Co-Event Suggestion Box

[JackHurst]

TCap's suggestion is the best solution. Graeme specifically added this feature to the most recent atropine release after several people requested it. So that would be:

- When drawing fixtures, favour equal number of wins between players, but (mostly) ignore points
- When producing final rankings, stick with tradition of wins then points.

And to illustrate the drawbacks of using wins and points for Bristol event with 5 rounds, let's imagine we repeated Co:Rea with 4 rexes and 44 prunes.

P1 Rex1
P2 Rex2
P3 Rex3
P4 Rex4
P5 Prune1
...
P48 Prune44

Assume a Rex maxes every round, and a prune scores 0 on every round, eventually solving a tiebreak conundrum.

Assuming the Rexes all miss each other in the Rd1 draw, then how does your tournament pan out every single time?

R2
You get 2 rex on rex games
You get 22 prune on prune games

R3
You still get 2 rex on rex games. The unbeaten rexes facing off, and the beaten rexes facing off
And again 22 prune on prune games

After Rd 3, one of the Rexes is on 1/3 wins! Loads of prunes still on 3/3 wins. It's already a mathematicaly certainty that there's going to be a prune finish above a rex at this point because you still have unbeaten prunes playing each other in rd 4.


I would hope hosts consider to start using the "Wins only" drawing for short events, and especially for Bristol It's not perfect, but it's the best option.

[Fiona T]

Yeah wins have to count or there's no incentive to take any risks in your games. But yep - will be interesting to see how the draws look at Brum!

[Fiona T]

For a Lincoln event, where there are two games happening on the same physical table, if possible make sure that the card sets used look different to avoid cross contamination!

[JackHurst]

I'm genuinely curious to know from hosts what would happen if a player deliberately beat prune 1-0 (or 0-0 on a TB) in order to boost their overall chances of winning an event.

I dunno if I'd ever have the balls to do it, but for example if Oxford were to have 5 or 6 more sign ups and I got prune in RD1, then I believe the optimal move would be to beat Prune on a tiebreak

[Callum Todd]

I'm genuinely curious to know from hosts what would happen if a player deliberately beat prune 1-0 (or 0-0 on a TB) in order to boost their overall chances of winning an event.

I dunno if I'd ever have the balls to do it, but for example if Oxford were to have 5 or 6 more sign ups and I got prune in RD1, then I believe the optimal move would be to beat Prune on a tiebreak

how many tiebreaks?

[Matthew Brockwell]

On the earlier subject of conundrum mishaps a piece of blu-tack or tape on the back of them to attach them to the table will stop the worst of the mishaps. Also giving them pre-folded should mitigate the "answer is visible" issue. I've wondered too if FOCAL could get some cheap (donated) Android phones and preload Jack's app on them along with a timer for hosts without either the app or a signed-in apto account.

I'm going to go against the grain of this thread and say I really like the 14 rounder as it delivers letters, numbers and conundrums in a 4:2:1 ratio which feels right to me. Also having two conundrums can mitigate bad organiser conundrum judgement as losing out on two buzzer races feels less like bad luck and more like being outplayed. Graeme's excellent post here: http://www.c4countdown.co.uk/viewtopic. ... um#p179160

[Gavin Chipper]

Can all CO-event c4c threads have the title of the top post edited to include the date please?

Edit - Reading and Sheffield are currently the only ones that do this. So not London, Birmingham, Durham and Oxford (though Oxford has the month).

This is obviously a good idea, and to me seems such an uncontroversial no-brainer that I've invoked my mod powers to do it myself.

There's a couple of newer ones around at the moment.

[Fiona T]

*Supernundrums for grand finals only -

assuming they could be displayed in a way that both finalists can see clearly, a good or bad way of deciding what is often a crucial?

*What is a supernundrum?

60 seconds timer rather than 30 seconds

9 not crazy difficult conundrums
first letter of each answer creates a new conundrum - the answer to the supernundrum is this word!

e.g.




Solution - https://ibb.co/8jP7Mxd

Pros: No instabuzz - reasonable chance of mistake in workings
Cons: Inaccessible to non-elite players

I'd really like to trial this !

Edit - maybe needs its own thread?

[Thomas Carey]

I rambled about this here http://www.c4countdown.co.uk/viewtopic. ... um#p200804

We used them a fair bit in lockdown zoom tournaments and it worked well. For the really good players, it meant buzz times separated by about 2 seconds rather than 0.2 - much less dependent on reaction time. For the rest of us we'd usually get them but towards the middle or end of the time. I'm all for seeing it tried (obviously not in heat games). If it doesn't work, we gave it a go, and I think most of us aren't fans of tournaments being decided by reaction tests.

[JackHurst]

Would love to see this given a go. Much more drama and tension than a standard Conundrum. From a gameplay perspective, it's a better test of consistency and composure which is probably what you want in a final. Also can't be Hansforded.

More downsides:
- 10 times more like the setter can make a mistake
- Getting the audience to be fully quiet for 60s might be impossible. Very important for a supernundrum as knocking somebody of their rhythm would affect what sub solution letters they have memorized
- You need a projector and screen for this even if you didn't use it for the rest of the event.
- Not currently openly accessible for people to practise

I used to have a website where you could play these, will look into reviving it.

[Fiona T]

- 10 times more like the setter can make a mistake

In terms of mistakes, the stuff I'm building into roboriley already checks that at an individual level that scrambles match solutions, and that the first letter of the solutions matches the final answer. What it doesn't currently do but feasibly could, is check that there is only one valid solution for any conundrum. In terms of valid conundrums (no regular plurals etc) I don't think I can do that unless someone maintains and is happy to share a full list of valid conundrums.

I used to have a website where you could play these, will look into reviving it.

That would be great

[Fiona T]

So inspired by a question on fb from a co-event newbie as to whether there would be buzzers on every table, I've knocked together this...

http://titcombe.co.uk/fiona/buzzer.html

I think it's pretty similar to the app that Beachy's got on his iphone, but web based so nothing needed to download and should work across different devices, but will require a network connection initially (once your phone's cached it, it will probably work without, maybe...)

I don't know if it's preferable to tapping the table but happy for people to try it, check they're happy it gives accurate results on close taps, and share the link if they want to give people the option of using it. I haven't tried lots of people accessing the page at once, but given the code basically runs in your browser, it should be fine...

Also for simultaneous buzzes, where an adjudicator and var would call a re-do, it's going to announce a winner - but I guess apterous does that!

I think very close buzzes are ok - but happy for people to test and tell me it's not accurate enough! It's ridiculously simple, so might not measure up!

[Gavin Chipper]

Ooh, that looks really good.

[JackHurst]

Looks good Fiona, nice job.

"Player 1 Buzzed" -> " This player buzzed"
"Player 2 Buzzed" -> "This player Buzzed"

Some sort of visual indication that makes it very obvious to the host because I don't think a host would necessarily remember who is P1/P2/Red/Blue when they've got several other things on their mind while reffing the conundrum.

[Eeshan Malhotra]

(This feels like the best thread to post this, but if it needs to be moved that's okay)
Some questions about situations that have arisen while hosting games (at a Lincoln style event) that I definitely don't know the answers to, but possibly aren't codified at all:

1. If a player is picking letters too slow, what should the host to?
My response to this has usually been to gently ask them to speed up, but at some point, should the host just pick a letter arbitrarily? (Assuming the min/max for vowels/consonants doesn't apply in the situation of course)
1b. How slow is too slow to warrant a warning or other intervention?
I don't think anyone wants to actually have a timer that starts each time someone is picking a letter, but if a host is expected to go by gut feel, it would be good to have a target to calibrate that feel to. 5 seconds?

2. Is it okay if someone declares a target not written down and breaks it down rather than building it up?
For example when orally solving a numbers game for 689, they go "689 is 684 plus 5, 684 comes from 57 times 12. The 57 is 50 plus 7 and the 12 is 3 times 4"
They clearly have a correct solution in their head, so I'm inclined to say yeah cool, but not sure if I'm doing the right thing.

3. Do you have to do computations for the player describing their calculation (like Rachel on TV), or ask for the full method without helping them?
If the player goes "(50+9+4) times 7", is the host required to calculate this (mentally or otherwise) and tell the player before they then describe the next step in their method as "then add 1". On one hand, this helps a player who lost track of their computation and isn't sure if the last number needs to be added or subtracted. On the other hand, this is a valid strategy on the show, so why not at Co events. I think both arguments have merits, but I'd really like to know the consensus (as both host and player)

Meta comment: In most games, these situations don't even arise. And most of the times they do, the players and host mutually sort it out amongst themselves. But if these sorts of things are codified on a reference page (a) it may make newer hosts feel confident they can tackle weird situation (b) in the rare situation there's a dispute among players, it's easy to point to a document

[Matt Rutherford]

(This feels like the best thread to post this, but if it needs to be moved that's okay)
Some questions about situations that have arisen while hosting games (at a Lincoln style event) that I definitely don't know the answers to, but possibly aren't codified at all:

1. If a player is picking letters too slow, what should the host to?
My response to this has usually been to gently ask them to speed up, but at some point, should the host just pick a letter arbitrarily? (Assuming the min/max for vowels/consonants doesn't apply in the situation of course)
1b. How slow is too slow to warrant a warning or other intervention?
I don't think anyone wants to actually have a timer that starts each time someone is picking a letter, but if a host is expected to go by gut feel, it would be good to have a target to calibrate that feel to. 5 seconds?

2. Is it okay if someone declares a target not written down and breaks it down rather than building it up?
For example when orally solving a numbers game for 689, they go "689 is 684 plus 5, 684 comes from 57 times 12. The 57 is 50 plus 7 and the 12 is 3 times 4"
They clearly have a correct solution in their head, so I'm inclined to say yeah cool, but not sure if I'm doing the right thing.

3. Do you have to do computations for the player describing their calculation (like Rachel on TV), or ask for the full method without helping them?
If the player goes "(50+9+4) times 7", is the host required to calculate this (mentally or otherwise) and tell the player before they then describe the next step in their method as "then add 1". On one hand, this helps a player who lost track of their computation and isn't sure if the last number needs to be added or subtracted. On the other hand, this is a valid strategy on the show, so why not at Co events. I think both arguments have merits, but I'd really like to know the consensus (as both host and player)

Meta comment: In most games, these situations don't even arise. And most of the times they do, the players and host mutually sort it out amongst themselves. But if these sorts of things are codified on a reference page (a) it may make newer hosts feel confident they can tackle weird situation (b) in the rare situation there's a dispute among players, it's easy to point to a document

01-I'd say about 5 seconds (roughly the same on Apterous), followed by verbal speed-up, then pick a letter.
02-Yeah. They've still done it-as long it's not obvious/take-the-piss-style bodgery
03-If you can keep up, all good. No problem in asking for clarity. Some references to help may be good, but most of this is very situational

[Matt Rutherford]

Some stuff arose at Birmingham that I wanted to dump here.

New Styles Of Drawing
-For Goat, it was Players On Same Number Of Wins Overall drawn against each other, it seemed to work OK. Some preferred it to Swiss, others said either this style (I'm calling it Brum-style draws, and will die on this hill) or Swiss-style would be fine. Jack's view of some more tests for further clarity is I think the right on. More feedback would be welcoming.
-For Co, it was Players On The Same Number Of Wins From The Previous Round Only drawn against each other, and from my taking of the general consensus it didn't seem to work quite as well. Not quite a horrendous pile of poo, but it did give strong players a further bolstering, and (in my view) one event was enough to establish that it doesn't quite work. I tried the drawing styles to see how they've worked-glad I did

Ways Of Resolving Ties In Standings
Had the unusual case at the Goat where two players were tied for 5th and dead-level on wins and points. Was agreed to just leave them level at 5th. In the event this could affect position, and for clarity, I propose separation on the following if possible...
-If they've played each other at any point in the tourney, then the winner of that game goes above
-If not, then whoever has the better spread
-If that's equal, then a single conundrum

These are only my own views. Feedback welcome!

[Fiona T]

01-I'd say about 5 seconds (roughly the same on Apterous), followed by verbal speed-up, then pick a letter.

Disagree here - on telly players pick letters one by one and have plenty of thinking time - none of this CV7 where the letters come out with barely time to write them down. I mean clearly someone can't sit there for five minutes, but allowing a few extra seconds is perfectly ok. By all means prompt if it's dragging on, but I'd never presume to pick a letter - I mean if it gets really silly then call the event host over!

[JackHurst]

01-I'd say about 5 seconds (roughly the same on Apterous), followed by verbal speed-up, then pick a letter.

Disagree here - on telly players pick letters one by one and have plenty of thinking time - none of this CV7 where the letters come out with barely time to write them down. I mean clearly someone can't sit there for five minutes, but allowing a few extra seconds is perfectly ok. By all means prompt if it's dragging on, but I'd never presume to pick a letter - I mean if it gets really silly then call the event host over!

The process I go by as host:
- 5-7ish seconds: hurry them along
- 9ish seconds threaten to pick for them if they don't hurry up
- ??? Seconds randomly pick

It's never got to the third step! People generally pick something once they've been warned twice.

I think it's a none problem. I'd go as far as saying if this has been a problem for anybody on a table, it's more likely due to overly officious hosting.

[Gavin Chipper]

Ways Of Resolving Ties In Standings
Had the unusual case at the Goat where two players were tied for 5th and dead-level on wins and points. Was agreed to just leave them level at 5th. In the event this could affect position, and for clarity, I propose separation on the following if possible...
-If they've played each other at any point in the tourney, then the winner of that game goes above
-If not, then whoever has the better spread
-If that's equal, then a single conundrum

These are only my own views. Feedback welcome!

I'm not sure what better spread means, but I'd use highest score as a tie-break and then 2nd highest score etc.

I'd probably avoid games against each other as it can cause further complications and ambiguities if three or more people are tied. (Edit - plus by avoiding it, players just have an objective performance score to look at.)

[JackHurst]

Ways Of Resolving Ties In Standings
Had the unusual case at the Goat where two players were tied for 5th and dead-level on wins and points. Was agreed to just leave them level at 5th. In the event this could affect position, and for clarity, I propose separation on the following if possible...
-If they've played each other at any point in the tourney, then the winner of that game goes above
-If not, then whoever has the better spread
-If that's equal, then a single conundrum

This is a very sensible resolution. It's a really good point to raise, because although it feels edge case, there's a good chance it could happen one day if we keep having 10+ events a year for many more years.

Graeme might implement this into atropine if you ask nicely. Probably doesn't need to be a configurable rule, can just be one that's uniformly applied to all events.

[Callum Todd]

Ways Of Resolving Ties In Standings
Had the unusual case at the Goat where two players were tied for 5th and dead-level on wins and points. Was agreed to just leave them level at 5th. In the event this could affect position, and for clarity, I propose separation on the following if possible...
-If they've played each other at any point in the tourney, then the winner of that game goes above
-If not, then whoever has the better spread
-If that's equal, then a single conundrum

There is already a rule for this. In Countdown, if players in a series are tied for wins and points then the tiebreaker for seed position is who solved the most conundrums in their run.

This has happened before at a co:event, In Blackpool 2029, when myself and Jack Worsley were tied for 2nd place (so it decided who got to the final). I was advanced to the final on this rule even though I had lost to Jack earlier in the day.

I think if we're playing Countdown we have to go by Countdown's rules. If we're playing Goatdown I suppose we can make up something else that seems more sensible, such as h2h or points difference as suggested above.

[Gavin Chipper]

I don't think everything has to be done by Countdown's rules (though 10-letter conundrums are an outrage). But head-to-heads are problematic as explained above. I still don't know what is meant by spread - too much jargon. Points difference? That's not particularly satisfactory anyway as you can't always stop your opponent scoring. Edit - also for a given score, arguably lower points difference is better as it suggests better opponents.

I still think highest individual score (and 2nd highest in the case of a tie and so on) is the simplest and most obvious solution.

Stick with conundrums if you want to be consistent with Countdown, but what's the second tie-break? On the show I think it probably is scores anyway.

[JackHurst]

... there's a good chance it could happen one day if we keep having 10+ events a year for many more years.

This has happened before at a co:event, In Blackpool 2029...

See I told you it would happen eventually . Apparently it happens 5 years from now according to a time travelling Callum Todd.

[JackHurst]

I don't think everything has to be done by Countdown's rules (though 10-letter conundrums are an outrage). But head-to-heads are problematic as explained above. I still don't know what is meant by spread - too much jargon. Points difference? That's not particularly satisfactory anyway as you can't always stop your opponent scoring.

I still think highest individual score (and 2nd highest in the case of a tie and so on) is the simplest and most obvious solution.

Stick with conundrums if you want to be consistent with Countdown, but what's the second tie-break? On the show I think it probably is scores anyway.

Head to head is used in international football tournaments. In the extremely rare event that there are 3 or more in a tie there is then a subsequent set criteria, all the way down to drawing straws.

I'm sure you can argue for and against the relative merits of each of the tiebreak criteria until the cows come home, but what are you trying to achieve by doing so? Just makes sense to agree something vaguely sensible so that there is a precedent, and then move on.

Yes spread is pts difference btw.

[Ben Wilson]

The problem with using conundrums is then you'd need some way of tracking them*, and when you have scoresheets coming at you left, right and centre and at Edinburgh-style events you're also dishing out conundrums, you kinda want the process to be as streamlined as possible. This includes leafing back through dozens of scoresheets at the end of the day, too. Fwiw I just always leave tied in positions as they are, and if it affects prize positions, I award two prizes (and always buy spares for this very reason). Thankfully it's not happened within the top 5 at COLIN yet but clearly it's something I should think about.

*yes at COLIN I do sometimes track them for spot prizes but I don't track it for every player, just those who are still in contention in that particular round.

[Fiona T]

one point I'd like to raise is that if event organisers can beg, borrow or steal a monitor and cable for public display of draws and standings it would be very useful and save the constant "what table am I?" , "who is player 1?" etc...

[Gavin Chipper]

I don't think everything has to be done by Countdown's rules (though 10-letter conundrums are an outrage). But head-to-heads are problematic as explained above. I still don't know what is meant by spread - too much jargon. Points difference? That's not particularly satisfactory anyway as you can't always stop your opponent scoring.

I still think highest individual score (and 2nd highest in the case of a tie and so on) is the simplest and most obvious solution.

Stick with conundrums if you want to be consistent with Countdown, but what's the second tie-break? On the show I think it probably is scores anyway.

Head to head is used in international football tournaments. In the extremely rare event that there are 3 or more in a tie there is then a subsequent set criteria, all the way down to drawing straws.

Yes, but I don't think it's very good, and it would mean coming up with more complex criteria that you wouldn't otherwise need to if head-to-head throws up anomalies.

I'm sure you can argue for and against the relative merits of each of the tiebreak criteria until the cows come home, but what are you trying to achieve by doing so? Just makes sense to agree something vaguely sensible so that there is a precedent, and then move on.

Well, the implication here seems to be that whatever suggestion is made first, as long as it isn't completely insane, gets implemented. I don't think that's a good way to go about things.

By the way, you might not have seen because I edited my post, but I think for a given score, lower points difference is actually "better" because it implies better opposition. But doing it that way would be counterintuitive, so I think it's best left out as a tie-breaker altogether.

The problem with using conundrums is then you'd need some way of tracking them*, and when you have scoresheets coming at you left, right and centre and at Edinburgh-style events you're also dishing out conundrums, you kinda want the process to be as streamlined as possible. This includes leafing back through dozens of scoresheets at the end of the day, too. Fwiw I just always leave tied in positions as they are, and if it affects prize positions, I award two prizes (and always buy spares for this very reason). Thankfully it's not happened within the top 5 at COLIN yet but clearly it's something I should think about.

*yes at COLIN I do sometimes track them for spot prizes but I don't track it for every player, just those who are still in contention in that particular round.

I was looking at the wiki, and it seems that on Countdown they go by number of conundrums solved as a first tie-break, followed by best score. So if using conundrums is awkward, then top score is probably what you should use.

I think ties are generally fine anyway, but it can make a difference if e.g. there is a final and 2nd/3rd are tied so there needs to at least be a tie-break mechanism available.

[Fiona T]

I think extra hoops such as a single conundrum are best avoided unless a very last resort!

[JackHurst]

I'm sure you can argue for and against the relative merits of each of the tiebreak criteria until the cows come home, but what are you trying to achieve by doing so? Just makes sense to agree something vaguely sensible so that there is a precedent, and then move on.

Well, the implication here seems to be that whatever suggestion is made first, as long as it isn't completely insane, gets implemented. I don't think that's a good way to go about things.

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

[Callum Todd]

Back from my time-travelling escapades, I think if we're following Countdown there's not much to decide on really. Just make sure we're aware of what the tie breaks are after conundrums.

If we're not then sure let's come up with an alternative and trial a few out. It might take a while to work out which is best given it's a fairly uncommon situation though.

[Gavin Chipper]

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

[Adam S Latchford]

Can we make a hard rule that all bristol events do the split in conundrums in the room (half the room get one, half the room get the other) so every single table has somebody adjudicating a crucial?

Like - we did it last time - but i really really really think this should be a rule that is enforced

[Gavin Chipper]

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

And the head-to-head rule is still under-defined. What do you do if three or more are tied, none, some or all of whom have played each other? That's why you have to give new rules some thought. You can end up not noticing that it still leaves ambiguities, and you can fail to realise that simpler rules were already in place and that the whole thing was unnecessary in the first place.

[Fiona T]

Can we make a hard rule that all bristol events do the split in conundrums in the room (half the room get one, half the room get the other) so every single table has somebody adjudicating a crucial?

Like - we did it last time - but i really really really think this should be a rule that is enforced

Do you think that's needed if you use a phone based buzzer? No controversy about who buzzed first.

[Matt Rutherford]

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

No, it isn't. The show and the tourneys are similar, but distinct in many ways. From the view of an organizer, it makes things a lot simpler in these situations. Conundrums solved creates faff that can be avoided. Agree with Jack's observations

Can we make a hard rule that all bristol events do the split in conundrums in the room (half the room get one, half the room get the other) so every single table has somebody adjudicating a crucial?

Like - we did it last time - but i really really really think this should be a rule that is enforced

Should have encouraged use of people filming/Fiona's buzzer at Brum for making it clearer who has buzzed first. Would encourage all people to do so moving forward. Phone buzzer would means issues could be avoided, but still think having two cons is good for Bristol events so video can be used as a back up if needed.

[Matt Rutherford]

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

And the head-to-head rule is still under-defined. What do you do if three or more are tied, none, some or all of whom have played each other? That's why you have to give new rules some thought. You can end up not noticing that it still leaves ambiguities, and you can fail to realise that simpler rules were already in place and that the whole thing was unnecessary in the first place.

Go for points difference straight out in that case. Still less faff

[Adam S Latchford]

Can we make a hard rule that all bristol events do the split in conundrums in the room (half the room get one, half the room get the other) so every single table has somebody adjudicating a crucial?

Like - we did it last time - but i really really really think this should be a rule that is enforced

Do you think that's needed if you use a phone based buzzer? No controversy about who buzzed first.

The phone buzzer does eliminate it but also relies on that being enough readily available on each of the tables. Table of 3 theres always a phone - guarantee there'll be some tables of 2 where someone doesn't want to use their phone and the others is dead or something - and splitting the cons at a bristol maybe adds 30 seconds work at maximum to the organiser

[Fiona T]

It's a great way to go about things. If your approach to decision making is to not sweat the small stuff, you get more done, and have more time and energy left to focus on the decisions that actually matter.

I think every workplace has at least one Gevin who just refuses to let trivial decisions happen quickly, and that person tends to be the enemy of progress

I almost feel like I am letting myself down just by continuing to engage at all!

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

No, it isn't. The show and the tourneys are similar, but distinct in many ways. From the view of an organizer, it makes things a lot simpler in these situations. Conundrums solved creates faff that can be avoided. Agree with Jack's observations

Can we make a hard rule that all bristol events do the split in conundrums in the room (half the room get one, half the room get the other) so every single table has somebody adjudicating a crucial?

Like - we did it last time - but i really really really think this should be a rule that is enforced

Should have encouraged use of people filming/Fiona's buzzer at Brum for making it clearer who has buzzed first. Would encourage all people to do so moving forward. Phone buzzer would means issues could be avoided, but still think having two cons is good for Bristol events so video can be used as a back up if needed.

Filming at a Lincoln style event is tricky - the host is already using their phone to operate timer (fair enough they can use a player's phone) but as well as that they're watching the timer to give 10 or 5 sec warning, watching the players to see who buzzed, and checking the buzzee answers promptly, clearly and correctly, not to mention goaty complications! I think attempting to film at the same time is likely to cause more problems than it solves. Conundrums always the trickiest part of these events!

[Gavin Chipper]

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

No, it isn't. The show and the tourneys are similar, but distinct in many ways. From the view of an organizer, it makes things a lot simpler in these situations. Conundrums solved creates faff that can be avoided. Agree with Jack's observations

And the head-to-head rule is still under-defined. What do you do if three or more are tied, none, some or all of whom have played each other? That's why you have to give new rules some thought. You can end up not noticing that it still leaves ambiguities, and you can fail to realise that simpler rules were already in place and that the whole thing was unnecessary in the first place.

Go for points difference straight out in that case. Still less faff

Or just the second tie-break from the show - highest individual score (and 2nd, 3rd highest etc. in case of further ties). I'd say that's the simplest and least faff. As said, points difference if anything negatively correlates with what you want.

[JackHurst]

Filming Conundrums is the way to go imo. You also then have a record of what actually happened which can't be argued with. Buzzer app is good for who buzzes first, but doesn't account for other things like player hesitation, or declarations made in a dubious way.

Filming isn't hard. When I play a game I usually record my conundrum by putting my phone on selfie video mode and propping up against a solid object on the table. You can set it recording at any point before the host reveals, it doesn't have to be right before. Pro tip: make sure the shuffle isn't visible on the phone screen so that players could see it there before it's revealed by the host.

Yes Bristol style events should absolutely split. How much time does an extra Con add to proceedings? I'd go as far as saying bristol cons should have 3 difficulty groups. Pitching conundrum difficulty to a whole room as really hard, so you can very easily end up with conundrums that are way too easy to be separating good conundrum players.

[Graeme Cole]

Well as Callum says, there's already a tie-break mechanism used on the show so it would be simpler to do that. Plus the suggestion is worse.

No, it isn't. The show and the tourneys are similar, but distinct in many ways. From the view of an organizer, it makes things a lot simpler in these situations. Conundrums solved creates faff that can be avoided. Agree with Jack's observations

And the head-to-head rule is still under-defined. What do you do if three or more are tied, none, some or all of whom have played each other? That's why you have to give new rules some thought. You can end up not noticing that it still leaves ambiguities, and you can fail to realise that simpler rules were already in place and that the whole thing was unnecessary in the first place.

Go for points difference straight out in that case. Still less faff

Or just the second tie-break from the show - highest individual score (and 2nd, 3rd highest etc. in case of further ties). I'd say that's the simplest and least faff. As said, points difference if anything negatively correlates with what you want.

Agreed for all these reasons. Points difference isn't a suitable tiebreak for Countdown, IMO.

As I understand it, on Countdown players are ranked by wins, then total points scored, then (at least according to very old anecdotal evidence) number of conundrums solved, then highest individual score in a game.

I don't see why co-events couldn't use the same. Rank the standings by wins then points, and if two players are still tied and it's for a position that matters (i.e. decides who makes the final), use conundrum count then highest individual score. In the highly unlikely event that that doesn't break the tie, we can cross that bridge when we come to it (sudden death conundrums, rock paper scissors, etc).

If it's a pain to go back through the scoresheets and count conundrums, then forget that, and use highest individual score after total points. But points difference doesn't make sense for Countdown. It's not football. When your opponent scores, it's because they played well, not because you didn't defend well.

[James Laverty]

Points difference would be a terrible tiebreaker for Countdown, as Graeme says, it's not like football or rugby, you aren't actively trying to stop your opponent scoring points (except maybe stymieing them with a 6s or 4l numbers choice tactically).

If there is somehow still a tie, I can think of a few options, but all have their problems

1. Total time it took to solve conundrums (not solved =30 secs)
2. Percentage of Max Score

I'm grouping these two together as why they could work, but as Graeme alludes above, it would be a hell of a pain to go through the game sheets, and I know a lot of hosts don't always bother with recording the max.

3. Head to Head score

Very unlikely to have happened, but if the two players played against each other at some point of the day, the victor prevails. Probably more likely to happen in a Bristol style tournament.

4. Face off/Mini game to decide

Would need to accomodate time in the schedule and events often struggle to run to time in the first place

[Callum Todd]

Good debate folks. I agree points difference isn't a great tiebreaker so should be as low down the list as possible. And h2h just causes a clusterfuck if it's more than a two-way tie.

Still think we should just do what Countdown does unless there's a very good reason to do otherwise, and in this case I don't believe there is a good enough reason. It's really not hard to work out how many conundrums each player has solved.

I'll be sure to pass all this feedback on in 2029 to see if that changes the decision.

[Fiona T]

I've put together a conundrum checker that event hosts can use to check the validity of their conundrums.

https://focaltools.azurewebsites.net/Co ... ecker.aspx

It checks the scramble matches the solution (Graeme's conundrum printing thingy already does that) but also checks that there is one and only one 9 letter word that matches the scramble (the same as I've built into the latest version of RoboRiley). It does not check for plurals and that kind of thing.

It's slow (took 5-10 seconds for a file with 15 conundrums), so if you have a large file it's gonna take a while... I've added some flashing text to annoy you while you wait. I might look at how to speed it up but I guess it's only going to be used a few times a year so not a biggie.

(I've also moved the mobile buzzer to this hosting platform https://focaltools.azurewebsites.net/buzzer/buzzer.html)

[Graeme Cole]

I've put together a conundrum checker that event hosts can use to check the validity of their conundrums.

https://focaltools.azurewebsites.net/Co ... ecker.aspx

It checks the scramble matches the solution (Graeme's conundrum printing thingy already does that) but also checks that there is one and only one 9 letter word that matches the scramble (the same as I've built into the latest version of RoboRiley). It does not check for plurals and that kind of thing.

You might want to exclude the scramble itself from the possible solutions. It's valid for the scramble to be a valid word if there's only one anagram of it, but if you feed the conundrum checker "RAUNCHIER HURRICANE", for example, it complains that there are two solutions.

It's slow (took 5-10 seconds for a file with 15 conundrums)

That sounds optimisable. What does it do to check a conundrum?

[Fiona T]

You might want to exclude the scramble itself from the possible solutions. It's valid for the scramble to be a valid word if there's only one anagram of it, but if you feed the conundrum checker "RAUNCHIER HURRICANE", for example, it complains that there are two solutions.

yeah I considered that but figured you could think "yes I know" But yeah easy enough to exclude it. (Edit - done)

It's slow (took 5-10 seconds for a file with 15 conundrums)

That sounds optimisable. What does it do to check a conundrum?

It's probably a bit quicker than that but still slower than I'd like.

It's basically doing the same as I've done with RoboRiley - I've got a trie structure with the dictionary in, I then get all the permutations (362880 for 9 letters without repeats) stick those in a trie and compare tries. Tries are pretty efficient - I think it's the permutations thing that needs optimising - will have a google...

[Graeme Cole]

You might want to exclude the scramble itself from the possible solutions. It's valid for the scramble to be a valid word if there's only one anagram of it, but if you feed the conundrum checker "RAUNCHIER HURRICANE", for example, it complains that there are two solutions.

yeah I considered that but figured you could think "yes I know" But yeah easy enough to exclude it. (Edit - done)

It's slow (took 5-10 seconds for a file with 15 conundrums)

That sounds optimisable. What does it do to check a conundrum?

It's probably a bit quicker than that but still slower than I'd like.

It's basically doing the same as I've done with RoboRiley - I've got a trie structure with the dictionary in, I then get all the permutations (362880 for 9 letters without repeats) stick those in a trie and compare tries. Tries are pretty efficient - I think it's the permutations thing that needs optimising - will have a google...

Yep, generating all the permutations of a 9 letter word is the red flag here.

If you've already got a list of all the valid words, you could arrange the letters within each valid word in alphabetical order, then build a sorted list of the resulting strings (or put them in a trie, if you prefer). For example, ACDEINOTU would appear in the list three times, consecutively. Now you only have to check one permutation of the scramble, look it up in your pre-computed sorted list, and count how many times it appears.

There are ways of improving on this (e.g. include the count in the list like ["ACDEINOTU", 3] rather than repeating the strings), but the main thing is you don't have to generate every permutation of the scramble.

[Fiona T]

Yeah makes sense for this specific scenario, but not sure how/whether that approach works with the more common scenario of finding all maxes/words from a selection (but typically just one selection at a time so slowness much less of an issue!) I've had a bit of a play and improved the permutation generation by about 50%. Not sure whether it's worth building something specifically for this which will probably be used a few times a year max - will have a think... (Mebbe we need a separate algorithms thread!)

[Gavin Chipper]

As I understand it, on Countdown players are ranked by wins, then total points scored, then (at least according to very old anecdotal evidence) number of conundrums solved, then highest individual score in a game.

I don't see why co-events couldn't use the same. Rank the standings by wins then points, and if two players are still tied and it's for a position that matters (i.e. decides who makes the final), use conundrum count then highest individual score. In the highly unlikely event that that doesn't break the tie, we can cross that bridge when we come to it (sudden death conundrums, rock paper scissors, etc).

You would go by second highest score if highest is tied, and so on. As I'm 99% sure they would do on the show.

[Graeme Cole]

Yeah makes sense for this specific scenario, but not sure how/whether that approach works with the more common scenario of finding all maxes/words from a selection (but typically just one selection at a time so slowness much less of an issue!) I've had a bit of a play and improved the permutation generation by about 50%. Not sure whether it's worth building something specifically for this which will probably be used a few times a year max - will have a think... (Mebbe we need a separate algorithms thread!)

What if you just iterate over the whole dictionary, checking every valid word to see if it's an anagram of the scramble? (Or, for the general case, that the word is available from the scramble?) I think even that would still be faster than checking every permutation of the scramble to see if it's a valid word. Checking if one string is an anagram of another is basically the same algorithmic complexity as finding a string in a trie, but there are four times as many permutations of 9 letters as there are 1-9 letter words in the dictionary.

[Fiona T]

Yeah makes sense for this specific scenario, but not sure how/whether that approach works with the more common scenario of finding all maxes/words from a selection (but typically just one selection at a time so slowness much less of an issue!) I've had a bit of a play and improved the permutation generation by about 50%. Not sure whether it's worth building something specifically for this which will probably be used a few times a year max - will have a think... (Mebbe we need a separate algorithms thread!)

What if you just iterate over the whole dictionary, checking every valid word to see if it's an anagram of the scramble? (Or, for the general case, that the word is available from the scramble?) I think even that would still be faster than checking every permutation of the scramble to see if it's a valid word. Checking if one string is an anagram of another is basically the same algorithmic complexity as finding a string in a trie, but there are four times as many permutations of 9 letters as there are 1-9 letter words in the dictionary.

Yeah it makes sense I agree - atm I don't have the dictionary uploaded to the website, just the serialised trie so will require a little effort - will get to it when I have some time.

[Fiona T]

Yeah makes sense for this specific scenario, but not sure how/whether that approach works with the more common scenario of finding all maxes/words from a selection (but typically just one selection at a time so slowness much less of an issue!) I've had a bit of a play and improved the permutation generation by about 50%. Not sure whether it's worth building something specifically for this which will probably be used a few times a year max - will have a think... (Mebbe we need a separate algorithms thread!)

What if you just iterate over the whole dictionary, checking every valid word to see if it's an anagram of the scramble? (Or, for the general case, that the word is available from the scramble?) I think even that would still be faster than checking every permutation of the scramble to see if it's a valid word. Checking if one string is an anagram of another is basically the same algorithmic complexity as finding a string in a trie, but there are four times as many permutations of 9 letters as there are 1-9 letter words in the dictionary.

Yeah it makes sense I agree - atm I don't have the dictionary uploaded to the website, just the serialised trie so will require a little effort - will get to it when I have some time.

Sorted for conundrums - simpler than I thought - works instantly now, cheers Graeme! Will think about the 'get words' ...

[Fiona T]

What if you just iterate over the whole dictionary, checking every valid word to see if it's an anagram of the scramble? (Or, for the general case, that the word is available from the scramble?) I think even that would still be faster than checking every permutation of the scramble to see if it's a valid word. Checking if one string is an anagram of another is basically the same algorithmic complexity as finding a string in a trie, but there are four times as many permutations of 9 letters as there are 1-9 letter words in the dictionary.

Yeah but I'm not checking 9! words - by putting the permutations into a trie, if the scramble is ABCXEANRI once CX is ruled out, then all the words that start with CX are ruled out etc... Anyway - probably sidetracking this thread from its original purpose - will have a play and do some comparisons.

[Fiona T]

I thought the pointy cards for word length declarations worked brilliantly well at co:lon and should definitely be used going forward! (For those who weren't there, Phil had created an A4 sheet with left and right columns of Nothing - 9 and table number in the middle - players pointed to their word length rather than saying out loud - was incredibly smooth and everyone seemed to manage it without problems!)

[Gavin Chipper]

The live results thing was great. But I can't see any reference on it to the final or the overall winner of the event. It seems that even after all these years that the winner of the event being an afterthought in the online results is still very much a thing.

[Gavin Chipper]

Also, in Bristol style conundrums, if someone buzzes, is the other person still supposed to buzz if they then see the answer? Or do they just write it down? Certainly on Fiona's buzzer thing it wouldn't register anyway.

And there was at least one event in the past where the person who buzzed first was supposed to allow their opponent to see their answer so that it's more like the show. I think the fact that this happened puts some doubt into people's minds about the standard method for this. Although if they buzz late there won't be enough time for their opponent to see their answer before committing to an answer themselves.

[Adam S Latchford]

I thought the pointy cards for word length declarations worked brilliantly well at co:lon and should definitely be used going forward! (For those who weren't there, Phil had created an A4 sheet with left and right columns of Nothing - 9 and table number in the middle - players pointed to their word length rather than saying out loud - was incredibly smooth and everyone seemed to manage it without problems!)

These were great

I think just adding simply "if its not written down, point repeatedly at your declaration" would be a good way to work around if both players are not written down (they can then proceed to quickly scramble it down)

[Gavin Chipper]

I don't think just saying "Not written down" is a problem though. It's when people round the room can hear all the lengths and the actual words that Bristol style can be a problem.

[Fiona T]

Also, in Bristol style conundrums, if someone buzzes, is the other person still supposed to buzz if they then see the answer? Or do they just write it down? Certainly on Fiona's buzzer thing it wouldn't register anyway.

And there was at least one event in the past where the person who buzzed first was supposed to allow their opponent to see their answer so that it's more like the show. I think the fact that this happened puts some doubt into people's minds about the standard method for this. Although if they buzz late there won't be enough time for their opponent to see their answer before committing to an answer themselves.

I mean I tapped the phone to indicate I was having a guess but if someone is scribbling it down that would indicate they're also 'buzzing'. In terms of seeing what the other person has written, that's fine IMO but maybe does need explaining in the preamble