
BGonline.org Forums
Early v. Late Round Byes
Posted By: Henrik Bukkjaer
Date: Wednesday, 27 November 2013, at 12:57 a.m.
In Response To: CT tournament review  brief (Bob Koca)
After having been away from bgonline a few months, I quickly browsed through recent treads that seemed to have draw a lot of interest and activity, and fell over this one.
Haven't read ALL the posts thoroughly, but I have a question to Bob & Jason:
You say that early vs. late round byes are the same. When you do so (from a mathematical perspective), have you considered:
1) A lot of tournaments have increasing match lengths.
2) Not all players are equally strong, thus the expected strength of an opponent is not the same in the first round, and in the semifinal.
3) When you get a bye, you also get a break. If you're playing a big 2day tournament with long matches, such a break will be much more worth, given late in the tournament. A break in the first round will carry the least value. (I know some directors prefer to "skew" the byes, in order to be able to start all matches in the first round  I think this is a misunderstanding from a sporting perspective, but I can see why it might be attractive from a schedule point of view). Still, a first round bye in a "skewed" tournament tree, will NOT carry the same "break value" as a late round bye will (eg. semi or quarter finals).
4) In DBgF we sometimes seed players (it depends on the type of tournament). I don't know if you ever seed players in ABT events (to ensure that the current ABT Leaders doesn't meet each other in round one, or to ensure approximate evenly distributed strength in your tournament trees). Have you considered the effect seeding will have on the value of first vs. late round byes?
5) Spectators, PR, appeal and sponsors. A late round bye can be a turnoff for other reasons than the mathematical ones (winning probability and equity). As an example, in the just completed Danish Championships, the field was cut to 32 players going into the final cup. Byes and other tiebreaks were used in the first rounds, to get to that nice 2^n number of players. The finals were sponsored by the national bookmaking company, who put odds on all final 32 players for the entire country to bet on. Late round byes would probably not have been perceived as "the same" by the sponsor, the bettors and the spectators following the last 32 finals.
So, even though (I think) I understood your points, I'm not fully convinced that your claim of mathematical equivalency is true.

BGonline.org Forums is maintained by Stick with WebBBS 5.12.