
« Reply #56 on: January 06, 2013, 12:33:49 am »
I've been tracking my Eddington number for a while now (http://www.buttonsofmymind.co.uk/eddington-number-cycling/). Using the definition of days as calendar dates I'm currently E56 (imperial) and my next target is E60 (I need 11 rides >60 miles). In metric I'm E79 but since it's not equivalent (100 days of 100 km is harder than 62 days of 62 miles) I've never looked at the metric version before :)

John

Wowbagger

  • Dez's butler
    • Musings of a Gentleman Cyclist
Re: What's your Eddington number ?
« Reply #57 on: January 06, 2013, 12:35:59 am »
Since Eddington himself used imperial, I don't really see that the concept of a "metric" Eddington number is of any value.

Why doesn't one of our august number of metric enthusiasts "invent" it and name it after themselves?  :P
Oh, Bach without any doubt. Bach every time for me.

AAO

Re: What's your Eddington number ?
« Reply #58 on: January 06, 2013, 12:53:29 am »
Metric could be called your Kiddington Number. 

simonp

  • Omnomnomnipotent.
Re: What's your Eddington number ?
« Reply #59 on: January 06, 2013, 04:01:50 am »
Yesterday I invented the K-number which is the number of calendar years, K, in which you have ridden at least K,000km. I think mine is about 6.

Re: What's your Eddington number ?
« Reply #60 on: January 06, 2013, 06:29:20 pm »
A whole new way to play with a decade's-worth of data.  :)

I spent a happy evening discovering that my Eddington number is 94.  If all goes well and I finally complete a RRTY, I should get it above 100.  It's a motivating thought for this time of the year.

MercuryKev

  • Maxin' n Audaxin'
Re: What's your Eddington number ?
« Reply #61 on: January 06, 2013, 06:36:54 pm »
A whole new way to play with a decade's-worth of data.  :)

I spent a happy evening discovering that my Eddington number is 94.  If all goes well and I finally complete a RRTY, I should get it above 100.  It's a motivating thought for this time of the year.


Strange to discover that we've been carrying out the same exercise in parallel :)  I only had 3 years of data to work with and I'm currently sitting at 71.

CrazyEnglishTriathlete

  • Miles eaten don't satisfy hunger
  • 3x Brimstone ancien 3x Pendle/Tan Hill DNF
    • CET Ride Reports and Blogs
Re: What's your Eddington number ?
« Reply #62 on: January 06, 2013, 10:20:14 pm »
Yesterday I invented the K-number which is the number of calendar years, K, in which you have ridden at least K,000km. I think mine is about 6.

How about the F-number, which is based on Furlongs.  Allows a decent score - mine would be over 400  :smug:
Eddington Numbers 125 (imperial), 168 (metric) 518 (furlongs)  111 (nautical miles)

What's your Eddington number ?
« Reply #63 on: January 06, 2013, 10:38:18 pm »
How about the F-number, which is based on Furlongs.  Allows a decent score - mine would be over 400  :smug:
Or the FFF number, which would denote the number of fortnights in which you have cycled FFF furlongs and also consumed FFF firkins (of beer; or, if metric, of wine).

What's your Eddington number ?
« Reply #64 on: January 06, 2013, 10:58:04 pm »
Strange to discover that we've been carrying out the same exercise in parallel :)

I've kept weekly and cumulative annual total distances since late 2003, when I got back into riding seriously, but I didn't have the individual ride distances needed for the E-calculation all in one place. So I went back through diaries, AUK records, blog-posts, GPS tracks and memory to create as good a list as I can of all my rides of over 100 km, right back to my first century ride (Ardgay-Laxford Bridge-Tongue, solo, aged 16) and including a 7-day LEJOG, Raids Pyrenean and Corsican, Wellington-Aukland (NZ) and holiday rides in France, Italy and Germany,

Happy memories. If I ever have to stop riding, I'll be able to relive them all. Meanwhile, more living to do: LEL and the Raid Alpine planned for this year.

clarion

  • Tyke
Re: What's your Eddington number ?
« Reply #65 on: January 07, 2013, 10:06:45 am »
I can't believe that I'm seriously considering dredging back through the records, adding up daily totals, and converting to miles before sorting and counting...
Getting there...

Re: What's your Eddington number ?
« Reply #66 on: January 07, 2013, 10:28:26 am »
I'm trying to fight against the temptation to do it.

(Putting it on my Jobs List will probably secure the win.)

clarion

  • Tyke
Re: What's your Eddington number ?
« Reply #67 on: January 07, 2013, 04:03:20 pm »
I can't believe that I'm seriously considering dredging back through the records, adding up daily totals, and converting to miles before sorting and counting...
49 (data only back to 2009)
One ride required to get to the magic 50.  Six rides for 51...

But at least I now have a spreadsheet for entering any daily totals over 80km.  And just 98 imperial centuries to get to 100 :-\
Getting there...

Re: What's your Eddington number ?
« Reply #68 on: January 07, 2013, 04:51:22 pm »
How to work out my Ed# using a spreadsheet?  ???
I've a column of numbers from 1 mile to 177 miles, 2098 entries deep. I think my Ed# is somewhere between 67 and 81.
What do I need to get my spreadsheet to do to work it out accurately?

clarion

  • Tyke
Re: What's your Eddington number ?
« Reply #69 on: January 07, 2013, 05:48:28 pm »
You need to sort them in descending order of distance, then read down to find where the row number (assuming no header rows) and the distance converge, and read off the lower distance value (rounded to whole numbers, natch).
Getting there...

Re: What's your Eddington number ?
« Reply #70 on: January 07, 2013, 05:54:25 pm »
You need to sort them in descending order of distance, then read down to find where the row number (assuming no header rows) and the distance converge, and read off the lower distance value (rounded to whole numbers, natch).
I was expecting a majick formula!  ::-)

My E = 76. Five rides of >77 will mean E = 77.

MercuryKev

  • Maxin' n Audaxin'
Re: What's your Eddington number ?
« Reply #71 on: January 07, 2013, 10:35:46 pm »
You need to sort them in descending order of distance, then read down to find where the row number (assuming no header rows) and the distance converge, and read off the lower distance value (rounded to whole numbers, natch).
I was expecting a majick formula!  ::-)



My E = 76. Five rides of >77 will mean E = 77.

Here you go:

In column A have a list of your rides in descending order - Name this range (the whole column) 'Dist'
In column B put this formula =IF((COUNTIF(Dist,">=" & A2))>=A2,(COUNTIF(Dist,">=" & A2)), "n/a") and copy the formula down the whole list

Look down column B and the first number returned after a load of n/as should be your Eddington number (if your rides are sorted in descending order)

This formula assumes that you have column headers in row 1



Re: What's your Eddington number ?
« Reply #72 on: January 08, 2013, 12:22:01 am »
You need to sort them in descending order of distance, then read down to find where the row number (assuming no header rows) and the distance converge, and read off the lower distance value (rounded to whole numbers, natch).
I was expecting a majick formula!  ::-)



My E = 76. Five rides of >77 will mean E = 77.

Here you go:

In column A have a list of your rides in descending order - Name this range (the whole column) 'Dist'
In column B put this formula =IF((COUNTIF(Dist,">=" & A2))>=A2,(COUNTIF(Dist,">=" & A2)), "n/a") and copy the formula down the whole list

Look down column B and the first number returned after a load of n/as should be your Eddington number (if your rides are sorted in descending order)

This formula assumes that you have column headers in row 1

Modifying that formula for OpenOffice Calc gives an Eddington number 1 too great for me. The first row with a number is the row under the actual Eddington number row. I've altered it just to take 1 off the return if the statement is true and it works OK.

=IF((COUNTIF(Dist;">="&D2))>=D2;(COUNTIF(Dist;">="&D2))-1;"n/a")

My distances are in column D starting in cell D2.

The way I do it is just to create a column adjacent to the descending distances with the numbers 1,2,3,4 etc. down it. You can just drag down the sequence to fill in the column. This number is the Eddington number in the last row where is still less than the distance.

i.e. in my table the relevant section looks like this...

NUMBER   DATE   RIDES   MILES
52   01/02/09   1   57.8
53   25/07/09   1   56.8
54   31/08/10   5   56.6
55   04/01/09   1   56.2
56   31/05/11   1   56.1
57   22/05/10   3   55.8
58   29/05/09   1   55.3
59   04/06/11   1   55.1
60   29/05/11   1   54.9

So my current Eddington number is E56.

I just add new rides to the bottom and sort columns B to D on distance descending leaving A as it is. Thus my Eddington number will increase to E57 once I've done 5 more days of more than 57 miles (I need to wipe out the 4 days of 56 and a bit miles and also do one more to bump the rows down to 57).

John

mcshroom

  • Mushroom
Re: What's your Eddington number ?
« Reply #73 on: January 08, 2013, 01:23:32 am »
I went for a slightly different countif.

Put the distances in column A, a set of ascending numbers in column B, and then this formula in column C: -

=COUNTIF($A$2:$A$1000, ">="&B2)     - Headers in row one for this

Copy the cells down in column C (so &B2 becomes &B3, &B4 etc.). At the highest number where Column A and Column B read the same number, that's your Eddington Number.

eg: -

Ride  Distance  Count
1.2          1      5
3.8          2      4
2.6          3      3
4.9          4      1
3.1          5      0


Eddington number here would be 3
Climbs like a sprinter, sprints like a climber!

clarion

  • Tyke
Re: What's your Eddington number ?
« Reply #74 on: January 08, 2013, 10:00:36 am »
Why?

Eddington had a simple system - count up the number exceeding a certain distance.  Reading the row numbers on a spreadsheet gives us that without any fancy calculations (well, other than the one converting my km into miles for the reference column).

Could. Not. Be Simpler.  Why complicate things unnecessarily?  Surely Eddington, as a scientist, would frown on anything beyond the simplest method required.
Getting there...