Bursting fourth with pride

[rhu]

Today was the second annual Boston Crossword Puzzle Tournament, once again wonderfully arranged by Eric Helmuth and Joon Pahk. You may recall that last year I came in eighth, with two puzzles that I "should" have done one clock-minute better on. (For example, in one case, I had 30 seconds left in the minute to check my results, and accidentally took 32.)

My goals for this year were (1) to have fun, (2) to manage my time better, and (3) to find out whether or not last year was a fluke.

I'm pleased that I succeeded at all three goals.

(1) I had a great time, both on and off the "playing field." I sat in front of Ryan and Brian, and they teased me mercilessly about not listening to their podcast, and we chatted about a whole bunch of things. I finally got to meet BEQ. And I got to spend a lot of time chatting with fauxklore, who was in town for the event, and with a bunch of my local NPL friends. I also saw my Zamir friend Deb S. and Arnold Z-K from my shul.

(2) On no puzzle did I feel that I should have scored in the previous minute; I think (based on the rustling of the room) that I fell one minute behind on puzzle 2, but that was because it legitimately took me that long to finish it. And on puzzle 4, by BEQ and Joon, I think I made up that minute. We'll know for sure later tonight when the detailed results are posted.

(3) And I came in fourth. So now, to some degree, the pressure is off. I don't expect to ever actually crack the top three, but I feel now that last year's top-ten finish wasn't just blind luck.

Comments

[cellio]

Congratulations!

Is there partial credit, such that you have to decide on the incremental benefit of going an extra minute, or do you have to complete a puzzle to score with it?

[530nm330hz.livejournal.com]

The actual scoring rules are somewhat complex, but what they boil down to for the people near the top is this:

* You give up 25 points every 60 seconds.
* Your first mistake costs you over 150 points, which basically knocks you out of contention.

So it's more important to solve clean than fast, and if you've got time left in the minute it is generally wise to use it to scan for mistakes.

[abbasegal.livejournal.com]

Interesting. You have clearly a clearly delineated "no incremental cost to check" point -- that is, until the next minute mark, any checking is free. How do you think the strategy would change if, instead, you were dinged 1 point every 2.4 seconds (or if you prefer, .4167 points per second, or some arbitrarily continuous equivalent)?

[530nm330hz.livejournal.com]

Uhhhh... that's a game theory question... that means math.... mmmm... we do word puzzles here.... :-)

Seriously, though, I think that would lead to constant weighing of "how far behind am I vs how confident am I that I didn't make any careless mistakes?" Because it's the careless mistakes you're looking for --- when you enter an answer, you KNOW if you're unsure, and you make a Ripstein mark, and you double-check it at the end regardless.

On puzzle 4, I knew I needed to make up time, and I'd already decided going in that if I finished with less than 15 seconds in the minute I wasn't going to do more than a cursory check for blank squares. As it was, I had 38 seconds left in the minute and did take 20 of them to scan the answers for nonsense.

Next year at the ACPT they're going to try moving to 30-second quanta. Which won't affect the strategy too much, although I could see people choosing to take a 30-second penalty where they might not be willing to take a 60-second penalty for checking.

[fauxklore]

Congratulations again. It was nice seeing you.