All games are tough, but some.....

[mkenuk]

Is it just me, or have some of the recent Standard games been more difficult than usual?

A while ago, just for my own amusement, I tried to devise a formula that would grade Chi games by level of difficulty.
I got the idea from a book of Sudoku games in which the puzzle grids were shown to be 'easy', 'difficult', 'very difficult' etc.

Simply counting the number of rosettes awarded for each game doesn't work, as the number of players varies so much from game to game; also it's impossible to know how many of the rosettes 'won' in any game have come through the use of private word-lists and online anagram solvers.

My system for grading Chi games is based on the number of common words missed by players.
For example, in the recent locksmith game, which I felt instinctively was much harder than usual,  no fewer than nine 'common' words (from a 'target' of thirty-three) were missed by more than half of those taking part; six of those nine words were missed by two thirds of players and four of the words (sitcom, shtick, coltish and schmo)  were seen by less than a quarter of those playing the game.
Ninety percent of players missed schmo.

In the recent postponed game, by contrast, only four words (from forty-five) were missed by more than half. None were missed by as many as two-thirds.
Using the scoring system I've devised, postponed would have a difficulty rating of one (easy) and locksmith would have a rating of five (the hardest).

Today's game still has twelve hours to go, but my betting is that it will be another 'five'! Possibly shown to be even tougher than locksmith.

[Hobbit]

I quite agree Mike.  I'm struggling with today's game but I put that down to being quite tired! I only hit on the 9 letter word by accident! In the locksmith game I did manage coltish but didn't get sitcom, shtick or schmo (which I don't think I've ever heard of!)  Having said that I'm generally only a "middle of the road" sort of player!

[Valerie]

I think you may be onto something, Mike.  I've thought the same about the standard games of late but decided it may just be me.  I found that LOCKSMITH game almost impossible.  Kept going back to it, willing the words to jump out at me before finally giving up.   But I find a few games to be like that.  

[ensiform]

I got schmo, but I never get a rosette or a trophy, so my triumph was probably based solely on the fact that I'm a schmo and so knew the word.

[Jacki]

A rating is a great idea! I use other players rosettes as a guide. If T Rex doesn't have a rosette I know I'm most probably cactus!

[Valerie]

Snap, Jacki!  I also think the same about Mike.  Funny how you develop player yardsticks.   

[mkenuk]

A rating is a great idea!

Unfortunately the scoring system I devised for Chi does not predict how difficult a game will be before you play it - it only seeks to objectify how difficult a game was after it has closed.

Puzzles such as Sudoku and Chess problems ('white to play and mate in two moves') are exercises in logic. There is usually only one solution to such problems, whereas with Chi games there are a huge number of variables which make predicting in advance the level of difficulty of any game very difficult.

Yardsticks / benchmarks?
If you can keep pace with the Doctor from Forbidden Planet you won't be far from the target!!

[TRex]

My measure for difficulty is the bar graph beneath the Scoreboard with the 'Rare' button set to not show rare word scores. If the left-to-right trend (not counting the leftmost bar) is upwards with the peak at the rightmost bar, then I figure the puzzle isn't unusually difficult. If the highest bar is somewhere towards the centre, it is more difficult. The further to the left the peak, the harder the puzzle.

[TRex]

A rating is a great idea! I use other players rosettes as a guide. If T Rex doesn't have a rosette I know I'm most probably cactus!

I frequently don't get a rosette until comparing results with my wife!

[Ozzyjack]

We don’t give each other the answers, but if we each get within a Bull’s roar of a rosette, we do word counts to identify where we have differences.  It becomes a lot easier if you know that a word exists in a general area and it saves a lot of time looking at other areas fruitlessly.

Of course, it doesn’t help if we both miss the same word. 

[Valerie]

Ditto TRex and Ozzyjack.  My partner and I are very competitive.  I do printed versions for each of us to play and compare scores.  Then I enter our combined results online to see how we fared.  Sounds drawn out and complicated but it works for us.  And we also learn new words from each other.  As demonstrated in the second non-common word of yesterday's DESECRATE challenge game.  I would never have come up with that word in a million years if it hadn't been for his maths/scientific brain!

Thanks Mike for your statistical input to the games.  Very enlightening.  I wouldn't know where to start.

[mkenuk]

My measure for difficulty is the bar graph beneath the Scoreboard with the 'Rare' button set to not show rare word scores.

Yes, that graph contains a mass of statistical information and, provided you join the game after it has been open for a while, it can give a good indication of how easy or difficult the game is.

[Jacki]

My husband and I don't play independently however sometimes when I can't get the main word I ask him to have a look. Mostly of course it's pretty easy to spot but some like PRETTIFIES and SUBLIMATE were not jumping out at me however my better half is more analytical than me and he can see the wood for the trees!