Ask Graeme?

Gavin Chipper · Post by **Gavin Chipper** » Sat May 04, 2013 4:51 pm

James Robinson wrote:
Peter Mabey wrote:We know that (notwithstanding Nick's remarks) several games are recorded in a day.

Eileen didn't seem to be her usual self yesterday - would that have been at the end of a long day for her?
Is there any evidence on how other multiple winners have fared at the end of a whole day recording?
Well no, because the end of the day was Wednesday's show, so it was only the 2nd game of the day was yesterday's show, and she seemed pretty competent on Thursday's show, but maybe it's just all those shows that have started to take their toll............

I think normally Monday to Friday is one recording day, which is why a DC guest is normally on from Monday to Friday, but that's not the case right now. Basically, if you want to go through old records, the best way to work out where in a recording day a game is, look at how many games the guest has done.

Graeme Cole · Post by **Graeme Cole** » Sun May 05, 2013 12:16 pm

Johnny Canuck wrote:Apologies if I'm asking too many questions, and sincere thanks for answering my earlier ones, but...

What is the most frequent max score for a 15-round game, and would it be possible to make a graph featuring number of games vs. max score?

The most frequent max in all "Old 15" games is 119, which occurred in 213 games out of 2596.

Frequencies of maxes:

Cumulative frequency (percentile):

Dave Preece · Post by **Dave Preece** » Mon May 06, 2013 12:47 pm

Neil Collins wrote:More of a trivia question than a stats question...

What is it that makes me unique amongst all Countdown contestants?

Played two matches and was never in front?

Dave Preece · Post by **Dave Preece** » Mon May 06, 2013 12:54 pm

Graeme Cole wrote:
Cumulative frequency (percentile):

Sorry to sound really thick (really), but I don't understand this graph?

#layplease

Chris Davies · Post by **Chris Davies** » Mon May 06, 2013 1:24 pm

Dave Preece wrote:
Sorry to sound really thick (really), but I don't understand this graph?

#layplease

The graph shows the percentage of games that had equal to or below a particular maximum score. So it can be seen that 50% of games had a maximum of 125 points or below.

Dave Preece · Post by **Dave Preece** » Mon May 06, 2013 2:38 pm

Chris Davies wrote:
Dave Preece wrote:
Sorry to sound really thick (really), but I don't understand this graph?

#layplease
The graph shows the percentage of games that had equal to or below a particular maximum score. So it can be seen that 50% of games had a maximum of 125 points or below.

H, thanks Chris, apre iate that (was expecting some 'clever' comments), cheers

Andy Platt · Post by **Andy Platt** » Mon May 06, 2013 3:00 pm

Dave Preece wrote:#layplease

I think your chances will be better on pof.com or match.com than on here mate

Rhys Benjamin · Post by **Rhys Benjamin** » Mon May 06, 2013 10:33 pm

Dave Preece wrote:
Graeme Cole wrote:
Cumulative frequency (percentile):

Sorry to sound really thick (really), but I don't understand this graph?

#layplease

Did you ever take O-Level/GCSE Maths? Simple cumulative frequency graph.

Dave Preece · Post by **Dave Preece** » Mon May 06, 2013 11:57 pm

Andy Platt wrote:
Dave Preece wrote:#layplease
I think your chances will be better on pof.com or match.com than on here mate

LOL

Like your sense of humour Andy!

Dave Preece · Post by **Dave Preece** » Tue May 07, 2013 12:03 am

Rhys Benjamin wrote:
Dave Preece wrote:
Graeme Cole wrote:
Cumulative frequency (percentile):

Sorry to sound really thick (really), but I don't understand this graph?

#layplease
Did you ever take O-Level/GCSE Maths? Simple cumulative frequency graph.

I did GCSE maths yeh, I only got a 'D' from middle band! I was told this was more or less max (I imagine 'C' was 100%), but have learnt far more about maths in the 23 years since, but not including this type of graph...

One thing I didn't ever have to learn, however, was to have the BALLS to ask when I didn't get something, sonny!!!

Ryan Taylor · Post by **Ryan Taylor** » Tue May 07, 2013 7:29 am

Dave Preece wrote:
Rhys Benjamin wrote: Did you ever take O-Level/GCSE Maths? Simple cumulative frequency graph.
I did GCSE maths yeh, I only got a 'D' from middle band! I was told this was more or less max (I imagine 'C' was 100%), but have learnt far more about maths in the 23 years since, but not including this type of graph...

One thing I didn't ever have to learn, however, was to have the BALLS to ask when I didn't get something, sonny!!!

You shouldn't have felt obliged to reply to that comment from such an ill-mannered and obnoxious child.

Dave Preece · Post by **Dave Preece** » Tue May 07, 2013 11:52 am

LOL!!!

Dave Preece · Post by **Dave Preece** » Tue May 07, 2013 11:55 am

Graeme Cole wrote:
Johnny Canuck wrote:Apologies if I'm asking too many questions, and sincere thanks for answering my earlier ones, but...

What is the most frequent max score for a 15-round game, and would it be possible to make a graph featuring number of games vs. max score?
The most frequent max in all "Old 15" games is 119, which occurred in 213 games out of 2596.

Frequencies of maxes:

Cumulative frequency (percentile):

Graham, would he first graph look much more 'bell curve' shaped if the 9's scored 9 points instead of 18?

Andy Platt · Post by **Andy Platt** » Tue May 07, 2013 12:07 pm

Good question, is it possible to apply flat scoring to the database?

I wouldn't mind seeing Dave's hypothetical bell.

Jon O'Neill · Post by **Jon O'Neill** » Tue May 07, 2013 12:12 pm

Andy Platt wrote:Good question, is it possible to apply flat scoring to the database?

I wouldn't mind seeing Dave's hypothetical bell.

I reckon a bell with a little dip for impossible numbers or something.

I wouldn't mind seeing the graph for maxes-only scoring.

Liam Tiernan · Post by **Liam Tiernan** » Tue May 07, 2013 8:54 pm

Rhys Benjamin wrote:
Did you ever take O-Level/GCSE Maths? Simple cumulative frequency graph.

It may be easy for you to recognise, Rhys, because you've studied these in the last year or so. Try going 20-odd years without seeing one and it won't come to you so easily. By the time you're Daves age, unless you find work in an field where a good knowledge of maths is useful, the odds are that you'll have forgotten at least half the maths you learn over the next three years.

Gavin Chipper · Post by **Gavin Chipper** » Tue May 07, 2013 9:28 pm

Jon O'Neill wrote:
Andy Platt wrote:Good question, is it possible to apply flat scoring to the database?

I wouldn't mind seeing Dave's hypothetical bell.
I reckon a bell with a little dip for impossible numbers or something.

I wouldn't mind seeing the graph for maxes-only scoring.

Stepdown too.

Dave Preece · Post by **Dave Preece** » Wed May 08, 2013 11:24 pm

Andy Platt wrote:Good question, is it possible to apply flat scoring to the database?

I wouldn't mind seeing Dave's hypothetical bell.

Is it bad that i felt a little aroused [joking], after reading that?

Philip Wilson · Post by **Philip Wilson** » Thu May 30, 2013 10:20 pm

Hi, so is it ok to ask a question about a certain contestant here? This has been bugging me for weeks now!!

Graeme Cole · Post by **Graeme Cole** » Thu May 30, 2013 10:45 pm

Philip Wilson wrote:Hi, so is it ok to ask a question about a certain contestant here? This has been bugging me for weeks now!!

Don't see why not.

Philip Wilson · Post by **Philip Wilson** » Fri May 31, 2013 11:11 pm

Ok, well a few weeks ago while the closing credits were rolling I spotted the two contestants were chatting to each other, which seemed rather good and a change from sitting there which can appear a bit 'awkward'. Anyway that's just my view, but it reminded me of a contestant a few years ago who used to do it every time. So therefore being new here I just tried to investigate further and try and find this contestant's name in the lists, as I have no idea who he was now, but was just interested in when it was etc, human nature stuff I guess.
Well I thought it would be really easy to find him in the list of shows but I just can't and I wondered if you know who it was!

I know he won several times, probably made it to an octochamp and he definitely came back for that series' finals, which makes it all the stranger that I can't find any info. I think he was on during Des O's or the beginning of Jeff's time, but not totally sure about that, but he didn't come back for the CofC in January. Oh and if it's any help he looked a bit like Shane Long [footballer] and was a Fulham fan!
Who was he!?!?!

James Robinson · Post by **James Robinson** » Fri May 31, 2013 11:22 pm

Philip Wilson wrote:Ok, well a few weeks ago while the closing credits were rolling I spotted the two contestants were chatting to each other, which seemed rather good and a change from sitting there which can appear a bit 'awkward'. Anyway that's just my view, but it reminded me of a contestant a few years ago who used to do it every time. So therefore being new here I just tried to investigate further and try and find this contestant's name in the lists, as I have no idea who he was now, but was just interested in when it was etc, human nature stuff I guess.
Well I thought it would be really easy to find him in the list of shows but I just can't and I wondered if you know who it was!

I know he won several times, probably made it to an octochamp and he definitely came back for that series' finals, which makes it all the stranger that I can't find any info. I think he was on during Des O's or the beginning of Jeff's time, but not totally sure about that, but he didn't come back for the CofC in January. Oh and if it's any help he looked a bit like Shane Long [footballer] and was a Fulham fan!
Who was he!?!?!

That sounds like Jeffrey Burgin from Series 61 to me.

Philip Wilson · Post by **Philip Wilson** » Fri May 31, 2013 11:49 pm

James Robinson wrote: That sounds like Jeffrey Burgin from Series 61 to me.

That's him! Thanks James, and so quick! I'm sure I googled that name before and someone else appeared. So finally beaten by Innis then

Jack Worsley · Post by **Jack Worsley** » Sun Jun 09, 2013 5:27 pm

This may have already been asked but has any numbers game ever come up more than once in the show's history? (Same numbers and target but order of the numbers doesn't matter).

Dave Preece · Post by **Dave Preece** » Wed Jun 12, 2013 1:56 am

Jack Worsley wrote:This may have already been asked but has any numbers game ever come up more than once in the show's history? (Same numbers and target but order of the numbers doesn't matter).

I very much doubt that!

Johnny Canuck · Post by **Johnny Canuck** » Wed Jun 12, 2013 10:20 am

Dave Preece wrote:
Jack Worsley wrote:This may have already been asked but has any numbers game ever come up more than once in the show's history? (Same numbers and target but order of the numbers doesn't matter).
I very much doubt that!

CoC XII GF, Round 14: 100 50 25 75 3 4 -> 880
Series 63 4th QF, Round 5: 50 100 75 25 4 3 -> 880

(Incidentally, it is impossible to get 10 points on this particular numbers game.

)

This is the only one I know off the top of my head, but I'm betting there are at least a few others.

Dave Preece · Post by **Dave Preece** » Thu Jun 13, 2013 3:43 am

I stand corrected, sorry!

Are the odds on that somewhere near 50,000/1?????????

Jon Corby · Post by **Jon Corby** » Thu Jun 13, 2013 9:03 am

Dave Preece wrote:I stand corrected, sorry!

Are the odds on that somewhere near 50,000/1?????????

I started to think about the calculation, but you'd want to weight it by the selection type. Clearly it's much more likely that a 4 large game would be repeated since 4 of the 7 components are guaranteed to be the same from the off, but then 4 large isn't picked very often. Then I thought I'd just wait for somebody else to do it.

Guy Barry · Post by **Guy Barry** » Thu Jun 13, 2013 10:39 am

I haven't done the calculation either, but given that there must have been something like 13,000 numbers games in the entire history of Countdown (I'm sure someone will supply the correct figure), it doesn't strike me as particularly remarkable that two of them turned out to be the same. It's a version of the so-called birthday problem - there are 365 possible birthdays, and yet you only need 57 people in a room to have a 99% probability that two have the same birthday.

Andy Platt · Post by **Andy Platt** » Thu Jun 13, 2013 1:12 pm

Wouldn't surprise me if there were more as I've had the same numbers and target in the same numbers attack (20 rounds) before a couple of times on apterous (although yeah, 4 large)

Guy Barry · Post by **Guy Barry** » Thu Jun 13, 2013 3:08 pm

OK, I've done a few rough calculations now. For simplicity I'll just look at the "one large" picks. From the graph posted by Graeme some time ago it looks as though they come up about 70% of the time, so that's about 9000 games played in total.

For further simplicity I'll just look at the ones where no number is repeated. Statistically these should come up around 52% of the time, so that's about 4680 games. There are four possible choices of large number, 252 (=10C5) choices of small numbers, and 900 choices of target. That comes to 907200 possible combinations. Using the approximate formula given on the Wikipedia page, the probability that two of these games have yielded the same combination is close to 1 - exp(-(4680^2)/(2*907200)), or around 99.9994%.

So I don't think anyone should be remotely surprised!

Graeme Cole · Post by **Graeme Cole** » Fri Jun 14, 2013 9:48 pm

Better late than never...

Dave Preece wrote:Graham, would he first graph look much more 'bell curve' shaped if the 9's scored 9 points instead of 18?

Andy Platt wrote:I wouldn't mind seeing Dave's hypothetical bell.

Jon O'Neill wrote:I reckon a bell with a little dip for impossible numbers or something.

Gavin Chipper · Post by **Gavin Chipper** » Sun Jun 16, 2013 4:38 pm

I don't think that dip's probably got much to do with impossible numbers.

Dave Preece · Post by **Dave Preece** » Tue Jun 18, 2013 1:30 am

Me neither Gavin!

Dave Preece · Post by **Dave Preece** » Tue Jun 18, 2013 1:32 am

Dave Preece wrote:
Graeme Cole wrote:
Johnny Canuck wrote:Apologies if I'm asking too many questions, and sincere thanks for answering my earlier ones, but...

What is the most frequent max score for a 15-round game, and would it be possible to make a graph featuring number of games vs. max score?
The most frequent max in all "Old 15" games is 119, which occurred in 213 games out of 2596.

Frequencies of maxes:

Cumulative frequency (percentile):

Graham, would he first graph look much more 'bell curve' shaped if the 9's scored 9 points instead of 18?

Thought so...

{genius LOL}

Dave Preece · Post by **Dave Preece** » Tue Jun 25, 2013 1:30 am

How would YOU rank the top 10 or more contestants?

Dave Preece · Post by **Dave Preece** » Tue Jun 25, 2013 1:30 am

Wins, %wins, comp wins, %comp wins, ???

Adam Gillard · Post by **Adam Gillard** » Tue Jun 25, 2013 6:28 pm

Dave Preece wrote:How would YOU rank the top 10 or more contestants?

Dave Preece wrote:Wins, %wins, comp wins, %comp wins, ???

Shag, Snog, Stalk, Marry, Maim, Kill, Avoid, Befriend, Berate, Worship?

Dave Preece · Post by **Dave Preece** » Tue Jun 25, 2013 11:02 pm

I guess the answer is still: I don't know?

Matt Morrison · Post by **Matt Morrison** » Wed Jun 26, 2013 7:32 am

I think it was Angus Deayton.

Graeme Cole · Post by **Graeme Cole** » Wed Jun 26, 2013 11:40 pm

Dave Preece wrote:How would YOU rank the top 10 or more contestants?

Dave Preece wrote:Wins, %wins, comp wins, %comp wins, ???

Dave Preece wrote:I guess the answer is still: I don't know?

Yep.

However, I'm curious - what does "comp wins" stand for? Competitive? Competition? Competent? Compost?

Dave Preece · Post by **Dave Preece** » Thu Jun 27, 2013 6:03 am

Graeme Cole wrote:
Dave Preece wrote:How would YOU rank the top 10 or more contestants?

Dave Preece wrote:Wins, %wins, comp wins, %comp wins, ???

Dave Preece wrote:I guess the answer is still: I don't know?
Yep.

However, I'm curious - what does "comp wins" stand for? Competitive? Competition? Competent? Compost?

So... Am I assuming that this is one of the most challenging questions you have been asked in this tread - hence to non-answer????

Dave Preece · Post by **Dave Preece** » Fri Jun 28, 2013 3:56 am

PS

Comp is short for competition.

Neil Collins · Post by **Neil Collins** » Tue Jul 02, 2013 8:39 pm

Dave Preece wrote:
Neil Collins wrote:More of a trivia question than a stats question...

What is it that makes me unique amongst all Countdown contestants?
Played two matches and was never in front?

Just realised that I never came back on this one.

I was the only contestant to take part in Countdown Masters who had never won a game, having lost in my original appearance the previous year. Richard made a big thing about my appearance on Masters being an experiment to see what would happen if a losing contestant was put up against a previous series champion (Brian Hudson)

The truth was though that the contestant who had been booked to appear against Brian hadn't turned up, and as I worked in the pub that John Meade (God rest his soul) drank in, I was dragged in with about an hours notice to fill in.

Graeme Cole · Post by **Graeme Cole** » Tue Jul 02, 2013 11:36 pm

Neil Collins wrote:
Dave Preece wrote:
Neil Collins wrote:More of a trivia question than a stats question...

What is it that makes me unique amongst all Countdown contestants?
Played two matches and was never in front?
Just realised that I never came back on this one.

I was the only contestant to take part in Countdown Masters who had never won a game, having lost in my original appearance the previous year. Richard made a big thing about my appearance on Masters being an experiment to see what would happen if a losing contestant was put up against a previous series champion (Brian Hudson)

The truth was though that the contestant who had been booked to appear against Brian hadn't turned up, and as I worked in the pub that John Meade (God rest his soul) drank in, I was dragged in with about an hours notice to fill in.

Heh, that's quite interesting. I suppose the result of the "experiment" was more balanced than might have been expected - looks like it was only the numbers games that won it for Brian.

sean d · Post by **sean d** » Tue Jul 09, 2013 12:25 pm

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.

Andy Platt · Post by **Andy Platt** » Tue Jul 09, 2013 1:50 pm

Apterous stats:

Max of 3 = negligibly small / Max of 4 = 0.2% / Max of 5 = 3% / Max of 6 = 19% / Max of 7 = 42% / Max of 8 = 30% / Max of 9 = 5%

Dave Preece · Post by **Dave Preece** » Wed Jul 10, 2013 4:51 pm

There are more (I'd go as far to say, far more) 9's on the show IMO.

COULD be more 4's too, others will be similar, just a guess though!

Graeme Cole · Post by **Graeme Cole** » Wed Jul 10, 2013 11:46 pm

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.

Andy Platt · Post by **Andy Platt** » Thu Jul 11, 2013 8:29 am

Looks like the conclusion is that the letters on countdown are 'shuffled' very slightly better than apterous's shuffle algorithm, but Charlie has also done a bloody good job of replicating it.

JackHurst · Post by **JackHurst** » Thu Jul 11, 2013 8:41 am

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.

sean d · Post by **sean d** » Thu Jul 11, 2013 5:01 pm

Thanks gents.

I'd suggest it is fairly significant. Andy's figures suggest 77% 7-or-better max on Apterous, the Countdown since 2010 figures suggest 84% 7-or-better max. Doing a quick weighted average (with 18 pts for a 9) I think the average max score per letters round on the show since 2010 is 7.807. On Apterous it's 7.538. That's about 3 points a game. Jack makes a good point too about the tendancy for contestants on the show (non-Apterites mainly!) to strangle rounds with 6 consonants by default. (I'm assuming 5C/4V yields a higher average max).
Anecdotally the letters on the show seem less random than Apterous... less batches of 3 A's or 4 L's, an uncommon amount of selections with 9 different letters, etc. But then perhaps I see more unfriendly selections on Apterous cos I might rattle through 50-100 letters rounds in a day.

Gavin Chipper · Post by **Gavin Chipper** » Thu Jul 11, 2013 8:40 pm

Andy Platt wrote:Looks like the conclusion is that the letters on countdown are 'shuffled' very slightly better than apterous's shuffle algorithm, but Charlie has also done a bloody good job of replicating it.

Or worse.

Edit - Basically I think it's probably a bit harsh to say they're shuffled better on Countdown, given that Apterous is trying to replicate Countdown rather than merely get the highest maxes possible. And obviously Countdown is always going to be nearer Countdown than Apterous is.

Peter Thomas · Post by **Peter Thomas** » Thu Jul 18, 2013 5:36 pm

On 10 July, the same numbers game target (831) came up twice in a row. Has this ever happened before?

Johnny Canuck · Post by **Johnny Canuck** » Fri Jul 19, 2013 12:24 am

A while back, I asked the same question. I believe 10 games have had two instances in a row of the same target... or 11 including this most recent one.

If Graeme will kindly accept ANOTHER two questions from me:
* What word has most frequently been used as a conundrum?
* What is the most frequently used conundrum scramble?

Andy Platt · Post by **Andy Platt** » Fri Jul 19, 2013 5:41 am

Interesting one ^

As part of my prep for going on TV I read through a lot of the conundrums that had been used in the past and I think the answer to part A will be INTUITION but keen to hear what Graeme has to say.

Edit: Hmm, only 4 times after checking, so probably not

Gavin Chipper · Post by **Gavin Chipper** » Fri Jul 19, 2013 9:43 am

Johnny Canuck wrote:A while back, I asked the same question.

So combining this all together, has the same question ever been asked twice in a row in this thread?

Ian Volante · Post by **Ian Volante** » Fri Jul 19, 2013 12:37 pm

Has the same question ever been asked twice in a row on this thread?

Dave Preece · Post by **Dave Preece** » Sat Jul 20, 2013 12:26 am

Has the same question ever been asked thrice in a row on this thread?

Guy Barry · Post by **Guy Barry** » Sat Jul 20, 2013 5:36 am

Dave Preece wrote:Has the same question ever been asked thrice in a row on this thread?

That wasn't the same as the last two questions, so I presume the answer's no.