Difficulty of Quick Crosswords by Setter

Author   starstruck_auPosted on Categories Announcement

For devotees of the QUITCH, there is now an analysis of the difficulty by setter. You can find it by starting here. There is also a link available at the top of the Trends page and under the recent crosswords on the main crosswords page. Please let me know of any issues.

There should be no change to the main SNITCH page, as setters are not currently listed for the 15×15 puzzles.

6 comments on “Difficulty of Quick Crosswords by Setter”

  1. Facinating – Very interesting to take a look. We have had a lot of commentary on why all of us to a certain extent find some crosswords easier than the blogger / quitch suggests and some harder.
    Had I actually timed myself and recorded the times I would have been interested to compare mine to the herd, and then I would have found out if it was individual fluctuations or a setter influence.
    To get a bit techhie, as the timing of anything is asymmetric data, percentiles are the correct way to show it, you have 25th and 75th percentiles, but then average in the middle. Any chance of adding the 50th percentile (the median), don’t drop the average, it is interesting how they differ.
    Also lots of people have commented that it does not include the DNFs which affects the hardest setters most, so presumably the impact of the setters is even more pronounced than the figures show. (And the paper solvers like me!)

    1. Thanks for the comments and for picking up the statistical inconsistency. (Mrs S, who qualified – and for some time practised – as an actuary, laments my happy-go-lucky engineering approach to statistics.) Now fixed.

      On the impact of DNFs, I can think of no practical way to incorporate them consistently. The Club approach, of deducting 20 points from the minimum error-free solve score, seems very arbitrary. I also think that a high level of DNFs represent a challenge for those who did make it through unscathed, as they may have taken additional time to avoid the bear traps. So, the SNITCH may reflect a higher level of difficulty as a result. Whether this really compensates for those who DNF is anyone’s guess.

      Thanks again for the feedback.

  2. Couple of extra entries at the bottom of the list for mispelled/mispelt/misspelt/misspelled/incorrectly typed Teazel and Trelawney

  3. Wow: I will be fascinated to get my head round this over the next few days and see if my general experience is reflected in the statistics. Thank you so much for this intriguing analysis.

  4. Brilliant. Just when I think the SNITCH family can’t get any better, it does!

Comments are closed.