SwanShadow Thinks Out Loud

LearnedLeague Season 58 concluded this week, and in the immortal words of Gloria Gaynor, “I will survive.”

After a grueling 25-day campaign, I managed to finish 17th in Rundle A West — without dispute, the league’s most talent-loaded bracket. By placing above the bottom 10 in our 32-player Rundle, I avoided relegation (the LL euphemism for “demotion”) to a lower bracket for next season. Not escaping that fate were several esteemed competitors whom I consider superstars in the trivia world.

Sometimes, it is indeed better to be lucky than good.

It’s worth noting that my placement in A West plummeted nine slots between last season (when I finished 8th) and this, even though my statistical performance in both seasons was similar. (My head-to-head matchplay record in LL57 was 11-9-5; this season, LL58, I went 10-9-6.) The primary contributing factor here was the disbanding after LL57 of the League’s previous top level, Rundle Championship, and the redistribution of its participants into the four A-level Rundles. A West inherited several former R-Champ members, raising the difficulty factor of our bracket exponentially. (Not that it needed to get more difficult. Rundle A West has long borne the nickname “A Murder” with good reason.) I would have to check name by name to be certain, but I’m pretty sure that every A West member who was in R-Champ in LL57 finished above me in LL58. So, there you go.

Now that I’ve completed three full LL seasons, the last two in A West, it’s a good time to analyze my overall performance in the League to date.

My win-loss-tie record stands at 42-21-12. That’s significantly skewed by my rookie season, in which I went 21-3-1 against other rookies and won my Rundle. None of my fellow R Central competitors had yet advanced to A-level as of LL58 (I believe one or two just earned promotion to A for next season), so it’s fair to say that I compiled that gaudy rookie record against less-stiff competition than what I’ve faced in A West the past two seasons. So, let’s call that first season’s 21 wins an outlier. In A-level competition, I’m a just–over-.500 hitter.

It’s also important to understand wins, losses, and ties in the context of LearnedLeague’s unique method of match play. In LL, defense — that is, the point values assigned to each day’s questions by each player, based on his or her estimation of that day’s opponent’s likelihood of answering each question correctly — plays a critical role. Quite frequently, a player wins or ties a match in which his or her opponent offers more correct answers — simply by virtue of more effective defense. Here’s an example: Player A gets four of the match’s six questions correct; Player B assigned those four questions values of 3, 2, 2, and 1. (Player A therefore missed two questions, valued at 1 and 0.) Player B gets five out of six, earning the following points: 0, 1, 1, 2, 2. (Player B missed the sixth question, valued by Player B at 3.) Player A’s score is 8(4) — that is, 8 points on 4 correct responses. Player B’s score is 6(5) — 6 points on 5 correct responses. Since only the match points, and not the number of correct answers, determines the outcome of the match, Player A wins, despite getting one less question right than Player B.

As a hardcore trivia guy, I sometimes find that system less than satisfying. Ideally, every trivia matchup would be decided purely on the basis of “who knows more stuff.” But the fact is, even Jeopardy!, the venue from which whatever minuscule trivia street cred I possess is derived, works the same way. I’ve certainly won games in my Jeopardy! career where one of my opponents answered more questions correctly, but I happened to get more of the high-dollar-value questions, or a Daily Double or two, correct. It’s how game creators make games competitive and exciting. I get that, and I’m cool with it.

I do, however, like to keep track of my own performance based strictly on my percentage of correct answers. When it comes to LearnedLeague, I’m pleased that I’ve continued to improve in this regard. In my rookie season, I notched 118 correct responses for a .787 batting average. In LL57, my first season in A West, I got 124 answers right, upping my average to .827. In the season just concluded, I scored 125 correct answers (.833). Some of that is pure luck, of course — you happen to get asked things that you know, or can figure out — but I’ve also been working on upping my game by reviewing material in categories where I could use a boost. I also spend at least a bit of time each evening playing quizzes on Sporcle. You just never know when knowing, say, the capital of Burkina Faso will come in handy. (It’s Ouagadougou, in case you were curious.)

Speaking of categories where I could use a boost…

To help facilitate defense, LearnedLeague publishes extensive statistical background on each player’s performance. At a glance, you can survey an opponent’s track record in every category, and see where his or her weaknesses lie. (You can — if you’re really into it — review every question your opponent has ever played, and discover which specific items he or she got right or wrong. I’m not quite that anal-retentive.)

Were you to review my statistical profile, you’d find few surprises if you know me well at all. After three seasons, my highest correct percentages are in Current Events (100%), Television (96.9%), Film (96.7%), Theatre (92.9%), Lifestyle (a catch-all category that encompasses such diverse areas as religion and fashion — 91.7%), and Games and Sports (90%). You’d have predicted that, yes?

Conversely, my nemeses are Art (60%), Classical Music (58.8%), and of course, Math (16.7%). Again, if you know me, you know that my ineptitude in mathematics rivals only my distaste for country music. In fact, I believe that Hell is an eternal algebra class with country music playing at ear-splitting volume over the loudspeakers.

I’ve been doing some brush-up reading on art, and trying to memorize some basic facts about the most notable classical composers. I think it’s helping. Nothing will help me get better at math. If you find yourself facing me in a future LearnedLeague match, and there’s a math question in the day’s sextet, you might as well slap a big fat 3 on that one. (Then again, I do pull one out of thin air 16.7% of the time. So, you never can be too certain.)

For what it’s worth, I’ve also attempted to work on my defense. I’m consistently a subpar — although not altogether horrible — defender, which means I do a mediocre job at assigning points based on my opponent’s perceived strengths and weaknesses. I could probably win an extra couple of games each season just through better defense, so I’m trying to take more time with that portion of each day’s quiz. My defensive efficiency rating improved to .672 this season, from the previous season’s .651, so I suppose I’m doing something right. Or at least, less wrong.

LearnedLeague 59 begins November 11. I’m looking forward to the next challenge.

This entry was posted on September 25, 2013 at 4:01 pm and is filed under Jeopardy!, LearnedLeague, SwanStuff, Trivial Pursuits, Uncategorized. You can subscribe via RSS 2.0 feed to this post's comments. You can comment below, or link to this permanent URL from your own site.