| |
BGonline.org Forums
How we will break 3-way ties at the London Open - a new approach
Posted By: Colin Owen In Response To: How we will break 3-way ties at the London Open - a new approach (stw672)
Date: Tuesday, 23 August 2016, at 3:23 p.m.
Everyone has APPROXIMATELY a 33.33% chance of going through (ignoring skill differences, of course). The Kazaross XG table gives 33.13% at 2a 4a , but 'Can a Fish Taste Twice as Good?' has as low as 31.4%. Whatever the chance really is, it cannot for sure be EXACTLY 33.33%. Giving the high roller the choice of whether to be A/B or C, is good of course, as it gives players the chance to exercise judgement here. But, unless only one player wished to be C, the roll-off will affect the odds.
Of course, rolling high dice to decide who has the choice of being A/B or C is much better than using those rolls to decide who goes through! But, if you want to combine skill with a truly level playing field, you have to have a round robin. That, of course, leaves the possibility, albeit small, of an interminable series of matches, with each player winning one out of two, and having to roll for high dice in the end anyway. But, at only a maximum 25% occurrence of a 3 way tie (less, if there are any skill differences) it shouldn't take too long. The method the UKBGF have decided upon certainly expedites the situation, and introduces more skill than 1 point playoffs, albeit at the expense of a slightly uneven playing field.
In the inaugural St Albans Open this spring, they used 1 point matches in a round robin for this purpose, but expedited the process by simply declaring the first player to win 2 consecutive matches to be the winner. This means that the draw to decide who plays in the first match will affect the players chances so, no level playing field, once again. Personally, I think the best way is to have a true round robin, probably one point matches, with mandatory clocks for these playoffs (which they did have at St Albans, at 12-4 TC for the one pointers). If it's still a tie after the 3 matches then you halve the reserve times, repeating this process each time (if necessary) until you get to a minimum of perhaps only 15s reserve. At some point, of course, you would have to decide the winner by lot, but the odds on it being unresolved after 5 sets of 3 matches is longer than 1000/1. With one pointers and the approximate (diminishing) TC suggested here, even two sets of three matches should take about the same time, on the average, as the London Open 5a 5a and 4a 2a method. A purist might prefer my method, but yours will certainly work well in practice.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.