The statistics that Mike has been carefully maintaining since Round 1000 are now such an expected fixture that it has only lately begun to occur to me that they are quite unlike most of the leader boards you will find for other games or sports.
Chess and Scrabble players are accorded an elaborately computed rating after every match or tournament. Our equivalent to that is (I suppose) the 5-round rolling scores report. But there are other, simpler league tables, such as you might find at your local squash or tennis club, that list wins, draws and losses. Our statistics, on the other hand, were called into being by rules that put the emphasis on amassing points, and will not even tell you how many times a player has won a round.
Now, the distinction win/draw/loss applies chiefly to games with two opposing players or teams. Coming second in a round of Dixonary can’t be accounted a loss: in fact the runner-up is often called “the real winner” because they will earn points from the subsequent round, and the actual real winner will not. And in Dixonary, ties (which you might also call draws) are very frequent.
Even so, there is such a thing as a “winning score”, but we don’t publish figures showing how many times a player has achieved it. So I thought I would see how that would pan out, and the results are below.
In the table, the “winning score” is the one numerically highest in the round, but with guess-assisted scores considered to rank lower. So, 6 > 6* > 5, and so 6 is higher than the other two, and is the winning score. In cases of equality, where the new dealer can be determined only by reference to the rolling or total scores, this table considers the highest score to be a winning score for all of the players who achieved it. An “outright win” is where a single player scored highest.
Of course, players can win rounds only by playing in them, and so these figures exhibit the same bias as the total score: the primary determinant of your number of wins is how long you have been playing. The customary remedy is to divide the number of interest by the number of rounds played. That is left as an exercise for the interested reader.
--
To view this discussion visit https://groups.google.com/d/msgid/dixonary/E1vO0Vc-000BgW-0B%40sslproxy07.your-server.de.