So there goes another series of ‘Only Connect’. Series 15.
I hope people have enjoyed it. Hearty congratulations to the 007s, who, I think you’ll agree, were extremely worthy of their crown. They went unbeaten throughout the series and showed an incredible breadth of knowledge and calm under pressure. One of the enjoyable things about this series was the number of teams we felt, from the start, were in with a shout – there were several strong contenders, and the line-up of the later rounds was never a foregone conclusion.
This has been the third series on which Mr Waley-Cohen (if he exists) and I have been the question editors and it took us past the 100-episode mark in charge (having been writers on the show for four series before that, and, separately, contestants earlier on). It was a joy to make at every stage, from the writing to the editing to the filming.
I wrote a blog about being question editor in the middle of our first series at the helm, and thought it might be a nice time to revisit. My intention is to focus less on the actual process of question editing and more on how the show is received. Inevitably, though, there’ll be a bit of both.
I tend to follow each episode when it’s being transmitted via the #onlyconnect hashtag on twitter, either live or retrospectively, and also check in on any comments on Facebook’s Quiz Discussion Group.
Though there’s a danger of being a glutton for punishment, and it’s also important to avoid getting into any kind of disagreement, tempting as it sometimes is, I’ve found it, on balance, a pretty reasonable way to judge how a show is going down, what questions worked well, what less well.
If people refer to the questions and the question setters at all, it is usually in the context of us being evil geniuses, setting out to bamboozle the teams and make the viewers at home feel stupid with the weight of our intricacies and obscurities. Of course, the show plays up to that a bit and it is all part of the fun and, in most cases, affectionately meant.
But, as it happens, my perspective as both a question setter and a question editor is different, in fact it’s almost the exact opposite.
First of all, I’m not a genius, nor am I unambiguously evil. Other question setters may well be, I can only speak for myself. I’m not a puzzler, a crossworder, a world-class quizzer. I’m a question writer by profession and have been for a long time.
Along with the writing, I ran a lot of quizzes for a long time, and one of my main aims when I ran quizzes was to leave no participant behind, to make sure the team finishing last had (almost) as much fun as the team finishing first, and to try to pitch the quiz so that the top score was not too high and the bottom score was not too low.
Mr Waley-Cohen and I bring the same approach to Only Connect. An occasional complaint I see is along the lines of “what’s the point in asking questions which no one can get?”. And I couldn’t agree more.
Which is why, in Only Connect, we try to have questions that teams do get (yes, yes, I know, this week’s final may not be the best example of that, but, come on, it was the final!)
A quick look at the average scores over the 15 series bears this out. We hope that the average for each episode, over the course of a series, will be steady, at around 40 points, and that’s pretty much exactly where it is (currently 40.1) – some series, it’s more like 38-39, some 41-42, but generally 40 is a good mark. That’s 20 per team, which, when you think about it, is plenty – a point for each of your R1 and R2 questions, full marks on the wall, 4 points on the vowels – it might break down like that.
Teams are generally getting questions right, and we want them to. If two questions in a row go past without either time scoring a point, we feel a bit sad.
So, why does it sometimes look/feel like the quiz is really hard? I simply think that’s built into the format – at home and in the studio, you do spend a lot of time being at a loss, before suddenly the answer is worked out.
I suppose that the complaint that we’re evil geniuses trying to bamboozle the teams is a bit like complaining that detective fiction writers spend too long concealing the identity of the killer, and why can’t they just cut to the chase and make it explicit whodunnit a bit earlier (not that Columbo isn’t one of the greatest shows of all time, but you get my point …)
Also, for those watching, the fact is, you’re at home, perhaps on your own, perhaps you’ve had a glass of wine, while the teams are with two other people, both almost certainly sober, and the value of their each throwing in ideas is incalculable. If you’re a step behind them, no wonder.
They’re also all very good at it, they’re not there by chance, they’ve passed auditions and (usually) been practising for ages.
And, most importantly and simply, we aim the questions at the teams, not the audience. It is a miraculous joy that so many viewers join the teams on the journey to each answer, but for us, as writers and editors, if we started worrying unduly about how people at home were doing, the show would fall apart.
It is a fine balance – in my head, when writing my own and when looking at other people’s questions, I’m always thinking that there shouldn’t be (with very very few exceptions) enough in Clue 1 to confidently guess (though it is lovely when teams can look back and work out in hindsight how Clue 1 fits in), and there never should be too little by Clue 4 (or Clue 3 in Round 2).
So we have to plot the journey very carefully in each question. It’s not always exactly the same, it couldn’t be – some questions get easier more quickly and some stay pretty hard all the way to the end. There are different types of question, and some are more “high-risk” than others.
There are some questions which we’re entirely confident fall into the sphere of General Knowledge, and the answer will pretty much have presented itself to any alert team by the end. Some, particularly sequences, are more specific – good examples in Series 15 were, if you recall, the ‘Panic’ maps sequence, and the ‘Inside Number 9’ series opening episode locations. The risk is that these questions fall completely flat, that no one on either team gets the cultural reference, and you end up looking like you’re dealing in wanton obscurity.
So as writer/editor, you are staking something on the kind of things people who go on Only Connect might know, and hoping for the best. These kind of questions are important to the show, whether they’re about cult TV shows or the small details of everyday life.
If I may wax a little philosophical … I feel like everyone closely involved with the show sees it as a little more than just a quiz show – we really hope it lives up to its title. It has to be a show about absolutely everything, where everything feels connected, and that is a big part of mine and Jack’s job (a part that he is particularly good at), making sure that as many topics as possible are covered.
Think of it this way; there are quizzing sweet spots – Beatles, Popes, Scrabble, Snooker, 80s films etc – they lend themselves well to Only Connect and to quizzing in general.
But it’s our duty as question editors to veer away from the sweet spots, to take risks, to have questions about everything and everything else (sometimes all within one question, sometimes not). The unique structure allows us to cover everything from spark plugs to Sparks, Ludacris to Lucretius, Iris Murdoch to Unsolved murders, Cat’s Eyes to catfish, Dora Maar to Moons of Mars (you can tell I’m quite enjoying this, but I’ll stop now …) and, most importantly, we can be obscure without being off-putting, because the obscurity can go early in a question, and the clarity can come later.
That’s why I take mild exception to the suggestion that Only Connect is forbidding or highbrow. In some ways, it’s anything but highbrow. Yes, there is the odd question in a hundred that asks you to know about Leonhard Euler or Latin conjugations, but there are plenty more on the likes of Back to the Future or Beyoncé. Nothing is too popular or populist. Nothing is off limits.
Our duty is to make it varied, fun, accurate and most of all, fair. I think we manage this pretty well – for the first two rounds, of course, some pot luck comes into it – sometimes teams just don’t get the run of the questions, but usually it evens itself out over the two rounds.
With Round 4 (Missing Vowels), it’s on the buzzer, so the only way that can look unfair is that there is a very specific topic on something one player on one team happens to be an expert in. That’s rare, but can happen.
The greatest focus on fairness is, of course, with Round 3, the walls. Two unmatched walls can make a fair difference. Again, I think we do a pretty good job, we look at the subject matters and the mechanisms, and the numbers of red herrings, and match them closely. Viewers’ perceptions are heavily influenced by how the teams do – I know there was one pair of walls this series, in particular, where a few people online were suggesting one was harder than the other, but actually, they had the same mechanisms, balance of wordplay and knowledge, it just happened that one team identified a (for some) tricky knowledge group early and that unlocked their wall, whereas that didn’t quite happen for the other team.
Funnily enough, when I myself was on Only Connect, many years ago, we blew a commanding position after R2 by floundering on a tricky wall. The other team scored 10 on their wall, and I was quite convinced we’d had a rough bit of luck. It was only looking back at the episode several years later, I examined the opposition’s wall and thought “you know, that was pretty tough too, I’m not sure we’d have done any better with that”.
Anyway, I hope that’s provided some insight. It’s such a joy to work on the show and it’s amazing that so many people have found their way to enjoying it. I hope it comes across what a kind and good-natured atmosphere there is on set, that the contestants are mainly having a great time, and I hope you’ll carry on watching.
Meanwhile, we’re waiting for Series 16 to be allowed out of isolation. Hope to see you all on the other side later in the year.