Let's have a definitive test of skill
Posted By: Bob Koca In Response To: Let's have a definitive test of skill (Phil Simborg)
Date: Tuesday, 24 September 2013, at 6:52 a.m.
In Response To: Let's have a definitive test of skill (Phil Simborg)
Agree with Rod's thought that it is meaningful but not as much as Phil seems to think. Here are some other issues (some of which overlap with Rods issues).
i) Suppose that player A plays an opening 43 one way and player B plays it another way. The match has to go along certain lines. It is unfair to the player who doesn't naturally get into such positions.
ii) You only want to ask tough questions. Brain farts on easy positions happen and making those only say 0.1% of the time compared to 0.5% of the time is important and not given credit in your system. Also here consider what %age of moves in a tournament are what you consider difficult plays worthy of being on such a quiz.
iii) Some think that some players have an ability to sense when an opponent will do a bad drop (or maybe a bad take of a TG position) and can benefit from a technically incorrect cube decision. Your test gives no value at all to that valuable skill.
iv) A match has a logical flow where thinking about one position leads you to the next position. In a match for example a player might make a certain play and already decide something such as cash if there is a dance. In your test though the player is suddenly being forced to deal with a new position.
v) Managing time wisely is a skill in a clocked match. Your test does not measure that skill and just giving a time limit to take the test is not the same thing.
vi) Even if a tournament is only using top players the best practical play might not be the XG play.
vii) Quiz factor on quizzes is a consideration.
Basically if you want to test how much skill one has at playing backgammon one should look at how well one plays backgammon. Variance reduced results would tremendously cut done on the noise.
Messages In This Thread
BGonline.org Forums is maintained by Stick with WebBBS 5.12.