Duel revision

[Charlie Reams]

Thanks to everyone who's filled out the Duel survey (closes tomorrow!). This thread is for raising any other points before the Duel gets overhauled at the end of this month.

My first question is, how would people feel about using total conundrum time as a tie-breaker?

[Ryan Taylor]

My first question is, how would people feel about using total conundrum time as a tie-breaker?

I would be fine with this. It's fair and rewards those who are good quick conundrum solvers.

[Innis Carson]

Personally I don't think I'd like this much, it's nice to be able to use the full time in duels to deliberate and check your answer and stuff without fear of it costing you. But I suppose as long as it's made clear that conundrum time matters, it's fair.

[Ian Volante]

Sounds good. I'd be happy to extend that to the total time, not just conundrum time. This may help in duels with no conundrum.

[Kai Laddiman]

I have a question: did you guess which name I used on that form?

[Graeme Cole]

My first question is, how would people feel about using total conundrum time as a tie-breaker?

I'd be mildly annoyed at this, but only because I'm not terribly good at conundrums. Objectively, I think it's a very good idea.

How would you deal with the case where two players got the same number of points overall, but one player solved more conundrums than the other? Would you count an unsolved conundrum as 30 seconds (or whatever the time limit was for that round)?

[Ryan Taylor]

How would you deal with the case where two players got the same number of points overall, but one player solved more conundrums than the other? Would you count an unsolved conundrum as 30 seconds (or whatever the time limit was for that round)?

I was under the impression that this was only for an all-Conundrum duel (like todays). It would be a bad idea to apply the conundrum times to a duel in the format of LLCLLC for example.

[Charlie Reams]

How would you deal with the case where two players got the same number of points overall, but one player solved more conundrums than the other? Would you count an unsolved conundrum as 30 seconds (or whatever the time limit was for that round)?

Yep.

It would be a bad idea to apply the conundrum times to a duel in the format of LLCLLC for example.

Why?

[Ryan Taylor]

[quote="Charlie "that's right, I'm at Google now" Reams"]

It would be a bad idea to apply the conundrum times to a duel in the format of LLCLLC for example.

Why?[/quote]

Example duel: LLLLLLLLLLC

A: 8, 7, 7, 8, 6, 7, 8, 7, 6, 8, 0 = 72
B: 7, 6, 6, 7, 5, 6, 7, 6, 5, 7, 10 = 72

A scores 72 maxing every single letter round then fluffs a diamond conundrum.

B comes along and scores 72 with no maxes except for solving the diamond conundrum on 29.9 seconds.

If I was player A I would feel pretty hard done by.

[Charlie Reams]

[quote="Charlie "that's right, I'm at Google now" Reams"]

It would be a bad idea to apply the conundrum times to a duel in the format of LLCLLC for example.

Why?

Example duel: LLLLLLLLLLC

A: 8, 7, 7, 8, 6, 7, 8, 7, 6, 8, 0 = 72
B: 7, 6, 6, 7, 5, 6, 7, 6, 5, 7, 10 = 72

A scores 72 maxing every single letter round then fluffs a diamond conundrum.

B comes along and scores 72 with no maxes except for solving the diamond conundrum on 29.9 seconds.

If I was player A I would feel pretty hard done by.[/quote]
I see what you mean, but it's hardly any different to what can (and regularly does) happen already:

A: 8, 7, 7, 8, 6, 7, 8, 7, 6, 7, 0 = 71
B: 7, 6, 6, 7, 5, 6, 7, 6, 5, 7, 10 = 72

Player A gets almost all maxes but Player B finishes higher. The all-or-nothing nature of the conundrum is brutal but that's the nature of the beast. I've toyed with a variant in which you get fewer points the longer you take to solve, but I wouldn't want to make that the default.

[Graeme Cole]

For what it's worth, I have a verrrrrrry dim and distant memory of Richard Whiteley explaining how two players' seed numbers had been decided. I think they were level on games won, aggregate score, highest score, neatest haircut and so on, so it went down to fastest conundrum time.

Am I imagining this? Maybe he was just explaining what would hypothetically happen in that situation. I think it was in the early 1990s. If true, it would seem that conundrums have long been considered "more valuable" than other rounds when everything else is equal.

[Charlie Reams]

For what it's worth, I have a verrrrrrry dim and distant memory of Richard Whiteley explaining how two players' seed numbers had been decided. I think they were level on games won, aggregate score, highest score, neatest haircut and so on, so it went down to fastest conundrum time.

Am I imagining this? Maybe he was just explaining what would hypothetically happen in that situation. I think it was in the early 1990s. If true, it would seem that conundrums have long been considered "more valuable" than other rounds when everything else is equal.

The rules have changed numerous times over the years, usually without warning. For example in Series 60, Jimmy Gough and Cate Henderson both finished on 8 wins and 782 points, but Cate was made #2 seed and Jimmy #3 since she'd solved more conundrums. The previous time this had happened, back in Series 46, Ben Wilson was given the higher seeding because his top score was higher. Ho-hum.

[Ryan Taylor]

Example duel: LLLLLLLLLLC

A: 8, 7, 7, 8, 6, 7, 8, 7, 6, 8, 0 = 72
B: 7, 6, 6, 7, 5, 6, 7, 6, 5, 7, 10 = 72

A scores 72 maxing every single letter round then fluffs a diamond conundrum.

B comes along and scores 72 with no maxes except for solving the diamond conundrum on 29.9 seconds.

If I was player A I would feel pretty hard done by.

I see what you mean, but it's hardly any different to what can (and regularly does) happen already:

A: 8, 7, 7, 8, 6, 7, 8, 7, 6, 7, 0 = 71
B: 7, 6, 6, 7, 5, 6, 7, 6, 5, 7, 10 = 72

Player A gets almost all maxes but Player B finishes higher. The all-or-nothing nature of the conundrum is brutal but that's the nature of the beast. I've toyed with a variant in which you get fewer points the longer you take to solve, but I wouldn't want to make that the default.

Hmm yeah, conundrums are brutal. I still think that it would be a good idea for an all-conundrum duel (which is what I thought you meant in the first place) because they are a test of conundrums, but when the conundrum(s) are few in number then I still think it is harsh for the fastest solve to be a deciding factor.

[Graeme Cole]

What about number of maxes, then total conundrum time? It deals with the case Ryan describes, but still makes reasonably sure of breaking the tie when many people get the same number of points and maxes (or max the whole duel, like yesterday).

[Ben Wilson]

I've toyed with a variant in which you get fewer points the longer you take to solve, but I wouldn't want to make that the default.

Please do this, it would be awesome. Also-

The previous time this had happened, back in Series 46, Ben Wilson was given the higher seeding because his top score was higher. Ho-hum.

That might actually have been a second tie-break as me and Phil Wass both got 5/6 conundrums.

[Matt Bayfield]

I would be strongly against Conundrum time being used as any sort of tie-breaker in a Duel, for a few reasons. These are mainly to limit my personal frustration at apterous to at least a manageable, rather than a stratospheric level.

Firstly, whilst I am reasonable at solving conundrums against a patzer (74% in Ascension), I am not particularly quick to solve them and buzz. In well over 5000 conundrums, I have only twice managed to solve in a time less than 1.0 seconds, and my typical time for a conundrum I spot "straight away" is in the range 1.5 to 2.5 seconds. That's way too slow to be competitive against any of the top players round here, so I would probably never win a Duel again - which I would find a little disappointing.

Secondly, using conundrum time as a tie-breaker is going to encourage people to buzz in straight away and increase the chances of them getting wrong a conundrum they would otherwise get right. For example, many more people will buzz in with something like CONQUEROR when the actual solution is RECONQUER. Alternatively, they will buzz in with the plausible-sounding wrong solution (I dunno, can't think of a particularly plausible example here, but say EXCERTION) when a few seconds' more thinking time would enable them to spot the correct solution (EXCRETION). Whilst one of the beauties of apterous gameplay is the degree of frustration it can cause, that is way too much frustration for me to find enjoyable.

Thirdly, as previously stated by others, using conundrum time places even more weight on being good at Conundrums (which are only 1/15 of a Normal 15-rounder, not that apterous should be trying to mirror Countdown). If the duel format were NC (add as many letters rounds to that as you like), then anyone solving the conundrum and fluffing the numbers would automatically be placed higher than someone solving the numbers and fluffing the conundrum. In short Duels (6 or fewer rounds) this could sometimes result in finishing 20 places lower in the day's standings.

Just my $0.02. I don't expect people to agree!

[Michael Wallace]

Is there a 'problem' with having tied duels? I agree with a fair bit of what Matt has to say - going down to a conundrum time tie-break just seems to be another way to help the super awesome people win, and to make it harder for us lamers to get our occasional bit of glory.

Conundrums are already (usually) quite overpowered anyway - if it's a standard game then dropping a conundrum is equivalent to missing a shedload of maxes, so adding a further bonus to it seems a bit unnecessary.

[Gavin Chipper]

You could have a separate conundrum at the end to be used as a tie-break, one that doesn't count in the actual score. Do all duels have conundrums anyway (I haven't played for ages)?