Page 23 of 31
Re: Ask Graeme?
Posted: Sun Jun 21, 2020 1:21 pm
by Gavin Chipper
Seems pretty random.
Re: Ask Graeme?
Posted: Sun Jun 21, 2020 4:13 pm
by JackHurst
Thomas Cappleman wrote: ↑Sun Jun 21, 2020 1:04 pm
If I've understood the system correctly, Conor's ended up on the total of everyone's score going into the 30BC? Several of whom (including you) were unbeaten, which in turn means he scoops up several series worth of contestants. As I think an octochamp who goes on to win the series basically ends up with a score roughly equal to the length of the series? So can't see that combination being beaten by anyone.
Yes I think it might work out that way actually. Probably now not as interesting as I thought!
Re: Ask Graeme?
Posted: Sun Jul 05, 2020 9:40 am
by Adam Dexter
Not sure if been covered elsewhere, but here goes.
Which word has the most stems, so should rarely be declared?
Are you able to show the most for each length of word?
Maybe this could be further split to determine words that if any extra vowel was added would yield a longer word - thus meaning your last pick should be a vowel if you already have this word and you're going for a high score.
Re: Ask Graeme?
Posted: Sun Jul 05, 2020 12:08 pm
by Charlie Reams
Adam Dexter wrote: ↑Sun Jul 05, 2020 9:40 am
Not sure if been covered elsewhere, but here goes.
Which word has the most stems, so should rarely be declared?
Are you able to show the most for each length of word?
Maybe this could be further split to determine words that if any extra vowel was added would yield a longer word - thus meaning your last pick should be a vowel if you already have this word and you're going for a high score.
Not sure this is the most, but CANISTER stems with 15 different letters which seems like a good start.
Re: Ask Graeme?
Posted: Sun Jul 05, 2020 12:38 pm
by Conor
ANGRIEST stems with 19 I think. I'd be surprised if that can be beaten.
For which words stem with any vowel:
http://wiki.apterous.org/Fruitful_8s
http://wiki.apterous.org/Fruitful_7s
http://wiki.apterous.org/Fruitful_6s
Re: Ask Graeme?
Posted: Sun Jul 05, 2020 2:06 pm
by Adam Dexter
Ahh very nice - I thought something like this would likely have already been covered.
Any idea if these are updated with dictionary updates?
Re: Ask Graeme?
Posted: Mon Jul 06, 2020 8:49 pm
by Adam Dexter
Who would be the current UCWC? Is anyone keeping track?
A bit like
https://www.ufwc.co.uk/ where the winner of the first match is champion until defeated (a la boxing)
Issue I can see is people who retire unbeaten - not entirely sure how this is dealt with in boxing?
Re: Ask Graeme?
Posted: Mon Jul 06, 2020 8:57 pm
by Callum Todd
I think in boxing they consider someone who retires as lineal champion to be the end of that lineage, and a new one begins with whoever picks up the title after their retirement.
The problem with Countdown is that everyone sort of 'retires' at the same time, so we'd just go through an endless cycle of the lineal champ being the CoC champs + the series winners since then.
So the current lineal champs would be:
(anyone who retired as lineal champ prior to 30BC)
Conor Travers
Dylan Taylor
Zarte Siempre
Mike Daysley
Dinos Sfyris
James Haughton
(any octochamps from current series)
(that octochamp who didn't get into the finals recently)
(current C1 from current series)
Re: Ask Graeme?
Posted: Mon Jul 06, 2020 8:59 pm
by Gavin Chipper
Adam Dexter wrote: ↑Mon Jul 06, 2020 8:49 pm
Who would be the current UCWC? Is anyone keeping track?
A bit like
https://www.ufwc.co.uk/ where the winner of the first match is champion until defeated (a la boxing)
Issue I can see is people who retire unbeaten - not entirely sure how this is dealt with in boxing?
Well, I'll make a start. In series 1, Helen Grayson retired as the daily champion but then came back in series 6 where she lost to Olivia Lloyd who lost the series final to Darryl Francis. Francis lost to David Trace in the CoC, and Trace lost to Harvey Freeman in the final. Harvey Freeman lost (but in a Masters game if that counts) to Mark Nyman. That run would end with either Freeman or Nyman.
Re: Ask Graeme?
Posted: Mon Jul 06, 2020 9:40 pm
by Thomas Carey
this is yet another question that can be answered with 'a bunch of the best ever players and Frank Mulvey'
Re: Ask Graeme?
Posted: Mon Jul 06, 2020 9:53 pm
by Ben Wilson
Inspired by the fact that today's c4c birthdays include someone who's 28 and who's 56 (not to mention the fact that I'm on a lot of painkillers right now) I'm wondering how many successful players' (defined as octochamps or finalists) debuts are exactly twice/half as 'old' as that of another successful player, either defined as days from 2nd November 1982 or by episode numbers.
On the one hand, this is a question no one else is ever likely to ask, but on the other hand, there's probably a good reason for that.
Re: Ask Graeme?
Posted: Tue Jul 07, 2020 10:45 am
by Conor
Adam Dexter wrote: ↑Sun Jul 05, 2020 2:06 pm
Ahh very nice - I thought something like this would likely have already been covered.
Any idea if these are updated with dictionary updates?
I updated them around the end of last year, so should be correct as of then. Will check again shortly with the latest dictionary.
Re: Ask Graeme?
Posted: Tue Jul 07, 2020 4:39 pm
by James Laverty
Gavin Chipper wrote: ↑Mon Jul 06, 2020 8:59 pm
Adam Dexter wrote: ↑Mon Jul 06, 2020 8:49 pm
Who would be the current UCWC? Is anyone keeping track?
A bit like
https://www.ufwc.co.uk/ where the winner of the first match is champion until defeated (a la boxing)
Issue I can see is people who retire unbeaten - not entirely sure how this is dealt with in boxing?
Well, I'll make a start. In series 1, Helen Grayson retired as the daily champion but then came back in series 6 where she lost to Olivia Lloyd who lost the series final to Darryl Francis. Francis lost to David Trace in the CoC, and Trace lost to Harvey Freeman in the final. Harvey Freeman lost (but in a Masters game if that counts) to Mark Nyman. That run would end with either Freeman or Nyman.
I did another run at this, starting with Season 34 (assuming Freeman retired unbeaten after Supreme Championship).##
James Martin was the only octochamp that series, so starting with the finals
James Martin lost in the semis to Harshan Lamabadusuriya
Harshan Lamabadusuriya lost in the final to Huw Morgan
Huw Morgan lost in CoC IX prelims to Simon Pryde
Simon Pryde lost in CoC IX semis to Pete Cashmore
Pete Cashmore lost in the COC IX final to Natascha Kearsey
Natascha Kearsey later beat COC X winner Scott Mearns in a special episode, so I'd declare her retired at this point.
If doing again, I'd go so from Series 42, but I'll leave that for another time
Re: Ask Graeme?
Posted: Tue Jul 07, 2020 7:26 pm
by Adam Dexter
All very interesting thanks everyone.
Now a question related to personal stats.
We have the Superstats on Apterous.
I almost always pick CCCVVVCVC however my opponents obviously have a wide range of picks.
Is there any way it could be seen what letter selections have the best max % for a player?
I'm thinking C could be 1 and V could be 0, thus converting any string into binary. This could then be analysed so my pick for example would be 453, and might have a max % of 60.
I know some people at events use boxes and others write the letters in "their" selection technique rather than what is actually out. I just wondered if there had ever been a study into whether this actually makes a tangible difference.
Re: Ask Graeme?
Posted: Tue Jul 07, 2020 7:50 pm
by Thomas Carey
Adam Dexter wrote: ↑Tue Jul 07, 2020 7:26 pm
All very interesting thanks everyone.
Now a question related to personal stats.
We have the Superstats on Apterous.
I almost always pick CCCVVVCVC however my opponents obviously have a wide range of picks.
Is there any way it could be seen what letter selections have the best max % for a player?
I'm thinking C could be 1 and V could be 0, thus converting any string into binary. This could then be analysed so my pick for example would be 453, and might have a max % of 60.
I know some people at events use boxes and others write the letters in "their" selection technique rather than what is actually out. I just wondered if there had ever been a study into whether this actually makes a tangible difference.
I'm one of those people. I started playing this stupid game at 12 years old, and always picked CVCVCVC (VC at the start before I could fish) and then a few years later (starting with my TV games, always a good time to try something new) I started writing all selections like that. I think it helps a little, partially by making me look at two different selections (should I write mine differently? I never did the 3x3 box thing) but also I like that pick because it spells out syllables that you can read in your head and do something with (and also you're more likely to get a funny thing spelt out). The clearest example of where this pick helped me was in my coc QF against Zarte - many people asked me how I saw TIRAMISUS and it was because I wrote the selection IATSUMIRS as TISAMURIS which basically spelled it out. But I'd be interested to know whether or not it actually helps me...
Re: Ask Graeme?
Posted: Wed Jul 08, 2020 2:38 pm
by Ian Volante
I like the binary idea. My most common picks are 181 or 182, and less likely 180 and 183. I only do the box thing when writing on paper, so never on Apterous. Not sure which works better at all, not convinced there's a difference.
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 8:05 am
by Adam Dexter
Bit of an aside, but did some Excellery last night and came up with this horrendous formula:
=IF(OR(LEFT(A1,1)="A",LEFT(A1,1)="E",LEFT(A1,1)="I",LEFT(A1,1)="O",LEFT(A1,1)="U"),"V","C")&IF(OR(MID(A1, 2, 1)="A", MID(A1, 2, 1)="E", MID(A1, 2, 1)="I", MID(A1, 2, 1)="O", MID(A1, 2, 1)="U"), "V", "C")&IF(OR(MID(A1, 3, 1)="A", MID(A1, 3, 1)="E", MID(A1, 3, 1)="I", MID(A1, 3, 1)="O", MID(A1, 3, 1)="U"), "V", "C")&IF(OR(MID(A1, 4, 1)="A", MID(A1, 4, 1)="E", MID(A1, 4, 1)="I", MID(A1, 4, 1)="O", MID(A1, 4, 1)="U"), "V", "C")&IF(OR(MID(A1, 5, 1)="A", MID(A1, 5, 1)="E", MID(A1, 5, 1)="I", MID(A1, 5, 1)="O", MID(A1, 5, 1)="U"), "V", "C")&IF(OR(MID(A1, 6, 1)="A", MID(A1, 6, 1)="E", MID(A1, 6, 1)="I", MID(A1, 6, 1)="O", MID(A1, 6, 1)="U"), "V", "C")&IF(OR(MID(A1, 7, 1)="A", MID(A1, 7, 1)="E", MID(A1, 7, 1)="I", MID(A1, 7, 1)="O", MID(A1, 7, 1)="U"), "V", "C")&IF(OR(MID(A1, 8, 1)="A", MID(A1, 8, 1)="E", MID(A1, 8, 1)="I", MID(A1, 8, 1)="O", MID(A1, 8, 1)="U"), "V", "C")&IF(OR(MID(A1, 9, 1)="A", MID(A1, 9, 1)="E", MID(A1, 9, 1)="I", MID(A1, 9, 1)="O", MID(A1, 9, 1)="U"), "V", "C")
Basically this will convert any 9 letter string into CV format.
Major issue is that Stephen R has worked out there are 336 combinations of CV ensuring the vowel selection is legal - might be a bit time heavy doing it manually!
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 6:25 pm
by Adam Dexter
Adam Dexter wrote: ↑Thu Jul 09, 2020 8:05 am
Major issue is that Stephen R has worked out there are 336 combinations of CV ensuring the vowel selection is legal - might be a bit time heavy doing it manually!
Just an update to say I have worked out how to get all the combinations - I have everything ready, just need a shitload of data.
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 7:53 pm
by Matt Morrison
Adam Dexter wrote: ↑Thu Jul 09, 2020 8:05 am
Bit of an aside, but did some Excellery last night and came up with this horrendous formula:
=IF(OR(LEFT(A1,1)="A",LEFT(A1,1)="E",LEFT(A1,1)="I",LEFT(A1,1)="O",LEFT(A1,1)="U"),"V","C")&IF(OR(MID(A1, 2, 1)="A", MID(A1, 2, 1)="E", MID(A1, 2, 1)="I", MID(A1, 2, 1)="O", MID(A1, 2, 1)="U"), "V", "C")&IF(OR(MID(A1, 3, 1)="A", MID(A1, 3, 1)="E", MID(A1, 3, 1)="I", MID(A1, 3, 1)="O", MID(A1, 3, 1)="U"), "V", "C")&IF(OR(MID(A1, 4, 1)="A", MID(A1, 4, 1)="E", MID(A1, 4, 1)="I", MID(A1, 4, 1)="O", MID(A1, 4, 1)="U"), "V", "C")&IF(OR(MID(A1, 5, 1)="A", MID(A1, 5, 1)="E", MID(A1, 5, 1)="I", MID(A1, 5, 1)="O", MID(A1, 5, 1)="U"), "V", "C")&IF(OR(MID(A1, 6, 1)="A", MID(A1, 6, 1)="E", MID(A1, 6, 1)="I", MID(A1, 6, 1)="O", MID(A1, 6, 1)="U"), "V", "C")&IF(OR(MID(A1, 7, 1)="A", MID(A1, 7, 1)="E", MID(A1, 7, 1)="I", MID(A1, 7, 1)="O", MID(A1, 7, 1)="U"), "V", "C")&IF(OR(MID(A1, 8, 1)="A", MID(A1, 8, 1)="E", MID(A1, 8, 1)="I", MID(A1, 8, 1)="O", MID(A1, 8, 1)="U"), "V", "C")&IF(OR(MID(A1, 9, 1)="A", MID(A1, 9, 1)="E", MID(A1, 9, 1)="I", MID(A1, 9, 1)="O", MID(A1, 9, 1)="U"), "V", "C")
Basically this will convert any 9 letter string into CV format.
Major issue is that Stephen R has worked out there are 336 combinations of CV ensuring the vowel selection is legal - might be a bit time heavy doing it manually!
I'm not much of an Excel master but it did occur to me that you could just use COUNTIF to work with an external range instead - in this instance D3 is my conundrum and B3:B7 is AEIOU.
Code: Select all
=IF(COUNTIF(B3:B7,LEFT(D3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,2,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,4,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,5,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,6,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,7,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,8,1)),"V","C")&IF(COUNTIF(B3:B7,RIGHT(D3,1)),"V","C")
I guess you could count occurrences of V in the produced string with something like
Code: Select all
=LEN(E3) - LEN(SUBSTITUTE(E3,"V",""))
where E3 is the CCVCVCVV string - this will give you the vowel count. Also need to check it's a 9 letter word though.
And I can't think how to put all of this in one formula without accepting it's time to learn Visual Basic.
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 8:15 pm
by Matt Morrison
Think I got it son.
=IF(LEN(D3)=9,IF(AND(LEN(D3) - LEN(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(D3,"A",""),"E",""),"I",""),"O",""),"U",""))>1,LEN(D3) - LEN(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(D3,"A",""),"E",""),"I",""),"O",""),"U",""))<6),IF(COUNTIF(B3:B7,LEFT(D3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,2,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,4,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,5,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,6,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,7,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,8,1)),"V","C")&IF(COUNTIF(B3:B7,RIGHT(D3,1)),"V","C"),"check your fucking vowels"),"not a fucking 9")
It's absolutely disgusting how much duplication you have to put in to Excel, most upsetting.
So will let you know if it's not a 9, if there are the wrong amount of vowels (didn't want to make it longer in providing "too few" and "too many" responses!), and then if all those conditions are met convert to CVC format.
It's a huge shame that SUBSTITUTE only works with a range when you supply a range. I thought you might be able to do something like
Code: Select all
SUBSTITUTE(D3,{"A","I","E","O","U"},"")
rather than need 5 SUBSTITUTEs but didn't work.
Oh, and if you need to sanitise the input then every instance of D3 needs to be UPPER(D3) too of course.
Will throw it open to any Excel gurus now!
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 8:23 pm
by Matt Morrison
Ah, you can do that with SUBSTITUTE. I needed SUMPRODUCT around the maths though.
=IF(LEN(D3)=9,IF(AND(SUMPRODUCT(LEN(D3) - LEN(SUBSTITUTE(D3,{"A","E","I","O","U"},"")))>1,SUMPRODUCT(LEN(D3) - LEN(SUBSTITUTE(D3,{"A","E","I","O","U"},"")))<6),IF(COUNTIF(B3:B7,LEFT(D3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,2,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,3,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,4,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,5,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,6,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,7,1)),"V","C")&IF(COUNTIF(B3:B7,MID(D3,8,1)),"V","C")&IF(COUNTIF(B3:B7,RIGHT(D3,1)),"V","C"),"check your fucking vowels"),"not a fucking 9")
done now.
Re: Ask Graeme?
Posted: Thu Jul 09, 2020 8:45 pm
by Adam Dexter
Tis a thing of beauty so it is. Still - mine works so will leave as is. Anyone wanna shove me a load of data?
Re: Ask Graeme?
Posted: Wed Jul 15, 2020 8:13 pm
by Noel Mc
Any idea of the highest intermediate number of a number round solution for which you must go via that number?
That's worded horribly.
Re: Ask Graeme?
Posted: Wed Jul 15, 2020 8:50 pm
by Gavin Chipper
Noel Mc wrote: ↑Wed Jul 15, 2020 8:13 pm
Any idea of the highest intermediate number of a number round solution for which you must go via that number?
That's worded horribly.
There is
this thread but there's nothing conclusive in it by the looks of things.
Edit -
some further discussion.
Re: Ask Graeme?
Posted: Wed Jul 15, 2020 9:07 pm
by Sam Cappleman-Lynes
Enjoyed reliving that thread. Sadly I think I've lost the raw data I produced while masquerading as Dave.
Re: Ask Graeme?
Posted: Wed Jul 15, 2020 10:21 pm
by Gavin Chipper
Sam Cappleman-Lynes wrote: ↑Wed Jul 15, 2020 9:07 pm
Enjoyed reliving that thread. Sadly I think I've lost the raw data I produced while masquerading as Dave.
You came and went pretty quickly.
Re: Ask Graeme?
Posted: Thu Jul 16, 2020 1:15 am
by Graeme Cole
Noel Mc wrote: ↑Wed Jul 15, 2020 8:13 pm
Any idea of the highest intermediate number of a number round solution for which you must go via that number?
That's worded horribly.
I hacked about with my solver for most of the evening and then found an efficient-ish way to look for this. So far I've only looked at 4 large, as that's where I think we'll find the answer, but of all possible 4 large numbers rounds, the selection and target with the highest unavoidable intermediate result is 100 75 50 25 3 3 -> 996.
The only solution is (((50 + 3) * 25 + 3) * 75) / 100, which goes via 99,600.
Re: Ask Graeme?
Posted: Thu Jul 16, 2020 2:15 pm
by Charlie Reams
This is super cool, I have wondered about this question before so very interesting to get an answer. Would love to see the answers for 6 small etc if you have the compute time to grind those out!
Re: Ask Graeme?
Posted: Thu Jul 16, 2020 9:02 pm
by Sam Cappleman-Lynes
I can corroborate Graeme's 4-large result. His hunch is correct, there aren't any games outside of 4-large for which such a large intermediate result is necessary. Here are the largest unavoidable intermediate results for each of the 5 selections:
4 Large - 99600: 996 = ((50 + 3) * 25 + 3) * 75 / 100
3 Large - 94200: 942 = (50 * 25 + 6) * 75 / (10 * 10)
2 Large - 60900: 812 = (100 * 6 + 9) * 10 * 10 / 75
1 Large - 10020: 835 = (100 * 10 + 2) * 10 / (6 + 6)
6 Small - 7400: 925 = (9 * 8 + 2) * 10 * 10 / 8
Re: Ask Graeme?
Posted: Thu Jul 16, 2020 9:37 pm
by Sam Cappleman-Lynes
I also did the same thing for the Nasty large numbers:
4 Large - 27492: 316 = ((62 * 12 - 1) * 37 + 1) / 87
3 Large - 32451: 877 = ((62 * 6 + 1) * 87 - 2) / 37
2 Large - 22950: 918 = (62 * 37 * 10 + 10) / (5 * 5)
1 Large - 10440: 580 = 87 * 8 * (8 + 7) / (9 + 9)
I briefly wondered what the results would be if you were allowed *any* large numbers, Aegilops-style. But a quick estimate tells me it would take a few months to exhaustively check every case. The main problem is that I've had to turn off the optimisation where I throw away partial calculations if there are multiple ways of reaching the same value with the same numbers, because here we genuinely care about the difference between (100 + 75) - 50 and 100 + (75 - 50).
Re: Ask Graeme?
Posted: Thu Jul 16, 2020 9:42 pm
by Gavin Chipper
Sam Cappleman-Lynes wrote: ↑Wed Jul 15, 2020 9:07 pm
Enjoyed reliving that thread. Sadly I think I've lost the raw data I produced while masquerading as Dave.
Well, this has led me to some shocking news. I remembered there was another poster on here who posted some numbers game stats but didn't post that often and I wondered if that might be you as well. Turns out not. He was
Ray Folwell and he posted quite a bit in the
Analysis of 4 large thread. Anyway, looking at his profile picture it seems he is the same person that
died after being hit by a train in 2011.
Here is a
funeral notice as well.
Edit - Ray Folwell (doesn't come up in a forum search presumably because of the link?)
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 1:20 pm
by Sam Cappleman-Lynes
Which octochamp has the smallest range of scores? Which has the smallest
coefficient of variation?
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 2:38 pm
by Innis Carson
Gavin Chipper wrote: ↑Thu Jul 16, 2020 9:42 pm
Sam Cappleman-Lynes wrote: ↑Wed Jul 15, 2020 9:07 pm
Enjoyed reliving that thread. Sadly I think I've lost the raw data I produced while masquerading as Dave.
Well, this has led me to some shocking news. I remembered there was another poster on here who posted some numbers game stats but didn't post that often and I wondered if that might be you as well. Turns out not. He was
Ray Folwell and he posted quite a bit in the
Analysis of 4 large thread. Anyway, looking at his profile picture it seems he is the same person that
died after being hit by a train in 2011.
Here is a
funeral notice as well.
That's very sad indeed, and even sadder that it was so long before we found out. RIP.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 2:49 pm
by Innis Carson
Once upon a time I tried to calculate this in the hope that it might be
me, but found that I was pipped by
Tom Hargreaves. Very possible that I missed someone, though, or that a new winner has emerged since.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 3:24 pm
by Gavin Chipper
Innis Carson wrote: ↑Tue Jul 21, 2020 2:49 pm
Once upon a time I tried to calculate this in the hope that it might be
me, but found that I was pipped by
Tom Hargreaves. Very possible that I missed someone, though, or that a new winner has emerged since.
I seem to remember Kai Laddiman from somewhere.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 5:38 pm
by Thomas Carey
Gavin Chipper wrote: ↑Tue Jul 21, 2020 3:24 pm
Innis Carson wrote: ↑Tue Jul 21, 2020 2:49 pm
Once upon a time I tried to calculate this in the hope that it might be
me, but found that I was pipped by
Tom Hargreaves. Very possible that I missed someone, though, or that a new winner has emerged since.
I seem to remember Kai Laddiman from somewhere.
this is kinda old and also not quite the same thing but basically is
reckon i'd be quite high up on your list sam, got a standard deviation of 4.95 and a higher mean than most of the people on that list. wiki's down atm so i can't check kai's or tom's or whatever but sometimes that's life
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 6:02 pm
by Innis Carson
Wow, I am amazed at how comprehensively I managed to misremember that whole exchange. Remarkable.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 6:19 pm
by Sam Cappleman-Lynes
Thomas Carey wrote: ↑Tue Jul 21, 2020 5:38 pmreckon i'd be quite high up on your list sam, got a standard deviation of 4.95 and a higher mean than most of the people on that list
Nice. Your higher mean brings the coefficient of variance down below both Tom's and Kai's. Looks like you and Hargreaves both have a range of 13 too, which is a good shout for being the smallest range and therefore an answer to my first question.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 6:33 pm
by Gavin Chipper
Sam Cappleman-Lynes wrote: ↑Tue Jul 21, 2020 6:19 pm
Thomas Carey wrote: ↑Tue Jul 21, 2020 5:38 pmreckon i'd be quite high up on your list sam, got a standard deviation of 4.95 and a higher mean than most of the people on that list
Nice. Your higher mean brings the coefficient of variance down below both Tom's and Kai's. Looks like you and Hargreaves both have a range of 13 too, which is a good shout for being the smallest range and therefore an answer to my first question.
I've heard rumours that you've been recording. It will be interesting to see where you fit in.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 9:55 pm
by Graeme Cole
This counts all octochamps in the 15-round era, lumping in old 15 rounders with new 15 rounders to annoy pedants. Only the heat games in the octorun are considered.
Smallest difference between lowest and highest score in an octorun (top 10 plus ties):
Code: Select all
name lowest highest range
-------------------- ------- ------- -------
John Hardie 99 110 11
Tom Hargreaves 98 111 13
Thomas Carey 109 122 13
Jack Welsby 96 112 16
Kai Laddiman 89 105 16
Dylan Taylor 113 129 16
Joe McGonigle 78 95 17
Toby McDonald 104 121 17
Tracey Mills 97 116 19
John Davies 87 106 19
Alex Fish 93 112 19
Shane Roberts 86 105 19
Giles Hutchings 111 130 19
Zarte Siempre 111 130 19
George Ford 95 114 19
Largest difference between lowest and highest score in an octorun (top 10 plus ties):
Code: Select all
name lowest highest range
-------------------- ------- ------- -------
Steven Briers 76 136 60
Stewart Holden 79 137 58
Robin Johnson 57 112 55
Carl Williams 62 116 54
Michael Pfeffer 65 119 54
Charlie Reams 73 126 53
Jonathan Liew 68 120 52
Mark Davies 85 137 52
Kevin Thurlow 72 123 51
Dave Taylor 61 112 51
John Osmond 68 119 51
Octorun scores: lowest coefficient of variation (top 10):
Code: Select all
name average stddev coeffvar
-------------------- ---------- ---------- ----------
Dylan Taylor 121.75 4.24 3.48%
John Hardie 105.62 3.71 3.51%
Thomas Carey 115.38 4.95 4.29%
Tom Hargreaves 106.25 4.74 4.46%
Toby McDonald 110.25 5.29 4.79%
Kai Laddiman 94.50 4.66 4.94%
Giles Hutchings 120.62 5.98 4.96%
Andy Platt 111.12 5.60 5.04%
Jack Welsby 103.88 5.51 5.30%
Tracey Mills 104.50 5.81 5.56%
Octorun scores: highest coefficient of variation (top 10):
Code: Select all
name average stddev coeffvar
-------------------- ---------- ---------- ----------
Kevin Thurlow 96.12 17.80 18.52%
Dave Taylor 86.38 15.92 18.43%
John Osmond 88.88 16.31 18.35%
Corrina Attwood 85.12 15.02 17.64%
Eoin Jackson 97.25 17.14 17.62%
Carl Williams 88.50 15.39 17.39%
Robin Johnson 93.12 15.99 17.17%
Tia Corkish 87.12 14.66 16.82%
Jimmy Gough 97.75 16.30 16.68%
Andrew Macleod 101.12 16.13 15.95%
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 10:00 pm
by Graeme Cole
To head off the inevitable questions of "where am I in the list?" I might as well post both lists in full so you can find your names.
Sorted by range:
Code: Select all
name lowest highest range
-------------------- ------- ------- -------
John Hardie 99 110 11
Tom Hargreaves 98 111 13
Thomas Carey 109 122 13
Jack Welsby 96 112 16
Kai Laddiman 89 105 16
Dylan Taylor 113 129 16
Joe McGonigle 78 95 17
Toby McDonald 104 121 17
Tracey Mills 97 116 19
John Davies 87 106 19
Alex Fish 93 112 19
Shane Roberts 86 105 19
Giles Hutchings 111 130 19
Zarte Siempre 111 130 19
George Ford 95 114 19
Joe Zubaidi 73 93 20
Andy Platt 100 120 20
Brendan Whitehurst 92 112 20
Innis Carson 94 115 21
Elliott Mack 92 113 21
Eoin Monaghan 100 121 21
Annie Humphries 87 108 21
Mark Murray 100 122 22
Paul Howe 91 113 22
Richard Brittain 94 116 22
Jeffrey Burgin 73 95 22
Tom Rowell 87 109 22
Jamie Washington 91 113 22
Tricia Pay 96 119 23
Amey Deshpande 79 102 23
Nick Wainwright 76 99 23
Jack Hurst 110 133 23
Ned Pendleton 69 92 23
Kevin Steede 85 108 23
James Slater 72 95 23
Dave Ashton 82 105 23
Brenda Jolley 77 101 24
Keith Maynard 86 110 24
Adam Gillard 102 126 24
Edward McCullagh 97 121 24
David Barnard 87 111 24
James Kennedy 80 105 25
David Thirlwall 77 102 25
Conor Travers 99 124 25
Craig Beevers 103 128 25
Tim Reypert 88 113 25
Cate Henderson 88 113 25
Jack Worsley 85 110 25
Chris Marshall 73 98 25
Alan Duval 68 93 25
Richard Heald 89 115 26
David Von Geyer 72 98 26
Jen Steadman 104 130 26
Wesley Jardine 78 104 26
Jamie Wing 96 122 26
Paul James 83 110 27
John Mayhew 91 118 27
Jon O'Neill 87 114 27
Jonathan Rawlinson 93 120 27
Rose Boyle 73 100 27
Tim Down 78 105 27
Robin McKay 102 129 27
Stephen Read 104 131 27
David Law 70 97 27
George Greenhough 86 114 28
Gary Male 84 112 28
Bradley Cates 95 123 28
Martin May 89 117 28
Steve Wood 64 92 28
Ann Dibben 99 128 29
Marcus Hares 87 116 29
Stephen Briggs 98 127 29
Ross Allatt 81 110 29
Stu Horsey 76 105 29
Antoinette Ryan 91 120 29
Dave Costello 74 103 29
Paul Harper 77 107 30
Grace Page 90 120 30
Stuart Earl 83 113 30
Judith Young 71 101 30
David O'Donnell 92 122 30
Neil Zussman 81 111 30
Jayne Wisniewski 72 102 30
Tom Cappleman 105 135 30
Stuart Solomons 86 117 31
Jeffrey Hansford 96 127 31
James Doohan 70 101 31
Danny Pledger 65 96 31
Jonathan Wynn 100 131 31
Paul Keane 73 104 31
Adam Rolston 68 99 31
Mike Daysley 79 110 31
James Haughton 103 134 31
Andrew Hulme 101 133 32
Chris Cummins 91 123 32
John Gray 78 110 32
Steven Moir 79 111 32
James Roberts 76 108 32
Martin Bishop 83 115 32
Heather Styles 82 114 32
Jordan Barker 72 104 32
Paul Erdunast 98 130 32
Steven Turnbull 64 96 32
Kirk Bevins 97 130 33
Michael Bowden 73 106 33
Jon Corby 95 128 33
James Hurrell 88 121 33
Steve Redfern 72 105 33
Julian Fell 104 138 34
Andrew Naylor 76 110 34
Jim Bentley 79 113 34
Oliver Garner 81 115 34
Daniel Pati 93 127 34
Graeme Cole 91 125 34
Elliott Mellor 118 152 34
Andy Evans 80 114 34
Morgan Sauer 81 115 34
Rupert Stokoe 82 117 35
Danny Hamilton 85 120 35
Martin Gardner 72 107 35
Richard Pay 70 105 35
Jeff Clayton 81 116 35
Scott Gillies 87 122 35
Liam Shaw 69 104 35
Jean Webby 76 112 36
Aaron Webber 75 111 36
Andy Noden 82 118 36
Gerry Tynan 91 127 36
Lawrie Collingwood 70 106 36
Chris Thorn 83 119 36
Maggie Barlow 73 109 36
Tom Silverlock 75 111 36
Chris Wills 92 129 37
Tony Warren 71 108 37
Chris McHenry 62 99 37
Tom Barnes 86 123 37
Noel McIlvenny 83 120 37
Tony Manwani 67 105 38
Mike Pullin 76 114 38
Paul Gallen 83 121 38
John Hunt 79 117 38
Mark Tournoff 79 117 38
John Brackstone 81 119 38
Matthew Shore 88 126 38
Tom Chafer-Cook 91 129 38
Graeme Staples 64 102 38
Andrew Jackson 78 116 38
Suzi Purcell 67 106 39
Laurence Killen 79 118 39
Phil Davies 75 114 39
Mark Deeks 86 126 40
Bradley Horrocks 82 122 40
Fiona Titcombe 69 109 40
Andrew Fenton 71 112 41
David Edwards 72 113 41
Peter Lee 78 120 42
Julia Wilkinson 80 122 42
Michael Macdonald-Cooper 84 126 42
Samir Pilica 78 120 42
Conrad Teixeira 82 124 42
Paul Nixon 93 136 43
Nik Von Uexkull 71 114 43
Eileen Taylor 67 111 44
Wendy Roe 79 123 44
Jimmy Gough 72 116 44
Chris Davies 95 139 44
Lee Hartley 72 117 45
Dan McColm 92 137 45
Tia Corkish 63 108 45
Matthew Tassier 75 120 45
Sweyn Kirkness 76 122 46
Brian Selway 63 109 46
Andy McGurn 64 110 46
Martin Hurst 83 130 47
Junaid Mubeen 74 121 47
Ryan Taylor 76 123 47
Andrew Macleod 77 124 47
John Shaw 73 121 48
Jonathan Coles 75 124 49
Moose Rosser 83 132 49
John Cowen 79 128 49
Eoin Jackson 74 123 49
Glen Webb 91 141 50
Corrina Attwood 68 118 50
Kevin Thurlow 72 123 51
Dave Taylor 61 112 51
John Osmond 68 119 51
Jonathan Liew 68 120 52
Mark Davies 85 137 52
Charlie Reams 73 126 53
Carl Williams 62 116 54
Michael Pfeffer 65 119 54
Robin Johnson 57 112 55
Stewart Holden 79 137 58
Steven Briers 76 136 60
Sorted by coefficient of variation:
Code: Select all
name average stddev coeffvar
-------------------- ---------- ---------- ----------
Dylan Taylor 121.75 4.24 3.48%
John Hardie 105.62 3.71 3.51%
Thomas Carey 115.38 4.95 4.29%
Tom Hargreaves 106.25 4.74 4.46%
Toby McDonald 110.25 5.29 4.79%
Kai Laddiman 94.50 4.66 4.94%
Giles Hutchings 120.62 5.98 4.96%
Andy Platt 111.12 5.60 5.04%
Jack Welsby 103.88 5.51 5.30%
Tracey Mills 104.50 5.81 5.56%
Jack Hurst 118.25 6.83 5.78%
Mark Murray 112.75 6.53 5.79%
George Ford 104.00 6.04 5.81%
Zarte Siempre 120.88 7.32 6.06%
Innis Carson 107.62 6.58 6.11%
Eoin Monaghan 112.25 6.89 6.14%
Alex Fish 99.75 6.22 6.24%
Elliott Mack 100.50 6.34 6.31%
Annie Humphries 96.88 6.19 6.39%
Joe McGonigle 89.50 5.72 6.39%
Stephen Read 114.38 7.48 6.54%
Brendan Whitehurst 102.62 6.76 6.59%
Jen Steadman 119.00 7.89 6.63%
Robin McKay 111.00 7.43 6.70%
Edward McCullagh 112.00 7.65 6.83%
Jack Worsley 102.25 7.08 6.93%
Tricia Pay 107.12 7.46 6.96%
John Davies 95.75 6.81 7.12%
Jamie Washington 103.12 7.37 7.15%
Craig Beevers 113.38 8.18 7.22%
Conor Travers 111.25 8.04 7.23%
Paul Howe 101.88 7.54 7.40%
Adam Gillard 112.88 8.40 7.44%
Tom Cappleman 115.88 8.65 7.47%
Elliott Mellor 132.62 10.01 7.55%
Kevin Steede 99.00 7.48 7.56%
Shane Roberts 95.75 7.28 7.60%
Jon O'Neill 100.50 7.68 7.64%
Cate Henderson 97.75 7.56 7.74%
Tom Rowell 96.75 7.51 7.76%
Tim Reypert 96.62 7.58 7.85%
Rose Boyle 87.62 6.91 7.88%
Paul James 99.25 7.90 7.96%
Richard Heald 99.38 7.97 8.02%
Antoinette Ryan 107.38 8.69 8.09%
Joe Zubaidi 83.12 6.81 8.19%
James Haughton 119.75 9.82 8.20%
Richard Brittain 102.50 8.43 8.22%
George Greenhough 102.12 8.45 8.27%
Jamie Wing 113.62 9.41 8.28%
James Kennedy 92.00 7.68 8.35%
Bradley Cates 107.75 9.01 8.36%
James Slater 84.00 7.05 8.40%
Marcus Hares 104.25 8.76 8.40%
Brenda Jolley 89.75 7.55 8.41%
Alan Duval 84.25 7.10 8.43%
Keith Maynard 98.12 8.28 8.44%
David Barnard 96.38 8.18 8.49%
Ann Dibben 110.50 9.45 8.55%
Julian Fell 115.50 9.89 8.56%
Dave Ashton 92.38 7.94 8.59%
Jonathan Wynn 115.25 9.91 8.60%
Amey Deshpande 89.75 7.74 8.63%
David Von Geyer 90.50 7.84 8.67%
Jonathan Rawlinson 106.25 9.22 8.67%
Martin May 104.88 9.10 8.68%
Paul Erdunast 115.62 10.09 8.72%
Stuart Solomons 99.50 8.76 8.80%
Andrew Hulme 116.25 10.24 8.81%
David O'Donnell 110.00 9.70 8.81%
John Mayhew 101.38 8.97 8.85%
Nick Wainwright 90.75 8.04 8.86%
Grace Page 103.62 9.27 8.95%
Mike Daysley 98.25 8.86 9.01%
Kirk Bevins 115.62 10.66 9.22%
James Doohan 87.75 8.12 9.25%
Michael Bowden 92.38 8.57 9.28%
Jeffrey Burgin 85.62 7.95 9.29%
Jeff Clayton 105.75 9.97 9.43%
David Thirlwall 88.00 8.38 9.52%
David Law 87.50 8.35 9.54%
Dave Costello 90.38 8.64 9.57%
Stephen Briggs 110.38 10.62 9.62%
Gary Male 93.75 9.02 9.63%
Chris Cummins 107.25 10.35 9.65%
Jeffrey Hansford 102.25 9.97 9.75%
Steve Wood 84.38 8.35 9.90%
John Gray 94.62 9.38 9.91%
Heather Styles 92.12 9.24 10.03%
James Roberts 92.00 9.26 10.07%
Ross Allatt 92.62 9.35 10.10%
Neil Zussman 96.00 9.70 10.10%
Chris Marshall 85.25 8.61 10.10%
James Hurrell 104.75 10.59 10.11%
Jon Corby 107.00 10.86 10.15%
Scott Gillies 101.25 10.32 10.19%
Chris Wills 109.38 11.19 10.23%
Bradley Horrocks 108.00 11.22 10.39%
Daniel Pati 105.00 10.92 10.40%
Paul Harper 92.38 9.64 10.44%
Stu Horsey 91.50 9.58 10.47%
Judith Young 88.38 9.29 10.51%
Maggie Barlow 99.25 10.45 10.53%
Gerry Tynan 108.38 11.43 10.55%
Noel McIlvenny 99.12 10.56 10.66%
Martin Bishop 101.12 10.79 10.67%
Rupert Stokoe 97.00 10.37 10.69%
Dan McColm 117.75 12.69 10.77%
Steven Moir 95.38 10.31 10.81%
Graeme Cole 101.62 11.07 10.89%
Tim Down 89.62 9.84 10.97%
Tom Chafer-Cook 112.00 12.31 10.99%
Danny Hamilton 95.12 10.47 11.01%
John Brackstone 102.75 11.32 11.02%
Matthew Shore 106.25 11.77 11.07%
Stuart Earl 100.88 11.17 11.08%
Paul Gallen 105.75 11.80 11.16%
Mark Deeks 103.00 11.58 11.24%
Andy Noden 103.50 11.64 11.25%
Mike Pullin 94.50 10.63 11.25%
Mark Tournoff 101.12 11.38 11.26%
Chris Davies 111.50 12.59 11.29%
Paul Nixon 111.62 12.61 11.30%
Andrew Naylor 97.62 11.17 11.44%
Wesley Jardine 94.12 10.82 11.50%
Paul Keane 93.00 10.74 11.54%
Samir Pilica 100.12 11.56 11.54%
Jean Webby 92.25 10.73 11.63%
Glen Webb 118.12 13.82 11.70%
Aaron Webber 96.62 11.38 11.78%
Steven Turnbull 78.62 9.33 11.86%
Wendy Roe 97.62 11.60 11.88%
Jordan Barker 89.12 10.59 11.88%
Steve Redfern 86.12 10.25 11.90%
Chris Thorn 95.00 11.40 12.00%
Morgan Sauer 96.88 11.87 12.25%
Martin Hurst 103.75 12.73 12.27%
Chris McHenry 80.38 9.92 12.35%
Richard Pay 91.50 11.41 12.47%
Tony Manwani 89.38 11.16 12.48%
Ned Pendleton 83.00 10.37 12.49%
Andy Evans 95.00 11.88 12.51%
Tom Barnes 102.75 12.86 12.52%
Jayne Wisniewski 88.12 11.08 12.58%
Michael Macdonald-Cooper 97.50 12.27 12.58%
Laurence Killen 99.25 12.55 12.64%
Danny Pledger 79.38 10.05 12.66%
Graeme Staples 84.75 10.79 12.73%
Oliver Garner 100.25 12.81 12.78%
Phil Davies 94.62 12.11 12.80%
Junaid Mubeen 98.75 12.68 12.84%
Adam Rolston 81.25 10.46 12.88%
Fiona Titcombe 90.00 11.60 12.89%
Tom Silverlock 88.00 11.45 13.01%
Ryan Taylor 99.00 12.97 13.10%
Martin Gardner 93.25 12.22 13.11%
Lawrie Collingwood 91.75 12.05 13.13%
Mark Davies 105.50 13.87 13.15%
Jim Bentley 94.50 12.46 13.19%
Julia Wilkinson 93.00 12.30 13.22%
Andrew Fenton 91.00 12.26 13.47%
John Cowen 99.88 13.58 13.59%
Liam Shaw 88.50 12.05 13.62%
John Hunt 98.50 13.43 13.63%
Eileen Taylor 94.25 12.89 13.68%
Peter Lee 100.12 13.72 13.71%
Sweyn Kirkness 95.62 13.28 13.89%
Andrew Jackson 95.50 13.34 13.97%
Nik Von Uexkull 93.62 13.11 14.01%
Lee Hartley 101.38 14.43 14.23%
David Edwards 92.12 13.18 14.30%
Moose Rosser 104.75 15.06 14.38%
Tony Warren 89.00 12.87 14.47%
Suzi Purcell 85.75 12.49 14.56%
Stewart Holden 108.75 15.86 14.58%
Conrad Teixeira 100.75 14.79 14.68%
Michael Pfeffer 96.25 14.32 14.88%
Andy McGurn 87.75 13.13 14.96%
Brian Selway 93.25 13.98 14.99%
Jonathan Coles 93.25 14.11 15.13%
Jonathan Liew 94.25 14.31 15.18%
John Shaw 96.75 14.69 15.19%
Charlie Reams 102.50 15.59 15.21%
Matthew Tassier 99.88 15.28 15.30%
Steven Briers 105.38 16.44 15.60%
Andrew Macleod 101.12 16.13 15.95%
Jimmy Gough 97.75 16.30 16.68%
Tia Corkish 87.12 14.66 16.82%
Robin Johnson 93.12 15.99 17.17%
Carl Williams 88.50 15.39 17.39%
Eoin Jackson 97.25 17.14 17.62%
Corrina Attwood 85.12 15.02 17.64%
John Osmond 88.88 16.31 18.35%
Dave Taylor 86.38 15.92 18.43%
Kevin Thurlow 96.12 17.80 18.52%
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 10:14 pm
by Gavin Chipper
Innis, you are way down the list.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 10:30 pm
by Innis Carson
On the contrary - as I recall, I posted the complete list just a couple of hours ago and I was second behind Harvey Freeman.
Re: Ask Graeme?
Posted: Tue Jul 21, 2020 10:33 pm
by Thomas Carey
Graeme Cole wrote: ↑Tue Jul 21, 2020 10:00 pm
Sorted by coefficient of variation:
Code: Select all
name average stddev coeffvar
-------------------- ---------- ---------- ----------
Jean Webby 92.25 10.73 11.63%
Glen Webb 118.12 13.82 11.70%
Aaron Webber 96.62 11.38 11.78%
love it. cheers! can't believe i got 3rd again, especially to those two. crushing
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 7:29 am
by Gavin Chipper
I don't actually think coefficient of variation is the best measure. If anything there's a ceiling effect which means that you'd expect lower variance with higher average score.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 7:37 am
by Sam Cappleman-Lynes
Gavin Chipper wrote: ↑Wed Jul 22, 2020 7:29 am
I don't actually think coefficient of variation is the best measure. If anything there's a ceiling effect which means that you'd expect lower variance with higher average score.
This would probably be approximately true if 9 letter words didn't score double. There's also the differences between the solvability of the various numbers selections to contend with. But it's all just a bit of fun.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 8:03 am
by Conor
I've had a tacit assumption that over the years the standard of the top contestants has generally risen while the standard of 'most' contestants has slightly dipped. I don't think there's much point testing the first part, but how about the second?
I'd suggest something like comparing the average maxes (grouped by letters, numbers, conundrum) per game taken from any player who didn't reach the finals. And this can be shown by series (as a time series). I'd still expect a small drop with the dictionary change in 2015. And will be interesting to see if the Apterous effect has trickled down at all.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 8:44 am
by Tom S
Conor wrote: ↑Wed Jul 22, 2020 8:03 am
I've had a tacit assumption that over the years the standard of the top contestants has generally risen while the standard of 'most' contestants has slightly dipped. I don't think there's much point testing the first part, but how about the second?
I'd say that this is an accurate assumption, but would be intrigued to see what the results are.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 9:43 am
by Gavin Chipper
Sam Cappleman-Lynes wrote: ↑Wed Jul 22, 2020 7:37 am
Gavin Chipper wrote: ↑Wed Jul 22, 2020 7:29 am
I don't actually think coefficient of variation is the best measure. If anything there's a ceiling effect which means that you'd expect lower variance with higher average score.
This would probably be approximately true if 9 letter words didn't score double. There's also the differences between the solvability of the various numbers selections to contend with. But it's all just a bit of fun.
Next question - what is the correlation between octotal and standard deviation?
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 10:21 am
by Sam Cappleman-Lynes
Gavin Chipper wrote: ↑Wed Jul 22, 2020 9:43 am
Next question - what is the correlation between octotal and standard deviation?
Very little correlation. The Spearman's rank coefficient comes out as -0.15, based on the data Graeme has provided.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 11:23 am
by Gavin Chipper
Sam Cappleman-Lynes wrote: ↑Wed Jul 22, 2020 10:21 am
Gavin Chipper wrote: ↑Wed Jul 22, 2020 9:43 am
Next question - what is the correlation between octotal and standard deviation?
Very little correlation. The Spearman's rank coefficient comes out as -0.15, based on the data Graeme has provided.
I was half-expecting it to be negative based on what I posted above. Obviously against that was the thing about 9-letter words. Also, while players who average in the 80s might generally have a high standard deviation to their score, those who actually do deviate very much get selected out of the gene pool so to speak. So octotals might be a poor sample for looking at standard deviation for lower averages.
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 12:04 pm
by Ian Volante
Graeme Cole wrote: ↑Tue Jul 21, 2020 10:00 pm
To head off the inevitable questions of "where am I in the list?" I might as well post both lists in full so you can find your names.
Sorted by range:
Code: Select all
name lowest highest range
-------------------- ------- ------- -------
John Hardie 99 110 11
Tom Hargreaves 98 111 13
Thomas Carey 109 122 13
Jack Welsby 96 112 16
Kai Laddiman 89 105 16
Dylan Taylor 113 129 16
Joe McGonigle 78 95 17
Toby McDonald 104 121 17
Tracey Mills 97 116 19
John Davies 87 106 19
Alex Fish 93 112 19
Shane Roberts 86 105 19
Giles Hutchings 111 130 19
Zarte Siempre 111 130 19
George Ford 95 114 19
Joe Zubaidi 73 93 20
Andy Platt 100 120 20
Brendan Whitehurst 92 112 20
Innis Carson 94 115 21
Elliott Mack 92 113 21
Eoin Monaghan 100 121 21
Annie Humphries 87 108 21
Mark Murray 100 122 22
Paul Howe 91 113 22
Richard Brittain 94 116 22
Jeffrey Burgin 73 95 22
Tom Rowell 87 109 22
Jamie Washington 91 113 22
Tricia Pay 96 119 23
Amey Deshpande 79 102 23
Nick Wainwright 76 99 23
Jack Hurst 110 133 23
Ned Pendleton 69 92 23
Kevin Steede 85 108 23
James Slater 72 95 23
Dave Ashton 82 105 23
Brenda Jolley 77 101 24
Keith Maynard 86 110 24
Adam Gillard 102 126 24
Edward McCullagh 97 121 24
David Barnard 87 111 24
James Kennedy 80 105 25
David Thirlwall 77 102 25
Conor Travers 99 124 25
Craig Beevers 103 128 25
Tim Reypert 88 113 25
Cate Henderson 88 113 25
Jack Worsley 85 110 25
Chris Marshall 73 98 25
Alan Duval 68 93 25
Richard Heald 89 115 26
David Von Geyer 72 98 26
Jen Steadman 104 130 26
Wesley Jardine 78 104 26
Jamie Wing 96 122 26
Paul James 83 110 27
John Mayhew 91 118 27
Jon O'Neill 87 114 27
Jonathan Rawlinson 93 120 27
Rose Boyle 73 100 27
Tim Down 78 105 27
Robin McKay 102 129 27
Stephen Read 104 131 27
David Law 70 97 27
George Greenhough 86 114 28
Gary Male 84 112 28
Bradley Cates 95 123 28
Martin May 89 117 28
Steve Wood 64 92 28
Ann Dibben 99 128 29
Marcus Hares 87 116 29
Stephen Briggs 98 127 29
Ross Allatt 81 110 29
Stu Horsey 76 105 29
Antoinette Ryan 91 120 29
Dave Costello 74 103 29
Paul Harper 77 107 30
Grace Page 90 120 30
Stuart Earl 83 113 30
Judith Young 71 101 30
David O'Donnell 92 122 30
Neil Zussman 81 111 30
Jayne Wisniewski 72 102 30
Tom Cappleman 105 135 30
Stuart Solomons 86 117 31
Jeffrey Hansford 96 127 31
James Doohan 70 101 31
Danny Pledger 65 96 31
Jonathan Wynn 100 131 31
Paul Keane 73 104 31
Adam Rolston 68 99 31
Mike Daysley 79 110 31
James Haughton 103 134 31
Andrew Hulme 101 133 32
Chris Cummins 91 123 32
John Gray 78 110 32
Steven Moir 79 111 32
James Roberts 76 108 32
Martin Bishop 83 115 32
Heather Styles 82 114 32
Jordan Barker 72 104 32
Paul Erdunast 98 130 32
Steven Turnbull 64 96 32
Kirk Bevins 97 130 33
Michael Bowden 73 106 33
Jon Corby 95 128 33
James Hurrell 88 121 33
Steve Redfern 72 105 33
Julian Fell 104 138 34
Andrew Naylor 76 110 34
Jim Bentley 79 113 34
Oliver Garner 81 115 34
Daniel Pati 93 127 34
Graeme Cole 91 125 34
Elliott Mellor 118 152 34
Andy Evans 80 114 34
Morgan Sauer 81 115 34
Rupert Stokoe 82 117 35
Danny Hamilton 85 120 35
Martin Gardner 72 107 35
Richard Pay 70 105 35
Jeff Clayton 81 116 35
Scott Gillies 87 122 35
Liam Shaw 69 104 35
Jean Webby 76 112 36
Aaron Webber 75 111 36
Andy Noden 82 118 36
Gerry Tynan 91 127 36
Lawrie Collingwood 70 106 36
Chris Thorn 83 119 36
Maggie Barlow 73 109 36
Tom Silverlock 75 111 36
Chris Wills 92 129 37
Tony Warren 71 108 37
Chris McHenry 62 99 37
Tom Barnes 86 123 37
Noel McIlvenny 83 120 37
Tony Manwani 67 105 38
Mike Pullin 76 114 38
Paul Gallen 83 121 38
John Hunt 79 117 38
Mark Tournoff 79 117 38
John Brackstone 81 119 38
Matthew Shore 88 126 38
Tom Chafer-Cook 91 129 38
Graeme Staples 64 102 38
Andrew Jackson 78 116 38
Suzi Purcell 67 106 39
Laurence Killen 79 118 39
Phil Davies 75 114 39
Mark Deeks 86 126 40
Bradley Horrocks 82 122 40
Fiona Titcombe 69 109 40
Andrew Fenton 71 112 41
David Edwards 72 113 41
Peter Lee 78 120 42
Julia Wilkinson 80 122 42
Michael Macdonald-Cooper 84 126 42
Samir Pilica 78 120 42
Conrad Teixeira 82 124 42
Paul Nixon 93 136 43
Nik Von Uexkull 71 114 43
Eileen Taylor 67 111 44
Wendy Roe 79 123 44
Jimmy Gough 72 116 44
Chris Davies 95 139 44
Lee Hartley 72 117 45
Dan McColm 92 137 45
Tia Corkish 63 108 45
Matthew Tassier 75 120 45
Sweyn Kirkness 76 122 46
Brian Selway 63 109 46
Andy McGurn 64 110 46
Martin Hurst 83 130 47
Junaid Mubeen 74 121 47
Ryan Taylor 76 123 47
Andrew Macleod 77 124 47
John Shaw 73 121 48
Jonathan Coles 75 124 49
Moose Rosser 83 132 49
John Cowen 79 128 49
Eoin Jackson 74 123 49
Glen Webb 91 141 50
Corrina Attwood 68 118 50
Kevin Thurlow 72 123 51
Dave Taylor 61 112 51
John Osmond 68 119 51
Jonathan Liew 68 120 52
Mark Davies 85 137 52
Charlie Reams 73 126 53
Carl Williams 62 116 54
Michael Pfeffer 65 119 54
Robin Johnson 57 112 55
Stewart Holden 79 137 58
Steven Briers 76 136 60
Sorted by coefficient of variation:
Code: Select all
name average stddev coeffvar
-------------------- ---------- ---------- ----------
Dylan Taylor 121.75 4.24 3.48%
John Hardie 105.62 3.71 3.51%
Thomas Carey 115.38 4.95 4.29%
Tom Hargreaves 106.25 4.74 4.46%
Toby McDonald 110.25 5.29 4.79%
Kai Laddiman 94.50 4.66 4.94%
Giles Hutchings 120.62 5.98 4.96%
Andy Platt 111.12 5.60 5.04%
Jack Welsby 103.88 5.51 5.30%
Tracey Mills 104.50 5.81 5.56%
Jack Hurst 118.25 6.83 5.78%
Mark Murray 112.75 6.53 5.79%
George Ford 104.00 6.04 5.81%
Zarte Siempre 120.88 7.32 6.06%
Innis Carson 107.62 6.58 6.11%
Eoin Monaghan 112.25 6.89 6.14%
Alex Fish 99.75 6.22 6.24%
Elliott Mack 100.50 6.34 6.31%
Annie Humphries 96.88 6.19 6.39%
Joe McGonigle 89.50 5.72 6.39%
Stephen Read 114.38 7.48 6.54%
Brendan Whitehurst 102.62 6.76 6.59%
Jen Steadman 119.00 7.89 6.63%
Robin McKay 111.00 7.43 6.70%
Edward McCullagh 112.00 7.65 6.83%
Jack Worsley 102.25 7.08 6.93%
Tricia Pay 107.12 7.46 6.96%
John Davies 95.75 6.81 7.12%
Jamie Washington 103.12 7.37 7.15%
Craig Beevers 113.38 8.18 7.22%
Conor Travers 111.25 8.04 7.23%
Paul Howe 101.88 7.54 7.40%
Adam Gillard 112.88 8.40 7.44%
Tom Cappleman 115.88 8.65 7.47%
Elliott Mellor 132.62 10.01 7.55%
Kevin Steede 99.00 7.48 7.56%
Shane Roberts 95.75 7.28 7.60%
Jon O'Neill 100.50 7.68 7.64%
Cate Henderson 97.75 7.56 7.74%
Tom Rowell 96.75 7.51 7.76%
Tim Reypert 96.62 7.58 7.85%
Rose Boyle 87.62 6.91 7.88%
Paul James 99.25 7.90 7.96%
Richard Heald 99.38 7.97 8.02%
Antoinette Ryan 107.38 8.69 8.09%
Joe Zubaidi 83.12 6.81 8.19%
James Haughton 119.75 9.82 8.20%
Richard Brittain 102.50 8.43 8.22%
George Greenhough 102.12 8.45 8.27%
Jamie Wing 113.62 9.41 8.28%
James Kennedy 92.00 7.68 8.35%
Bradley Cates 107.75 9.01 8.36%
James Slater 84.00 7.05 8.40%
Marcus Hares 104.25 8.76 8.40%
Brenda Jolley 89.75 7.55 8.41%
Alan Duval 84.25 7.10 8.43%
Keith Maynard 98.12 8.28 8.44%
David Barnard 96.38 8.18 8.49%
Ann Dibben 110.50 9.45 8.55%
Julian Fell 115.50 9.89 8.56%
Dave Ashton 92.38 7.94 8.59%
Jonathan Wynn 115.25 9.91 8.60%
Amey Deshpande 89.75 7.74 8.63%
David Von Geyer 90.50 7.84 8.67%
Jonathan Rawlinson 106.25 9.22 8.67%
Martin May 104.88 9.10 8.68%
Paul Erdunast 115.62 10.09 8.72%
Stuart Solomons 99.50 8.76 8.80%
Andrew Hulme 116.25 10.24 8.81%
David O'Donnell 110.00 9.70 8.81%
John Mayhew 101.38 8.97 8.85%
Nick Wainwright 90.75 8.04 8.86%
Grace Page 103.62 9.27 8.95%
Mike Daysley 98.25 8.86 9.01%
Kirk Bevins 115.62 10.66 9.22%
James Doohan 87.75 8.12 9.25%
Michael Bowden 92.38 8.57 9.28%
Jeffrey Burgin 85.62 7.95 9.29%
Jeff Clayton 105.75 9.97 9.43%
David Thirlwall 88.00 8.38 9.52%
David Law 87.50 8.35 9.54%
Dave Costello 90.38 8.64 9.57%
Stephen Briggs 110.38 10.62 9.62%
Gary Male 93.75 9.02 9.63%
Chris Cummins 107.25 10.35 9.65%
Jeffrey Hansford 102.25 9.97 9.75%
Steve Wood 84.38 8.35 9.90%
John Gray 94.62 9.38 9.91%
Heather Styles 92.12 9.24 10.03%
James Roberts 92.00 9.26 10.07%
Ross Allatt 92.62 9.35 10.10%
Neil Zussman 96.00 9.70 10.10%
Chris Marshall 85.25 8.61 10.10%
James Hurrell 104.75 10.59 10.11%
Jon Corby 107.00 10.86 10.15%
Scott Gillies 101.25 10.32 10.19%
Chris Wills 109.38 11.19 10.23%
Bradley Horrocks 108.00 11.22 10.39%
Daniel Pati 105.00 10.92 10.40%
Paul Harper 92.38 9.64 10.44%
Stu Horsey 91.50 9.58 10.47%
Judith Young 88.38 9.29 10.51%
Maggie Barlow 99.25 10.45 10.53%
Gerry Tynan 108.38 11.43 10.55%
Noel McIlvenny 99.12 10.56 10.66%
Martin Bishop 101.12 10.79 10.67%
Rupert Stokoe 97.00 10.37 10.69%
Dan McColm 117.75 12.69 10.77%
Steven Moir 95.38 10.31 10.81%
Graeme Cole 101.62 11.07 10.89%
Tim Down 89.62 9.84 10.97%
Tom Chafer-Cook 112.00 12.31 10.99%
Danny Hamilton 95.12 10.47 11.01%
John Brackstone 102.75 11.32 11.02%
Matthew Shore 106.25 11.77 11.07%
Stuart Earl 100.88 11.17 11.08%
Paul Gallen 105.75 11.80 11.16%
Mark Deeks 103.00 11.58 11.24%
Andy Noden 103.50 11.64 11.25%
Mike Pullin 94.50 10.63 11.25%
Mark Tournoff 101.12 11.38 11.26%
Chris Davies 111.50 12.59 11.29%
Paul Nixon 111.62 12.61 11.30%
Andrew Naylor 97.62 11.17 11.44%
Wesley Jardine 94.12 10.82 11.50%
Paul Keane 93.00 10.74 11.54%
Samir Pilica 100.12 11.56 11.54%
Jean Webby 92.25 10.73 11.63%
Glen Webb 118.12 13.82 11.70%
Aaron Webber 96.62 11.38 11.78%
Steven Turnbull 78.62 9.33 11.86%
Wendy Roe 97.62 11.60 11.88%
Jordan Barker 89.12 10.59 11.88%
Steve Redfern 86.12 10.25 11.90%
Chris Thorn 95.00 11.40 12.00%
Morgan Sauer 96.88 11.87 12.25%
Martin Hurst 103.75 12.73 12.27%
Chris McHenry 80.38 9.92 12.35%
Richard Pay 91.50 11.41 12.47%
Tony Manwani 89.38 11.16 12.48%
Ned Pendleton 83.00 10.37 12.49%
Andy Evans 95.00 11.88 12.51%
Tom Barnes 102.75 12.86 12.52%
Jayne Wisniewski 88.12 11.08 12.58%
Michael Macdonald-Cooper 97.50 12.27 12.58%
Laurence Killen 99.25 12.55 12.64%
Danny Pledger 79.38 10.05 12.66%
Graeme Staples 84.75 10.79 12.73%
Oliver Garner 100.25 12.81 12.78%
Phil Davies 94.62 12.11 12.80%
Junaid Mubeen 98.75 12.68 12.84%
Adam Rolston 81.25 10.46 12.88%
Fiona Titcombe 90.00 11.60 12.89%
Tom Silverlock 88.00 11.45 13.01%
Ryan Taylor 99.00 12.97 13.10%
Martin Gardner 93.25 12.22 13.11%
Lawrie Collingwood 91.75 12.05 13.13%
Mark Davies 105.50 13.87 13.15%
Jim Bentley 94.50 12.46 13.19%
Julia Wilkinson 93.00 12.30 13.22%
Andrew Fenton 91.00 12.26 13.47%
John Cowen 99.88 13.58 13.59%
Liam Shaw 88.50 12.05 13.62%
John Hunt 98.50 13.43 13.63%
Eileen Taylor 94.25 12.89 13.68%
Peter Lee 100.12 13.72 13.71%
Sweyn Kirkness 95.62 13.28 13.89%
Andrew Jackson 95.50 13.34 13.97%
Nik Von Uexkull 93.62 13.11 14.01%
Lee Hartley 101.38 14.43 14.23%
David Edwards 92.12 13.18 14.30%
Moose Rosser 104.75 15.06 14.38%
Tony Warren 89.00 12.87 14.47%
Suzi Purcell 85.75 12.49 14.56%
Stewart Holden 108.75 15.86 14.58%
Conrad Teixeira 100.75 14.79 14.68%
Michael Pfeffer 96.25 14.32 14.88%
Andy McGurn 87.75 13.13 14.96%
Brian Selway 93.25 13.98 14.99%
Jonathan Coles 93.25 14.11 15.13%
Jonathan Liew 94.25 14.31 15.18%
John Shaw 96.75 14.69 15.19%
Charlie Reams 102.50 15.59 15.21%
Matthew Tassier 99.88 15.28 15.30%
Steven Briers 105.38 16.44 15.60%
Andrew Macleod 101.12 16.13 15.95%
Jimmy Gough 97.75 16.30 16.68%
Tia Corkish 87.12 14.66 16.82%
Robin Johnson 93.12 15.99 17.17%
Carl Williams 88.50 15.39 17.39%
Eoin Jackson 97.25 17.14 17.62%
Corrina Attwood 85.12 15.02 17.64%
John Osmond 88.88 16.31 18.35%
Dave Taylor 86.38 15.92 18.43%
Kevin Thurlow 96.12 17.80 18.52%
Is this just up to series 81 then?
Re: Ask Graeme?
Posted: Wed Jul 22, 2020 7:17 pm
by Graeme Cole
Ian Volante wrote: ↑Wed Jul 22, 2020 12:04 pm
Is this just up to series 81 then?
Yes, sorry, I should have mentioned. I update the database after the end of each series.
Your range is 29 and your coefficient of variation is 9.68%.
Re: Ask Graeme?
Posted: Thu Jul 23, 2020 12:40 pm
by Ian Volante
Graeme Cole wrote: ↑Wed Jul 22, 2020 7:17 pm
Ian Volante wrote: ↑Wed Jul 22, 2020 12:04 pm
Is this just up to series 81 then?
Yes, sorry, I should have mentioned. I update the database after the end of each series.
Your range is 29 and your coefficient of variation is 9.68%.
My apparently default position of upper middle.
Re: Ask Graeme?
Posted: Thu Jul 23, 2020 8:58 pm
by Noel Mc
What is the most common final score in the current format?
Re: Ask Graeme?
Posted: Thu Jul 23, 2020 8:59 pm
by Noel Mc
Single player and then final result.
Re: Ask Graeme?
Posted: Thu Jul 23, 2020 10:04 pm
by Graeme Cole
From the start of the current 15-round format to the end of series 81...
Single player: 78 is the most common score (68 times) followed by 80 and 82 (64 times each).
Final score (ignoring order, writing the winner's score first in all cases): 74-69 (6 times) followed by 78-62, 80-70 and 98-77 (5 times each).
Re: Ask Graeme?
Posted: Mon Jul 27, 2020 6:31 am
by Noel Mc
Graeme Cole wrote: ↑Thu Jul 23, 2020 10:04 pm
From the start of the current 15-round format to the end of series 81...
Single player: 78 is the most common score (68 times) followed by 80 and 82 (64 times each).
Final score (ignoring order, writing the winner's score first in all cases): 74-69 (6 times) followed by 78-62, 80-70 and 98-77 (5 times each).
Awesome, thanks
One more:
What is the furthest, starting at round one, that two games have got, being identical in their selections.
- With letter/number order in a round important
- Without letter/number order in a round important
Re: Ask Graeme?
Posted: Mon Jul 27, 2020 6:53 pm
by Graeme Cole
Noel Mc wrote: ↑Mon Jul 27, 2020 6:31 am
What is the furthest, starting at round one, that two games have got, being identical in their selections.
- With letter/number order in a round important
- Without letter/number order in a round important
No two letters selections on Countdown have ever been exactly identical (same letters in the same order). Nine selections match nine other selections in their first eight letters, though - for example,
round 2 of episode 2085 and
round 12 of episode 3853.
If we just look at R1 selections and find the longest matching prefix between two such selections, we find that episodes
1055 and
3721 were identical in their first seven letters, as were episodes
3948 and
4827.
If we disregard the order of the letters within a selection, there are loads of episodes with the same R1 selection as some other episode (I make it 444, if you count each occurrence of the same R1 selection separately).
There is only one instance of two episodes having the same R1 selection and the same R2 selection. These were episodes
153 and
154, so it was probably a shuffling failure rather than a remarkable coincidence.
Re: Ask Graeme?
Posted: Mon Jul 27, 2020 7:59 pm
by Noel Mc
You're like a wizard.
Statistically, it shouldn't be surprising that there have never been two identical rounds, but somehow it is.