.

.

Townharbour Trophy (32)
Report by Michael Crane

The weekend got off to a flying start in the Friday 500 (14) for Nicky Check. After months of trying to get into Top 16 he finally made it by beating Mardi Ohannessian in the final. He was propelled into 12th place and secures his shot at the £500. See here.

For the Main we had a perfect draw of 32, which nicely split into 16 each for the Main and the Consolation. In the Main (16) we almost had an outright winner when Mark Heidenfeld went into the last round unbeaten on 5/5, I say almost because here he was stopped by Jerry Limb. Jerry's victory complicated matters somewhat for we then ended up with four players on 5/6, Jerry, Mark, Mardi and Peter Christmas! The 'head-to-head' couldn't show an outright winner for although Peter had beaten Mardi and Jerry, Mark had beaten Peter and Jerry had beaten Mark; as one player rose to the top another replaced him! So I was left with the average ranking score of each players' opponents. This returned Jerry as the winner with Mardi Runner-up and Mark 3rd and Peter Christmas 4th. See more on this here. Tim Cross, the 'sponsor of the TT kindly gave £100 which was shared between the four players; a gesture they all appreciated.

The last round match between Jerry Limb and Mark Heidenfeld saw Jerry playing as Advanced and Mark as World Class. Each of them made 11 errors of which two were blunders, so they we evenly matched; and their error rates were acceptable at 2.881 for Mark and 3.548 for Jerry.

They both made blunders when playing a 21. Here's Mark's. He is playing as black.
.

11 point match
Black 0 White 0
Black to play 21

.

Mark's blunder @ -0.115 is moving both checkers 7/5 7/6. Jerry is very likely to get his 13-point checker safe next roll; but just in case he doesn't, the split, 17/16 7/5 is the better play. Even if hit, Mark will probably get a shot off the bar, especially with 5s duplicated. Jerry rolled 43 and safetied the runner and the race began. Soon they were down to five checkers each and a three-roll finish when Mark ships across the cube:
.

Black cube action

A live-cube rollout reveals this as a redouble/pass, but Jerry takes - and loses four points!

Jerry's 21 blunder is a bit larger than Mark’s @ -0.134:
.

Black 4 White 0
White to play 21

Going for another home-board point Jerry moves 11/9 8/7; but Snowie thinks this too risky and prefers the safe play of 16/13. Jerry's play did return slightly more gammons - 19.6% to 18.5%, but his game winning chances we reduced by a fairly large margin - 66.5% down to 60.9%. However, Jerry didn't make another point on his next roll, but he did win the game and two points.

From hereon Mark wins only one more point and in Game 12, Jerry later bears off against a 2-point anchor to win two points and the match 11-5, and in doing so secures himself a place in the UK Finals.

	Mardi and Jerry

.

Because I wanted to record the match between Jerry Mark it was necessary for Jerry to play in the playing-room instead of in the bar where he'd played all weekend. However, the problem was, he was unable to sit on the chairs in the playing-roomand reach the board. The solution?Sit on a cupboard!   Picture courtesy of David Startin

.

Meanwhile in the Consolation (16), Arthur Wright had come through with 3/3 and he then awaited the outcome of the match between Nicky Check and Geoff Conn. It was Geoff that emerged the victor; and on the tie-break it was he that come out the winner.

Arthur	Geoff
.

On the Saturday night the Poker (8) Grand Prix reached a climax when Paul Gilbertson emerged the winner with Tony Fawcett 2nd and Paul Learmount 3rd. We now have our Top Table of eight players (see page 27) who will fight it out for the prize money at the UK Finals.

Also on the Saturday night Geoff Conn won the £50 Jackpot (4); and I won the Team (7) event beating Ian Davidson in the Final. Our initial team's victory was particularly sweet seeing as we were trailing 3-0 down and we went on to win it! In the Round Robin (14), Jonathan Powell took home all the dosh. I failed to qualify this time . . . although I did have only one attempt instead of my usual five!

Finally
It was a good end to the year's backgammon. The weekend went smoothly and everyone enjoyed themselves. We concluded our Ranking and Grand Prix Championships and the results are here.

.

.

.

.

Swiss Format Tie-Breaker
Working out the winner in Swiss Formats when there are more than two players sharing the same number of wins is something we have yet to achieve wherein the criteria is fair to all. I fear we may never be able to do it. The present method using the average ranking score of opponents has proven to be better than the method it replaced - sum of opponent's wins - purely because in the previous method any player who dropped out of the competition had a lower win ratio than those that remained in. The 'average' method reasoned that the lower the average ranking, the 'weaker' the opponents, ergo, the 'easier' the draw. Of course, this is far from true, but it was a fairer way then using win ratios.

The 'head-to-head' method was introduced and is used prior to using the average method and, on occasions it has proven adequate: it has either returned a result of 1st, 2nd, 3rd; or it has failed to do so and the average has been the decider. During the Townharbour, the 'head-to-head' couldn't show an outright winner, therefore I have devised a head-to-head method that might prove more decisive; and unless someone can come up with something better or point out glaring errors in its usage it will be used starting next year.

The method is:
a. 1 point is given for each victory over your those sharing top wins.
The result of this is that Mardi gets 0 points, Mark and Jerry get 1 point each, and Peter gets 2 points. Therefore, Peter is 1st and Mardi is 4th.

b. When two or more share the same number of points another 'head-to-head' is employed. If this doesn't return a position then c. is used.
Mark and Jerry both had 1 point each; however, they had played each other and Jerry had beaten Mark, therefore Jerry would be 2nd, leaving Mark 3rd.
c. If the players haven't played each other then shared opponents are compared to return a position, else the 'average' method is used.
If Mark and Jerry had not played each other then shared opponents would have shown that they both layer Peter and Mark won which would have placed Mark 2nd and Jerry 3rd.

I am not suggesting this is the answer, just that it is an answer to the solution. I am more than happy to hear other solutions and will gladly adopt them should they prove worthwhile and workable.

TOP