Page 12 of 30
Re: Ask Graeme?
Posted: Sat Dec 07, 2013 4:58 pm
by Gavin Chipper
Graeme Cole wrote:Jack Worsley wrote:Graeme, could you please post a table off all old 15 octochamps' estimated totals in the new 15 using the following formula:
(average points scored per numbers round - average points scored per letters round) x 8, + original total, then rounded to the nearest whole number?
How much the new format changes someone's projected score depends on how good they are at letters and numbers. I'm interested to see how many people would have a lower projected score in the new format (I think there will be a few but mostly at the lower end of the octochamp totals). Could you also include a column which compares each octochamp's position in the table to the old 15? For example, if the sixth highest scoring octochamp in the old 15 has the fourth highest projected new 15 total, there could be a column which reads "+2" or something like that. Cheers
I make it this:
Code: Select all
OLD NEW RANK
TOTAL TOTAL +/- +/-
1. Jack Hurst 946 963 +17 +0
2. Andrew Hulme 930 943 +13 +0
3. Kirk Bevins 925 940 +15 +0
4. Julian Fell 924 930 +6 +0
5. Craig Beevers 907 928 +21 +0
6. Adam Gillard 903 921 +18 +0
7. Eoin Monaghan 898 919 +21 +0
7. Edward McCullagh 896 919 +23 +1
9. Conor Travers 890 905 +15 +1
10. Chris Davies 892 900 +8 -1
11. David O'Donnell 880 899 +19 +0
12. Chris Wills 875 893 +18 +0
13. Chris Cummins 858 884 +26 +2
14. Stewart Holden 870 874 +4 -1
15. Jon Corby 856 873 +17 +1
16. Innis Carson 861 871 +10 -2
17. Tom Hargreaves 850 868 +18 +0
18. Jonathan Rawlinson 850 867 +17 -1
19. Steven Briers 843 866 +23 +2
20. Paul Gallen 846 863 +17 +0
21. Matthew Shore 850 855 +5 -4
22. Mark Deeks 824 850 +26 +5
22. Jack Welsby 831 850 +19 +3
24. Marcus Hares 834 849 +15 +0
24. James Hurrell 838 849 +11 -1
26. Jack Worsley 818 848 +30 +6
27. Daniel Pati 840 843 +3 -5
28. Richard Brittain 820 840 +20 +2
29. Grace Page 829 838 +9 -3
30. Paul Howe 815 837 +22 +5
31. Scott Gillies 810 835 +25 +8
32. John Brackstone 822 833 +11 -4
32. Graeme Cole 813 833 +20 +4
34. Charlie Reams 820 831 +11 -4
34. Tom Barnes 822 831 +9 -6
34. Jon O'Neill 804 831 +27 +9
37. George Greenhough 817 827 +10 -3
38. Mark Tournoff 809 825 +16 +2
39. Lee Hartley 811 824 +13 -2
40. Stuart Earl 807 823 +16 +2
41. Martin Bishop 809 822 +13 -1
42. Oliver Garner 802 820 +18 +2
43. Junaid Mubeen 790 817 +27 +7
44. Ryan Taylor 792 816 +24 +5
45. Jeffrey Hansford 818 813 -5 -13
46. John Hunt 788 811 +23 +5
47. John Mayhew 811 810 -1 -10
48. Richard Heald 795 808 +13 -1
49. Peter Lee 801 805 +4 -4
49. Stuart Solomons 796 805 +9 -3
51. Paul James 794 799 +5 -3
52. Cate Henderson 782 798 +16 +1
53. Tom Rowell 774 796 +22 +5
54. John Davies 766 794 +28 +10
55. Keith Maynard 785 793 +8 -3
56. Jimmy Gough 782 791 +9 -3
57. Rupert Stokoe 776 788 +12 +0
57. Aaron Webber 773 788 +15 +2
59. David Barnard 771 787 +16 +2
60. Jim Bentley 756 783 +27 +10
60. Mike Pullin 756 783 +27 +10
60. Kevin Thurlow 769 783 +14 +2
63. Wendy Roe 781 782 +1 -8
64. Tim Reypert 773 778 +5 -5
65. Neil Zussman 768 776 +8 -2
65. Shane Roberts 766 776 +10 -1
67. Steven Moir 763 775 +12 +0
68. Danny Hamilton 761 773 +12 +0
69. John Gray 757 772 +15 +0
70. Michael Macdonald-Cooper 780 771 -9 -14
71. Sweyn Kirkness 765 770 +5 -5
72. Kai Laddiman 756 765 +9 -2
73. Paul Keane 744 762 +18 +5
73. Michael Bowden 739 762 +23 +8
75. David Edwards 737 760 +23 +8
75. Gary Male 750 760 +10 -2
77. Stu Horsey 732 759 +27 +9
77. Nik Von Uexkull 749 759 +10 -3
79. Richard Pay 732 758 +26 +7
80. Martin Gardner 746 757 +11 -5
81. Brian Selway 746 756 +10 -6
81. James Roberts 736 756 +20 +4
83. Nick Wainwright 726 754 +28 +5
84. Heather Styles 737 753 +16 -1
85. Jonathan Coles 746 752 +6 -10
86. Jean Webby 738 751 +13 -4
86. Ross Allatt 741 751 +10 -6
88. Julia Wilkinson 744 737 -7 -10
89. David Von Geyer 724 734 +10 +0
90. Amey Deshpande 718 729 +11 +0
91. James Doohan 702 728 +26 +7
92. Liam Shaw 708 723 +15 +1
93. Tony Warren 712 721 +9 -1
93. Jayne Wisniewski 705 721 +16 +3
93. Brenda Jolley 718 721 +3 -3
96. Carl Williams 708 720 +12 -3
97. Rose Boyle 701 717 +16 +3
98. David Thirlwall 704 714 +10 -1
99. Andy McGurn 702 712 +10 -1
100. Judith Young 707 709 +2 -5
101. Dave Taylor 691 707 +16 +1
102. Chris Marshall 682 704 +22 +3
103. Tia Corkish 697 702 +5 -2
104. Suzi Purcell 686 695 +9 -1
105. Jeffrey Burgin 685 691 +6 -1
106. Steve Wood 675 688 +13 +0
107. Ned Pendleton 664 684 +20 +1
108. Joe Zubaidi 665 672 +7 -1
109. Chris McHenry 643 654 +11 +0
110. Danny Pledger 635 650 +15 +0
The "average points scored per letters round" and "average points scored per numbers round" refer to the average points scored in the appropriate type of round by that player in their heat games only. This average only counts points that were actually scored in the game, not raw scores, so if a player got an eight-letter word but was beaten by their opponent's nine, their score for that round still counts as zero.
I was just looking at this. So the translation of someone's old score into their new format score is Letters*10/11 + Numbers*4/3 + Conundrum. (Jack's description is horrendous.)
For the top 10 on the list, the average increase in score over the octorun is 15.7, for the top 20, it's 16.3, for the top 30 it's 16.2, and for the top 40 it's 16.1. I'm quite happy to say that the average difference is 16, so as a standard translation, the new system is worth 2 points a game.
Re: Ask Graeme?
Posted: Sun Dec 08, 2013 8:35 pm
by JackHurst
Graeme Cole wrote:sean d wrote:Forgive me if this has been dealt with already- I know it was discussed before but I can't find it.
My question relates to the relative difficulty of Countdown and Apterous. Do we have figures for average max particularly for letters rounds, as I suspect letters rounds are considerably 'nicer' on the show.
For all letters rounds for which the max is known, up to 1st March 2013...
3: 0.0042% (2)
4: 0.0933% (44)
5: 1.7800% (839)
6: 14.4985% (6834)
7: 43.4360% (20474)
8: 34.5723% (16296)
9: 5.6156% (2647)
For all 15-round games up to that date...
3: 0.0035% (1)
4: 0.0315% (9)
5: 1.2257% (350)
6: 12.5263% (3577)
7: 42.5795% (12159)
8: 37.2496% (10637)
9: 6.3839% (1823)
Since 2010...
3: 0.0128% (1)
4: 0.0511% (4)
5: 1.6471% (129)
6: 14.2237% (1114)
7: 43.4755% (3405)
8: 34.8059% (2726)
9: 5.7840% (453)
I don't see any particularly strong deviation from the apterous figures Andy's given. Note that the wiki, and therefore the database from which these figures are derived, gives letters maxes according to ODE3 for games since S64, and ODE2r for all earlier series. This means the letters maxes for series which used earlier editions (series 54 and earlier) might be wrong. A bit.
JackHurst wrote:Good analysis Graeme, but could you improve it by comparing to apterous when taking into account number of vowels picked (i.e compare the figures for 3, 4 and 5 vowel selections separately)? I feel like the average max on the show is lowered somewhat by the number of contestants going 3 vowel all of the time.
I don't think this question of mine was answered. The consensus from the information given seemed to be that letters rounds on the show yield a slightly larger max on average. I think if we looked at the data in the way that I suggested then we'd see a bigger difference.
Re: Ask Graeme?
Posted: Mon Dec 09, 2013 5:38 pm
by sean d
On the Apto v Countdown thing, I'd be interested to see how many letters rounds feature 9 different letters in each format I expect it's considerably higher on the show.
Re: Ask Graeme?
Posted: Mon Dec 09, 2013 10:27 pm
by Graeme Cole
Hey everyone, I'm back. Now I'm going to reply to posts that are nearly a month old.
sean d wrote:Thanks Graeme. That's a big and increasing average margin of victory. A couple of factors, I suppose, are the increase in the number of really good top end players and possibly a drop off in the quality at the other end.... certainly they seem to be finding it harder to find the 200+ contestants required per year.
I'm not sure there's necessarily been a drop-off in contestant quality. The new-15 stats are only based on about four months' worth of shows, as they only go up to the end of series 68. Also, I think the greater average margin in the new format, if that does carry on into series 69 and beyond, is probably more to do with the extra numbers round. In a numbers round you've got a good chance of gaining 10 points over your opponent. If you're a 4-large expert against an average opponent, for example, the chance is quite high that you'll win the round 10-0.
Re: Ask Graeme?
Posted: Mon Dec 09, 2013 11:01 pm
by Graeme Cole
Jon O'Neill wrote:Graeme Cole wrote:Here's the first bit - 15 round octochamps ordered by the number of maxes out of 120 in heat games. A maxed round is where the player got the most points available. Tiebreaks aren't counted.
The 9-rounder one is going to need a bit more thought to exclude people who had more than one run, so I'll come back to that one another day.
Code: Select all
1. Edward McCullagh 95
2. Julian Fell 94
3. Jack Hurst 91
4. Craig Beevers 89
4. Jonathan Rawlinson 89
6. Stewart Holden 86
7. Kirk Bevins 85
8. Chris Davies 84
9. David O'Donnell 82
10. Eoin Monaghan 81
11. Conor Travers 80
12. Chris Wills 79
12. Graeme Cole 79
14. Adam Gillard 78
14. George Greenhough 78
14. John Mayhew 78
14. Oliver Garner 78
14. Paul Gallen 78
19. Jack Welsby 77
20. Andrew Hulme 76
20. Innis Carson 76
20. Paul Howe 76
23. Peter Lee 75
24. Daniel Pati 74
25. Marcus Hares 73
26. Jimmy Gough 72
26. Martin Bishop 72
28. Chris Cummins 71
28. Ryan Taylor 71
28. Tom Barnes 71
31. Aaron Webber 70
31. David Barnard 70
31. Grace Page 70
34. Jon Corby 69
34. Richard Heald 69
36. Lee Hartley 68
36. Matthew Shore 68
38. Charlie Reams 67
38. Mark Deeks 67
38. Mark Tournoff 67
38. Neil Zussman 67
38. Paul James 67
38. Steven Briers 67
44. Jack Worsley 66
45. Andy McGurn 65
45. James Hurrell 65
45. Jeffrey Hansford 65
45. John Brackstone 65
45. Tom Rowell 65
50. Stuart Earl 64
50. Tom Hargreaves 64
52. John Hunt 63
52. Kevin Thurlow 63
52. Scott Gillies 63
55. John Davies 62
55. Junaid Mubeen 62
55. Stuart Solomons 62
55. Wendy Roe 62
59. Jean Webby 61
59. Jon O'Neill 61
59. Martin Gardner 61
62. Richard Brittain 60
62. Shane Roberts 60
64. Cate Henderson 59
64. John Gray 59
64. Jonathan Coles 59
64. Michael Bowden 59
64. Ross Allatt 59
64. Steven Moir 59
70. Brian Selway 58
70. David Edwards 58
70. Mike Pullin 58
70. Stu Horsey 58
74. Danny Hamilton 57
74. Jim Bentley 57
74. Keith Maynard 57
74. Liam Shaw 57
74. Rose Boyle 57
79. Nik Von Uexkull 56
80. Richard Pay 55
80. Rupert Stokoe 55
82. David Von Geyer 54
82. Jeffrey Burgin 54
82. Sweyn Kirkness 54
82. Tim Reypert 54
86. James Roberts 53
86. Kai Laddiman 53
86. Nick Wainwright 53
86. Paul Keane 53
90. Amey Deshpande 52
90. Jayne Wisniewski 52
90. Ned Pendleton 52
93. Carl Williams 51
93. Dave Taylor 51
93. Heather Styles 51
93. Michael Macdonald-Cooper 51
97. Gary Male 50
97. Judith Young 50
97. Julia Wilkinson 50
100. Danny Pledger 49
101. Tony Warren 48
102. James Doohan 46
103. Joe Zubaidi 44
104. Brenda Jolley 43
104. Tia Corkish 43
106. Chris Marshall 42
106. David Thirlwall 42
108. Chris McHenry 37
108. Steve Wood 37
108. Suzi Purcell 37
Edited to include Rose Boyle, David Barnard and Heather Styles from series 67, and Kevin Thurlow and Richard Pay who weren't in the list due to muppetry on my part.
Sorry for asking for an update, but I'd love to see this one up-to-date. Also if possible with what series they were from or what date they debuted
I'll post an update of this table after I've put series 69 into the database, which will be after the end of series 69. The executive summary is that Dylan and Jen now top the table, in that order.
Re: Ask Graeme?
Posted: Mon Dec 09, 2013 11:31 pm
by Graeme Cole
JackHurst wrote:Which contestant who played all of their games in the old format would have the biggest difference between their total score then, and their total score converted into new money (ie their numbers score multiplied by 4/3 and their letters score multiplied by 10/11)?
Jack Worsley. He got 818 points in eight games, which comprised 541 letters points, 237 numbers points and 40 conundrum points. With your formula, this would translate to 847.8 under the new format.
Full list, ordered by points gained by multiplying letters score by 11/10 and numbers score by 4/3:
Code: Select all
OLD NEW DIFF
1. Jack Worsley 818 847.82 +29.82
2. John Davies 766 793.76 +27.76
3. Nick Wainwright 726 753.52 +27.52
4. Mike Pullin 756 783.45 +27.45
5. Jon O'Neill 804 831.39 +27.39
6. Junaid Mubeen 790 817.03 +27.03
7. Jim Bentley 756 782.61 +26.61
7. Stu Horsey 732 758.61 +26.61
9. Chris Cummins 858 884.36 +26.36
9. Mark Deeks 824 850.36 +26.36
11. Richard Pay 732 757.88 +25.88
11. James Doohan 692 717.88 +25.88
13. Scott Gillies 810 834.55 +24.55
14. Ryan Taylor 792 816.36 +24.36
15. John Hunt 778 801.27 +23.27
15. Edward McCullagh 896 919.27 +23.27
17. David Edwards 737 760.24 +23.24
18. Michael Bowden 739 761.88 +22.88
19. Steven Briers 843 865.58 +22.58
20. Tom Rowell 774 796.30 +22.30
21. Chris Marshall 682 704.18 +22.18
22. Paul Howe 815 836.67 +21.67
23. Eoin Monaghan 898 918.97 +20.97
24. Craig Beevers 907 927.64 +20.64
25. Ned Pendleton 664 684.30 +20.30
26. Graeme Cole 813 833.03 +20.03
27. Richard Brittain 820 839.76 +19.76
28. James Roberts 736 755.52 +19.52
29. Jack Welsby 831 850.30 +19.30
30. David O'Donnell 880 898.79 +18.79
31. Adam Gillard 903 921.03 +18.03
32. Paul Keane 744 762.00 +18.00
33. Tom Hargreaves 850 867.64 +17.64
33. Oliver Garner 802 819.64 +17.64
35. Chris Wills 875 892.61 +17.61
36. Jack Hurst 946 963.45 +17.45
37. Paul Gallen 846 863.15 +17.15
38. Jonathan Rawlinson 850 866.97 +16.97
39. Jon Corby 856 872.73 +16.73
40. Cate Henderson 782 798.24 +16.24
41. Dave Taylor 691 707.21 +16.21
41. Jayne Wisniewski 705 721.21 +16.21
41. Rose Boyle 701 717.21 +16.21
44. Stuart Earl 807 823.15 +16.15
45. Mark Tournoff 809 824.97 +15.97
46. Heather Styles 737 752.91 +15.91
47. David Barnard 771 786.85 +15.85
48. Marcus Hares 834 849.33 +15.33
49. Aaron Webber 773 788.30 +15.30
50. Kirk Bevins 925 940.12 +15.12
51. John Gray 757 772.00 +15.00
52. Conor Travers 890 904.97 +14.97
53. Danny Pledger 635 649.94 +14.94
54. Liam Shaw 708 722.79 +14.79
55. Kevin Thurlow 769 782.85 +13.85
56. Jean Webby 738 751.45 +13.45
57. Richard Heald 795 808.24 +13.24
58. Steve Wood 675 688.12 +13.12
59. Martin Bishop 809 821.88 +12.88
60. Lee Hartley 811 823.82 +12.82
61. Andrew Hulme 930 942.61 +12.61
62. Danny Hamilton 761 772.85 +11.85
63. Carl Williams 708 719.76 +11.76
64. Rupert Stokoe 776 787.64 +11.64
65. Steven Moir 763 774.61 +11.61
66. Martin Gardner 746 757.33 +11.33
67. John Brackstone 822 833.21 +11.21
67. James Hurrell 838 849.21 +11.21
69. Charlie Reams 820 831.03 +11.03
70. Chris McHenry 643 653.88 +10.88
71. Amey Deshpande 718 728.55 +10.55
72. Nik Von Uexkull 749 759.33 +10.33
73. David Thirlwall 704 714.12 +10.12
74. Shane Roberts 766 776.06 +10.06
75. Gary Male 750 760.00 +10.00
76. Innis Carson 861 870.97 +9.97
77. Andy McGurn 702 711.76 +9.76
78. George Greenhough 817 826.73 +9.73
79. Brian Selway 746 755.70 +9.70
80. Ross Allatt 741 750.67 +9.67
81. David Von Geyer 724 733.64 +9.64
82. Suzi Purcell 686 695.27 +9.27
83. Kai Laddiman 756 764.97 +8.97
84. Grace Page 829 837.94 +8.94
85. Stuart Solomons 796 804.85 +8.85
86. Jimmy Gough 782 790.79 +8.79
87. Tom Barnes 822 830.73 +8.73
88. Tony Warren 712 720.67 +8.67
89. Chris Davies 892 900.42 +8.42
90. Neil Zussman 758 766.24 +8.24
91. Keith Maynard 785 792.79 +7.79
92. Joe Zubaidi 665 672.24 +7.24
93. Jeffrey Burgin 685 690.67 +5.67
94. Julian Fell 924 929.58 +5.58
95. Jonathan Coles 746 751.52 +5.52
96. Sweyn Kirkness 765 770.36 +5.36
97. Paul James 794 799.33 +5.33
98. Matthew Shore 850 854.91 +4.91
99. Tim Reypert 773 777.76 +4.76
99. Tia Corkish 697 701.76 +4.76
101. Stewart Holden 870 874.48 +4.48
102. Peter Lee 801 805.33 +4.33
103. Daniel Pati 840 843.45 +3.45
104. Brenda Jolley 718 721.39 +3.39
105. Judith Young 707 708.85 +1.85
106. Wendy Roe 781 782.06 +1.06
107. John Mayhew 811 810.24 -0.76
108. Jeffrey Hansford 818 812.79 -5.21
109. Julia Wilkinson 744 737.21 -6.79
110. Michael Macdonald-Cooper 780 771.27 -8.73
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 3:24 am
by Dave Preece
I'm assuming the bottom few are very good conundrum/average numbers round contestants???
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 3:25 am
by Dave Preece
The top few are average conundrum and top, top numbers round players compared to letter rounds???
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 3:26 am
by Dave Preece
This would explain why Conor is in the middle, a jack of no trades and a master of all???
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 3:28 am
by Dave Preece
I am surprised Jack is 36th on this list... Neither near top, middle or bottom???
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 9:40 am
by Jon Corby
Dave Preece wrote:I am surprised Jack is 36th on this list... Neither near top, middle or bottom???
A Conor of no trades and a master of all?
Re: Ask Graeme?
Posted: Tue Dec 10, 2013 9:43 am
by Gavin Chipper
Dave Preece wrote:I'm assuming the bottom few are very good conundrum/average numbers round contestants???
Conundrums don't make a difference. But yeah, more likely to be people who are shit at numbers.
Re: Ask Graeme?
Posted: Sat Dec 14, 2013 2:39 pm
by Fred Mumford
Has a player ever missed out on a finals place as a result of winning the final preliminary game of a series? (eg. player finishes series x with several wins, but loses the first game of series y and misses the series y finals, when losing the last game in series x would have got them into the series x finals).
Re: Ask Graeme?
Posted: Sat Dec 14, 2013 6:24 pm
by Dave Preece
Fred Mumford wrote:Has a player ever missed out on a finals place as a result of winning the final preliminary game of a series? (eg. player finishes series x with several wins, but loses the first game of series y and misses the series y finals, when losing the last game in series x would have got them into the series x finals).
I doubt it.
Re: Ask Graeme?
Posted: Sat Dec 14, 2013 6:31 pm
by Jennifer Steadman
Had one of the top 8 been unable to play in the Series 65 finals, that would've happened (Phyl Styles would've played instead of Matt Croy, having lost to some no-hoper called Jack Worsley who we never heard of again in the final prelim of the series). Dunno if it's happened properly but it wouldn't surprise me. Assuming of course that this isn't including people who've become octochamps in the last prelim of the series, but on accumulated wins and points were already in the top 8 prior to that game - if it is, then this would be the case in series 66, among others.
Re: Ask Graeme?
Posted: Sat Dec 14, 2013 6:31 pm
by Gavin Chipper
Dave Preece wrote:Fred Mumford wrote:Has a player ever missed out on a finals place as a result of winning the final preliminary game of a series? (eg. player finishes series x with several wins, but loses the first game of series y and misses the series y finals, when losing the last game in series x would have got them into the series x finals).
I doubt it.
I think it would have come up in discussions because that sort of thing has been discussed.
Re: Ask Graeme?
Posted: Sat Dec 14, 2013 7:59 pm
by Fred Mumford
Jennifer Steadman wrote:Assuming of course that this isn't including people who've become octochamps in the last prelim of the series
Thanks Jen. Correct assumption - I was specifically wondering about contestants who straddled series, not octochamps who completed their run at the very end of a series.
Re: Ask Graeme?
Posted: Sun Dec 15, 2013 10:12 pm
by Martin Thompson
You could win the last 6 games of a series to put you 8th in the list, but then lose the first in the next series and come 9th in the rankings. But if you hadn't won that 6th game you wouldn't have qualified anyway. So really it would have to be 7 wins and losing the next one, but you must surely always qualify with that?
Re: Ask Graeme?
Posted: Sun Dec 15, 2013 10:25 pm
by Dave Preece
As I said... I doubt it!
Re: Ask Graeme?
Posted: Sun Dec 15, 2013 11:53 pm
by Gavin Chipper
Jennifer Steadman wrote:Had one of the top 8 been unable to play in the Series 65 finals, that would've happened (Phyl Styles would've played instead of Matt Croy, having lost to some no-hoper called Jack Worsley who we never heard of again in the final prelim of the series). Dunno if it's happened properly but it wouldn't surprise me. Assuming of course that this isn't including people who've become octochamps in the last prelim of the series, but on accumulated wins and points were already in the top 8 prior to that game - if it is, then this would be the case in series 66, among others.
I'm not sure I follow. So who would have missed out on the basis of winning a game?
Re: Ask Graeme?
Posted: Mon Dec 16, 2013 1:05 pm
by Dave Preece
Oops!
Re: Ask Graeme?
Posted: Sat Dec 21, 2013 12:43 pm
by Mark Deeks
Martin Sinclair wrote:I've just done some research on something which I'm sure many will have briefly thought about doing (apologies if it already has and I'm an idiot). It's the calculation for what your score would've been if your opponent maxed every single round throughout. So, the way to do this is by taking all the rounds which you did score on (but didn't max on), and subtract the score which got in those rounds from your overall total. You can ignore any rounds which you failed to score on altogether. Also, I feel all the points which you scored on in the conundrum rounds need to be subtracted from the total, as your opponent could've hansforded you each time. Harsh, but true. I've looked up a few impressive octoruns, and here's how it currently stands.
1) Dylan Taylor, 823 (151 points lost). Octorun maximum: 1050
2) Jen Steadman, 786 (166 points lost). Octorun maximum: 1024
3) Giles Hutchings, 748 (217 points lost). Octorun maximum: 1060
4) Edward McCullagh, 744 (152 points lost). Octorun maximum: 979
5) Julian Fell, 737 (187 point lost). Octorun maximum:
However, some octoruns obviously have higher scores available in theirs than others, so perhaps it would be best to judge it by the lowest amount of points lost. As you can see, Dylan's still top. Another good way would be to measure the percentage achieved from your maximum with your new "impossible to get lower" score. With this, here's the new leaderboard:
1) Dylan Taylor: 78.380952381%
2) Jen Steadman: 76.7578125%
3) Edward McCullagh: 75.9959141982%
4) Jack Hurst: 71.4839961203%
5) Julian Fell: 71.4285714286%
Martin, as a staunch Dylan supporter, do you feel that perhaps this analysis of his performance, which selectively ignores conundrums, is perhaps biased in his favour, given that he was the weakest conundrum player of the list? This ultimately proved to be his undoing and I would be interested to see how you incorporate this.
Re: Ask Graeme?
Posted: Sat Dec 21, 2013 11:36 pm
by Martin Sinclair
You are right, I am a big fan of Dylan. Still, that has nothing to do with the conundrum rounds being eliminated altogether, as the thing with the timing of them makes it extremely difficult to measure. And, I did do a leaderboard (I think) of what the "impossible to get lower than" scores would have been if conundrums were included, and he's still top.
Re: Ask Graeme?
Posted: Sat Dec 21, 2013 11:39 pm
by Martin Sinclair
... Or he was for his octorun. I'll have to do the further calculations now.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 4:08 pm
by Mark Deeks
I appreciate that, yet it seems like an arbitrary distinction. Indeed, the entire metric seems somewhat arbitrary. From this end, it seems like a convoluted means of finding ways to list Dylan at the top of something, via what ever criteria are necessary. Therefore, in light of all that which your formulae have uncovered, as well being mindful of the outcome of series 69, Martin, how would you construct a Dave Preece-like list of the best players of all time?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 4:19 pm
by Zarte Siempre
Mark Deeks wrote:I appreciate that, yet it seems like an arbitrary distinction. Indeed, the entire metric seems somewhat arbitrary. From this end, it seems like a convoluted means of finding ways to list Dylan at the top of something, via what ever criteria are necessary. Therefore, in light of all that which your formulae have uncovered, as well being mindful of the outcome of series 69, Martin, how would you construct a Dave Preece-like list of the best players of all time?
I'm really enjoying your style, Mark
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 4:28 pm
by Martin Sinclair
My top 3 would be:
1) Conor Travers
2) Dylan Taylor
3) Jack Hurst
I feel that winning a series doesn't mean much, as Dylan has better statistics than any other winner of a regular series, but did he win it? No. Just look, he has the highest amount of maxes from a series by some way - 140/65, and the second best is 131/165. But, he was far from the best conundrum player, and Conor's CoC performance was just out of this world. He has two more trophies than him, so is #1 in my eyes. Jack's #3 because I feel he has the 2nd best statistics in a regular series, and he won his.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 4:51 pm
by Mark Deeks
Again though, Martin, that feels strongly like a contrived metric. To say that Dylan's 140 out of 165 is of more importance than winning a series in determining greatness is to say that meeting the threshold of an arbitrary, strained, artifical measurement is more valuable than beating the opponent. The maxes count is relevant only to those who choose to measure it, not to the field in which performance is measured, which is solely that of winning games. Therefore, if greatness is measured by achievement, then surely that which the game itself ascribes as the ultimate achievement - winning the bugger - should be the most valuable determination of all?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:01 pm
by Mark James
Martin Sinclair wrote:
I feel that winning a series doesn't mean much,
This is the kind of stupid thing Liverpool haters (I'm not a Liverpool fan btw) were saying in 2005. "Sure they won the competition but they weren't the best team in it". Well they were. That's what winning the thing means. As someone who participated in this years series all I cared about was getting my name on the trophy. I wasn't worried about max games or points totals.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:10 pm
by Martin Sinclair
Well, this brings me to another point of mine - eras. If Dylan, or Jen, had been on a few years ago at the same standard now, then they'd have won their series, for sure. So they're unlucky that they were both in a solid series. Also, I've found the perfect reason as to why collection of maxes is a better way of measuring than % of max, but this is getting a bit off-topic, unless if you want me to say why that method's better?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:14 pm
by Mark Deeks
Don't be afraid. Post it. I'd like to hear it!
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:37 pm
by Martin Sinclair
Example: two people play an octorun which has exactly the same maximum, in this case, it's 1000. Player 1 scores 964 and gets 118/120 maxes. Player 2 scores 965 and gets 85/120 maxes. As you can see, the only way in which Player 1 could have dropped 36 points from two rounds is if he/she had failed to score in both of those rounds, and that there was a nine available both times. However, Player 2 can be let of the hook much more easily. He/she could've afforded to drop maxes on 35 rounds and still get a higher score than Player 1, and therefore a higher % of max, with the maxes both being the same. So, Player 1 did better in 35 more rounds but still ended up on the losing side to the best % of max.
... And that's why maxes total is a better way of measuring.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:38 pm
by Martin Sinclair
and therefore have a higher % of max*
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:39 pm
by Mark Deeks
Why's Dylan not the best ever, then?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:41 pm
by Zarte Siempre
Martin Sinclair wrote:Example: two people play an octorun which has exactly the same maximum, in this case, it's 1000. Player 1 scores 964 and gets 118/120 maxes. Player 2 scores 965 and gets 85/120 maxes. As you can see, the only way in which Player 1 could have dropped 36 points from two rounds is if he/she had failed to score in both of those rounds, and that there was a nine available both times. However, Player 2 can be let of the hook much more easily. He/she could've afforded to drop maxes on 35 rounds and still get a higher score than Player 1, and therefore a higher % of max, with the maxes both being the same. So, Player 1 did better in 35 more rounds but still ended up on the losing side to the best % of max.
... And that's why maxes total is a better way of measuring.
I feel like I've heard this argument somewhere before... I can't QUITE put my finger on it though...
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:42 pm
by Martin Sinclair
Because he was trophyless (I don't feel that's the #1 factor) and just look at Conor's amount of maxes. I think it was 86/90, which overshadows the rest. Amazing.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:46 pm
by Mark Deeks
So what makes Jack Hurst third? Giles Hutchings, for example, recorded 94 maxes and won the series. Edward McCullagh recorded 95 and won his series. Why do you feel as though Jack Hurst had the second best statistics? Where do your yardsticks lie here, Martin?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:49 pm
by Martin Sinclair
Because he did better in the CoC than Edward, and Giles probably shouldn't have won his series final, statistically.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:50 pm
by Mark Deeks
Giles probably shouldn't have won his series final, statistically.
Why do you say that, Martin?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:54 pm
by Zarte Siempre
Mark Deeks wrote:So what makes Jack Hurst third? Giles Hutchings, for example, recorded 94 maxes and won the series. Edward McCullagh recorded 95 and won his series. Why do you feel as though Jack Hurst had the second best statistics? Where do your yardsticks lie here, Martin?
Yeah, that's a good question. See, to me, people remember Giles and Ed and Jack. I feel that whilst people on here would remember Dylan for breaking the all-time points record, the average-joe at home who just enjoys watching along probably wouldn't remember anyone who hadn't won a series for more than the next series or so. They might remember "That guy who lost in the final of the series where there were massive upsets" but that's about as much as he's likely to get I think.
I've got the highest score of series 69, but am I deluded enough to think anyone outside the kind of person likely to hang around this forum would know that or care in the slightest beyond the couple of weeks of attention it got me on Twitter at the time? Not at all.
So yeah, in essence I disagree with your logic Martin (though I respect your opinions and the work you've clearly put in).
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:55 pm
by Martin Sinclair
Andy was very unlucky that ODORIZER wasn't in. It used to be and is set to be added again next year. He only got 11 maxes in his final, Jack H got 14. Also, I don't class a numbers game where you get 7/5 points but a better 7/5 points is available, and I know Giles had a few of these in his octorun.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 5:57 pm
by Martin Sinclair
7/5 points is available as a max*
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:03 pm
by Martin Sinclair
But yeah, I can totally understand why some of you don't feel Dylan should be #1. It's just like the seeding system with Countdown - more wins first, then points.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:06 pm
by Mark Deeks
Martin Sinclair wrote:But yeah, I can totally understand why some of you don't feel Dylan should be #1.
Nor do you, by the sounds of it, Martin.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:07 pm
by Martin Sinclair
From the neutral's point of view, there's no way he'd me remembered for being the best. From the person who knows much more about Countdown (Apterites etc.), they may see where I'm coming from and agree with me, perhaps. Just remember, if Dylan had been in any regular series before his time, he'd have almost certainly have won it, statistically (this is related to my point about eras).
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:10 pm
by Martin Sinclair
Anyway, we're posting all of this in the wrong thread! It's wasting space so if you'd like to keep debating, I'd suggest putting all future posts in the "Best ever contestants" thread.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:10 pm
by Zarte Siempre
I wish it was easy to agree with you Martin, cause I'd quite like to have only lost to the "greatest player ever" or somewhere close to it, but after losing his final, I don't even know if you could put him in the top 10. I think (though I'm sure someone would disagree with me) as well, that if someone was to make a definitive list of the "greatest" in any field, personality WILL always come into it. Take boxing. Muhammad Ali is regarded as probably the greatest boxer ever. But he actually lost 5 of his fights, a record that puts his nowhere near the best. People loved him and rated him as much for his personality as his skill. Now I don't know Dylan particularly well, but on TV at least he was... bland at best and a personality vacuum at worst. For him to be forever remembered by the ordinary fans of the show out there, I think he'd need to have made them warm to him as well.
But as I say, I'm sure people will disagree with that
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:10 pm
by Mark Deeks
See, to me, people remember Giles and Ed and Jack.
Beautiful people, too. Certainly amongst the most attractive contestants of all time, if not the most.
Martin, Giles did all that he had to do to win. He outscored his opponent in the final, Martin. It doesn't matter, Martin, that he didn't outscore Jack's opponent. Remember also, Martin, that not all missed maxes are created equal. There's a difference between missing TOASTED and missing NAPROXEN, Martin. Do your max-centric analyses, Martin, account for this?
Also, Martin, your citing of the "eras" is interesting. Different eras, Martin, produce different quality players. But they do so in part, Martin, because of the means available for the contestants to practice. Martin, look at another all-time great, Kirk Bevins. Kirk was on twice, remember Martin - first in 2004, and then in 2009. Does his three max loss change his subsequent greatness in your eyes, Martin?
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 6:16 pm
by Zarte Siempre
Mark Deeks wrote:See, to me, people remember Giles and Ed and Jack.
Beautiful people, too.
Absolutely.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 7:12 pm
by JackHurst
FUCK YES I AM #3
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 7:30 pm
by Graeme Cole
Graeme Cole wrote:Jon O'Neill wrote:Graeme Cole wrote:Here's the first bit - 15 round octochamps ordered by the number of maxes out of 120 in heat games. A maxed round is where the player got the most points available. Tiebreaks aren't counted.
Sorry for asking for an update, but I'd love to see this one up-to-date. Also if possible with what series they were from or what date they debuted
I'll post an update of this table after I've put series 69 into the database, which will be after the end of series 69. The executive summary is that Dylan and Jen now top the table, in that order.
Updated to series 69...
Code: Select all
SERIES DEBUT MAXES
1. Dylan Taylor 69 2013-08-06 101
2. Jen Steadman 69 2013-08-30 99
3. Edward McCullagh 64 2011-02-02 95
4. Julian Fell 48 2002-10-16 94
5. Giles Hutchings 68 2013-04-03 93
6. Jack Hurst 63 2010-10-13 91
7. Craig Beevers 57 2007-10-23 89
7. Jonathan Rawlinson 66 2012-06-11 89
9. Andy Platt 68 2013-03-19 87
10. Stewart Holden 51 2004-02-23 86
11. Kirk Bevins 60 2009-03-02 85
12. Chris Davies 61 2009-08-28 84
12. Glen Webb 69 2013-06-05 84
14. David O'Donnell 58 2008-01-24 82
14. Bradley Cates 69 2013-10-18 82
16. Eoin Monaghan 63 2010-09-15 81
17. Conor Travers 54 2005-11-23 80
18. Chris Wills 47 2002-01-14 79
18. Graeme Cole 65 2011-06-07 79
20. George Greenhough 48 2002-11-01 78
20. Paul Gallen 52 2004-08-31 78
20. John Mayhew 53 2005-02-17 78
20. Oliver Garner 62 2010-01-12 78
20. Adam Gillard 64 2011-01-12 78
25. Jack Welsby 52 2004-07-21 77
26. Paul Howe 54 2006-02-04 76
26. Innis Carson 61 2009-06-25 76
26. Andrew Hulme 61 2009-07-09 76
29. Peter Lee 66 2012-02-20 75
30. Daniel Pati 63 2010-09-03 74
31. Marcus Hares 63 2010-11-01 73
32. Martin Bishop 59 2008-11-04 72
33. Chris Cummins 50 2003-07-21 71
33. Jimmy Gough 60 2009-05-11 71
33. Ryan Taylor 61 2009-11-05 71
33. Tom Barnes 64 2011-05-10 71
37. Grace Page 48 2002-07-02 70
37. Aaron Webber 56 2007-04-23 70
37. David Barnard 67 2012-11-20 70
40. Richard Heald 53 2005-06-02 69
40. Jon Corby 54 2006-04-28 69
42. Lee Hartley 47 2002-02-18 68
42. Matthew Shore 54 2006-02-17 68
44. Mark Tournoff 52 2004-11-15 67
44. Steven Briers 55 2006-11-03 67
44. Charlie Reams 59 2008-10-15 67
44. Neil Zussman 60 2009-02-16 67
44. Mark Deeks 65 2011-07-12 67
44. Paul James 67 2012-07-06 67
50. Jack Worsley 66 2011-12-07 66
51. John Brackstone 53 2005-05-12 65
51. James Hurrell 57 2007-06-07 65
51. Jeffrey Hansford 57 2007-08-27 65
51. Tom Rowell 63 2010-07-22 65
51. Andy McGurn 64 2011-03-08 65
56. Tom Hargreaves 47 2001-12-26 64
56. Stuart Earl 50 2003-11-14 64
58. Kevin Thurlow 47 2002-01-28 63
58. John Hunt 52 2004-10-14 63
58. Scott Gillies 63 2010-06-03 63
58. Alex Fish 69 2013-09-13 63
62. Wendy Roe 47 2002-04-23 62
62. John Davies 49 2003-05-02 62
62. Stuart Solomons 50 2003-10-28 62
62. Junaid Mubeen 59 2008-09-22 62
66. Martin Gardner 49 2003-05-14 61
66. Jon O'Neill 53 2005-04-01 61
66. Jean Webby 56 2007-02-26 61
69. Richard Brittain 55 2006-09-11 60
69. Shane Roberts 60 2009-04-09 60
71. John Gray 52 2004-09-27 59
71. Steven Moir 52 2004-11-03 59
71. Michael Bowden 54 2006-01-25 59
71. Jonathan Coles 58 2008-05-27 59
71. Cate Henderson 60 2009-03-23 59
76. Mike Pullin 47 2002-05-10 58
76. Ross Allatt 53 2005-04-14 58
76. Stu Horsey 55 2006-10-05 58
76. David Edwards 57 2007-07-13 58
76. Brian Selway 61 2009-10-19 58
76. Eileen Taylor 68 2013-04-24 58
82. Danny Hamilton 48 2002-07-24 57
82. Jim Bentley 50 2003-08-28 57
82. Keith Maynard 54 2006-01-16 57
82. Liam Shaw 67 2012-09-17 57
82. Rose Boyle 67 2012-10-15 57
87. Nik Von Uexkull 51 2004-04-21 56
88. Rupert Stokoe 47 2002-03-25 55
88. Richard Pay 51 2004-02-06 55
90. Sweyn Kirkness 51 2004-05-05 54
90. David Von Geyer 57 2007-09-17 54
90. Tim Reypert 58 2008-02-08 54
90. Jeffrey Burgin 61 2009-09-16 54
94. Nick Wainwright 56 2007-05-07 53
94. James Roberts 57 2007-11-21 53
94. Kai Laddiman 59 2008-10-02 53
94. Paul Keane 65 2011-11-10 53
94. Jonathan Liew 69 2013-11-21 53
99. Amey Deshpande 56 2007-01-16 52
99. Ned Pendleton 64 2011-02-17 52
99. Jayne Wisniewski 65 2011-09-23 52
102. Michael Macdonald-Cooper 58 2008-01-08 51
102. Dave Taylor 65 2011-08-23 51
102. Carl Williams 65 2011-09-06 51
102. Heather Styles 67 2012-11-07 51
106. Julia Wilkinson 48 2002-06-06 50
106. Gary Male 51 2004-03-23 50
106. Judith Young 53 2005-01-21 50
106. Joe McGonigle 68 2013-05-13 50
110. Danny Pledger 62 2010-05-14 49
111. Tony Warren 55 2006-05-10 48
112. James Doohan 60 2009-05-22 46
113. Joe Zubaidi 48 2002-11-21 44
114. Brenda Jolley 47 2002-03-01 43
114. Tia Corkish 67 2012-10-03 43
116. David Thirlwall 52 2004-06-08 42
116. Chris Marshall 67 2012-07-18 42
118. Chris McHenry 55 2006-11-22 37
118. Steve Wood 61 2009-10-06 37
118. Suzi Purcell 66 2012-05-24 37
As before, a maxed round is where the player got the maximum points available in that round. So if in a numbers round the best possible was 1 away, and you got 2 away, it's still a max. This is how apterous defines a max now.
"Debut" is the date of the first game in the player's octorun, and because some players had already appeared before, this might not be the same as their actual debut. The "series" column is the series in which their last heat game fell - in other words, which finals they qualified for.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 7:33 pm
by Martin Sinclair
I feel the numbers where you got 2 away but you could have got away as a max is up to opinion, as I know a few people don't think it should be classed as a max (I'm one of them).
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 7:46 pm
by Zarte Siempre
Martin Sinclair wrote:I feel the numbers where you got 2 away but you could have got away as a max is up to opinion, as I know a few people don't think it should be classed as a max (I'm one of them).
You're wrong.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 7:53 pm
by Gavin Chipper
Martin Sinclair wrote:I feel the numbers where you got 2 away but you could have got away as a max is up to opinion, as I know a few people don't think it should be classed as a max (I'm one of them).
I think Graeme does it that way because it's easier for him to calculate.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 8:19 pm
by Michael Wallace
I've been having a think about the high scores 'record' thing, as it's a moderately interesting problem. Obviously the basic conclusion is "Jack has the old-15 record, Dylan has the new-15 one", because a straight comparison is always going to involve some subjectivity. However, I thought a moderately neat (and crucially, easy) way to do a direct comparison would be to take Jack's octorun and say "what if one of his letters rounds was replaced by a numbers round in each of his shows?".
To do this I generated a bunch of 'pseudo' octoruns where in each game a randomly chosen letters round was replaced by a score picked from Jack's numbers round scores on the show (i.e. assuming his performance on these hypothetical rounds match his performance on the ones that actually happened). Unfortunately (although predictably) this proves inconclusive; Dylan and Jack are inseparable from a statistical perspective using this approach as it spits out a likely range for Jack's estimated octotal of 942-980, which comfortably includes Dylan's 974. (For the more statistically minded, that's a 95% confidence interval.)
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 8:26 pm
by Gavin Chipper
Michael Wallace wrote:I've been having a think about the high scores 'record' thing, as it's a moderately interesting problem. Obviously the basic conclusion is "Jack has the old-15 record, Dylan has the new-15 one", because a straight comparison is always going to involve some subjectivity. However, I thought a moderately neat (and crucially, easy) way to do a direct comparison would be to take Jack's octorun and say "what if one of his letters rounds was replaced by a numbers round in each of his shows?".
To do this I generated a bunch of 'pseudo' octoruns where in each game a randomly chosen letters round was replaced by a score picked from Jack's numbers round scores on the show (i.e. assuming his performance on these hypothetical rounds match his performance on the ones that actually happened). Unfortunately (although predictably) this proves inconclusive; Dylan and Jack are inseparable from a statistical perspective using this approach as it spits out a likely range for Jack's estimated octotal from 942-980, which comfortably includes Dylan's 974. (For the more statistically minded, that's a 95% confidence interval.)
That is quite interesting. Normally I'd say do the letters*10/11 + numbers*4/3 + conundrum score to convert from old to new, or do the add two points per game rule of thumb. But if we're looking at confidence intervals, perhaps we should really do this for everyone. Jack's score converts to slightly less than Dylan's using the methods I've suggested. But once that's done, the problem here isn't old v new - it's statistical variation, so people scoring close to Dylan and indeed Dylan himself could be given confidence intervals treatment.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 8:29 pm
by Graeme Cole
JackHurst wrote:
JackHurst wrote:Good analysis Graeme, but could you improve it by comparing to apterous when taking into account number of vowels picked (i.e compare the figures for 3, 4 and 5 vowel selections separately)? I feel like the average max on the show is lowered somewhat by the number of contestants going 3 vowel all of the time.
I don't think this question of mine was answered. The consensus from the information given seemed to be that letters rounds on the show yield a slightly larger max on average. I think if we looked at the data in the way that I suggested then we'd see a bigger difference.
Code: Select all
All-time
MAX 3 VOWELS 4 VOWELS 5 VOWELS
3 1 (0.004%) 1 (0.004%) 0
4 19 (0.081%) 23 (0.095%) 1 (0.082%)
5 363 (1.549%) 440 (1.825%) 44 (3.601%)
6 3220 (13.74%) 3482 (14.45%) 295 (24.14%)
7 10716 (45.74%) 9915 (41.14$) 515 (42.14%)
8 8098 (34.56%) 8528 (35.38%) 328 (26.84%)
9 1011 (4.315%) 1714 (7.111%) 39 (3.191%)
TOTAL 23428 24103 1222
Since 2010
MAX 3 VOWELS 4 VOWELS 5 VOWELS
3 0 1 (0.017%) 0
4 2 (0.060%) 2 (0.034%) 1 (0.224%)
5 42 (1.254%) 95 (1.613%) 12 (2.691%)
6 443 (13.23%) 797 (13.54%) 103 (23.09%)
7 1535 (45.84%) 2474 (42.02%) 188 (42.15%)
8 1188 (35.47%) 2100 (35.67%) 128 (28.70%)
9 139 (4.150%) 419 (7.116%) 14 (3.139%)
TOTAL 3349 5888 446
As before, the maxes are given by the wiki, and this was given by the recap writer, which doesn't know about dictionaries prior to ODE2r. So for series before the introduction of the ODE2r, the maxes might be slightly wrong.
Can anyone find the statland page on apterous that gives the max breakdown of each vowel pick? I can't.
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 8:38 pm
by Graeme Cole
Gavin Chipper wrote:Martin Sinclair wrote:I feel the numbers where you got 2 away but you could have got away as a max is up to opinion, as I know a few people don't think it should be classed as a max (I'm one of them).
I think Graeme does it that way because it's easier for him to calculate.
It used to be for that reason, but I've got a column in a table now that tells me what the best possible declaration in a numbers game was. Now apterous defines a numbers max the "easy" way, I'll do that here as well. However, it being easier is still part of the reason. When writing a database query, saying "it's a max if score = max score" is still easier than saying "if it was a numbers round, it's a max if abs(contestant's declaration - target) = abs(best declaration - target) and it wasn't disallowed, or if the max was 0; if it wasn't a numbers round, then it's a max if score = max score".
Re: Ask Graeme?
Posted: Sun Dec 22, 2013 8:42 pm
by Michael Wallace
Gavin Chipper wrote:That is quite interesting. Normally I'd say do the letters*10/11 + numbers*4/3 + conundrum score to convert from old to new, or do the add two points per game rule of thumb. But if we're looking at confidence intervals, perhaps we should really do this for everyone. Jack's score converts to slightly less than Dylan's using the methods I've suggested. But once that's done, the problem here isn't old v new - it's statistical variation, so people scoring close to Dylan and indeed Dylan himself could be given confidence intervals treatment.
Yeah, the basic problem is that to do any sort of comparison you're going to have to make assumptions to deal with the 'what if?', which necessarily come with some kind of uncertainty. With relatively few data (you can only really use on-screen scores, imo) I'm sceptical there's any reasonable approach that wouldn't find any difference in estimated scores between Jack and Dylan could likely just be attributed to chance.
Re: Ask Graeme?
Posted: Mon Dec 23, 2013 1:16 am
by Mark Deeks
For the record, Martin Sinclair was Dylan posting under a false name.