Hot streak

My dice have been hot lately.  I won in Flint recently, and also tied for best individual record in the Michigan versus Ohio challenge.  Below is a copy of the item from flintbg.com, which I wanted to capture before it disappears.

***************

CONGRATULATIONS to BILL CALTON, our 2015 FLINT CLUB CHAMPION!

October 18th: 37th Flint Area Club Championships
Holiday Inn-Gatewood Centre, Flint, Michigan

OPEN (26)
1-BILL CALTON (Harper Woods)
2-Dan Wittkopp (Traverse City)
3-Greg Merriman (Ann Arbor)
4/5/6-Jim Airgood (Swartz Creek) / Ben Friesen (Davison)
/ Emil Mortuk (St. Clair Shores)

INTERMEDIATE (14)
1-Dick Hunt (Swartz Creek)
2-John Claybourne (Detroit)
3-Mark Van Poucker (Harper Woods)

***************************

September 26-27: 5th OHIO-MICHIGAN TEAM CHALLENGE

MICHIGAN WINS: 34-29!
Team Michigan: Paul Farah / Dmitriy Obukhov / Paul Berg / Ben Friesen /
Bill Calton / Rick Todd / Dan Smith (not in photo) / Carol Joy Cole (Captain).
***********************

  

Bill Calton, Ben Friesen & Christopher Yep (OH)
tied for the best individual records of 6-2,
winning USBGF prizes and splitting the Side Pool.
[Photo taken by our hostess Karen Davis in Toledo, Ohio.]

 

Team Ohio: Karen Davis / Farhad Forudi / Mary Hickey / Captain Joe Miller /
Alireza Saatchi / Howard Shirey / Mike Vasilatos / Christopher Yep.

Beware Konrad Froeschl

Warning to the general public.

Konrad Froeschl, handle "KONFRO," has stiffed me for a significant amount of cash in online money play.  Play him at your own risk.  Please spread the word about his bad conduct and sportsmanship far and wide.

Thank you

Bill

Review of Conquering Backgammon

I prepared this review for Carol Joy Cole's Flint Backgammon News about six months ago.  I haven't seen a new issue in some time.  I don't believe this has been published though a version of the below may appear in the future. Enjoy, Bill 

***
edited 12/1/15 to match the review as published in the newsletter

*****************

Ed Rosenblum’s “Conquering Backgammon” is an ambitious work.  Ed aims to take the reader from novice to advanced payer, which he loosely defines as a club or tournament player.  The book has two self-contained volumes. The first 100 pages or so cover the basics while the last 150 pages cover more advanced topics.  Seasoned players are invited to read the book as well, reinforcing what you already know while picking up some new things along the way.  

Here is a short review of the contents.  The basic section starts with the rules, then quickly moves into technical aspects like dice probabilities and shot counting.  Next is a section on basic strategy, followed by more technical stuff on opening moves, when to double and take, and pip counting.  Ed then annotates a running game move by move over about 30 pages.  He concludes with more technical guides on bearing in, bearing off, playing for the gammon, and saving the gammon.

The advanced section begins by explaining the concept of equity and how to interpret computer evaluations.  There is a relatively lengthy section discussing game plans.  Ed considers Race, Prime, Attack (hit now) and his own invention Ambush (hit later).  Attack would be similar to Blitzing while Ambush includes Holding games and Backgames.  Each of these game plans are considered separately.  Next up is a lightning review of strategic concepts and playing tactics.  Ed returns to the topic of Doubling, going more in depth.  There is a lengthy section on Match Play and a concluding selection of checker play problems.

The several reference type materials are one of the most successful aspects of the book.  These are styled similar to Ed’s popular “Expert’s Guide to Winning Backgammon.”  Among these are guides for opening moves, opening replies, match equity, doubling process and race formulas.  Ed clearly put a lot of time into formatting these guides and it shows.  I expect the reader will likely refer to these time and again over a long period.  

Another strength is the section on Pip counting.  This is lacking in most books, so it was nice to see more detailed coverage here.  Adding a few quiz questions is a nice touch too.

I also liked the section on doubling.  Ed appears to be influenced by Phil Simborg’s teaching methods.  Here Ed recommends using Joe Sylvester’s well-known TRAP (Threat Race and Position), John O’Hagan’s Market Losers, Woolsey’s Law and Simborg’s law to guide your cube decisions.  While these sections are brief, he has summarized the material well.

I have mixed feeling regarding the physical book.  The book itself is a beautiful 7 x 10 hardcover printed on high-quality paper and features lots of multi-color diagrams.  You can tell Ed is emphasizing quality, and is expecting the reader to use this book over a period of years.  Yet it feels cramped.   I wonder if a larger 8 x 11 softcover format using cheaper materials might have been a better choice.  This would space things out a bit more, allowing the reader to take notes directly in the book, and possibly lower the price tag a bit too.

One weakness of the book is in organization.  Robertie and Magriel bring the reader along slowly.  After the rules, they dive straight into a lightly annotated running game, introducing other topics as they occur in the flow of the game.  Contrast with Rosenblum.  I fear the novice player would be intimidated to learn dice odds, pip counting, opening moves, and doubling strategy before you even start playing backgammon on page 61!  

I also find Rosenblum lacking insight.  Consider the section on dice odds for example.  Ed spends a few pages discussing the number of ways to hit a checker at various distances.  While counting the exact number of shots is a useful skill, Ed fails to turn the data into knowledge.  Magriel does when he observes that if you have to leave a direct shot, move closer to lessen the shots, and if you have to leave an indirect, leave it further away to lessen the shots.  Another example is on page 63 where he analyzes the opening sequence 31 point, 41 split, 53.   He analyzes four different moves over a page and a half before concluding that hitting is correct with 24/16*.  Of course it is.  But where is the insight into basic goals of opening play that Robertie might offer?  Or even some basic opening maxims like a good checker play usually hits, escapes, or makes a point?

Further I find the book lacking in depth.  Take one of the meaty sections of the book, on game plans.  There is good material here, but 5 or 6 pages on each game plan is hardly enough to conquer it.  Each of these topics are deserving of a full length book in their own right.  Looking at Backgames, we are presented a handful of positions over six pages.  We get a bit on timing but we don’t get discussions from the backgamers point of view on circulating checkers, keeping pure, when to hit, containment after the hit, when to take.  Defending the backgame, we don’t learn how to build the prime to bust the timing, to recirculate our own checkers, when to double, to control the outfield, etc.  It is good for what it is, an introduction to backgames and no more.

Finally I acknowledge an apparent inconsistency in my criticism.  On one hand I feel the book overwhelms the reader with too much detail too fast.  Yet on the other hand I feel the book is lacking in depth.  Will the aspiring beginner embrace or be turned off by the heavy dose of technical material in this book?  Are there really enough details on game plans for the beginner to learn what he needs to know?

Overall I feel the book falls short of the classic texts by Magriel, Robertie, and Trice.  Of course this a tough bar to measure against!  As described above, I find the book lacking in organization, depth and insight.  To Ed’s credit, he has perhaps done a better job in some aspects like Match Equity, Pip counting and the various reference materials  What the heck - if you absorb everything in the book you will indeed be an advanced player.  

Bill

Julius High

Below are selected errors from my match against Julius High in Cleveland.  Unfortunately Julius has only only been able to provide me with a portion of the match.  These positions cover the first several games only.


*************

1) First game of 9 point match.  Cube Action?

is Julius High score: 0 pip: 139

9 point match

pip: 127 score: 0 is Bill Calton

XGID=-aB-BaCBA---cC---bBdb-b---:0:0:1:00:0:0:0:9:10

on roll, cube action?

Analyzed in XG Roller++	No double	Double/Take
Player Winning Chances:	64.01% (G:24.71% B:0.51%)	64.08% (G:25.28% B:0.52%)
Opponent Winning Chances:	35.99% (G:9.31% B:0.34%)	35.92% (G:9.42% B:0.35%)
Cubeless Equities	+0.435	+0.906
Cubeful Equities
No double:	+0.597 (-0.043)
Double/Take:	+0.640
Double/Pass:	+1.000 (+0.360)
Best Cube action: Double / Take

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I rolled.  Computer gives this as a clear double.  Let's take a quick look at PRAT.

Julius has the better board but I am anchored and he isn't.  (POSITION about even)

I am up in the race (a modest RACE lead)

but the key is the position is volatile and gammonish. (Big time THREATS)

Most likely I hit loose or hit two (point on him if lucky) when Julius is under pressure to perform from the bar.  The position is an easy take of course, since Julius has plenty of ways to win this game.  The big sin is to win a gammon with the cube on one (which is actually what happened).

**************

2) Same game.  53 to play.

is Julius High score: 0 pip: 145

9 point match

pip: 103 score: 0 is Bill Calton

XGID=bAB-CBBAA---cA---bBdb-b---:0:0:1:53:0:0:0:9:10

to play 53

1.	XG Roller++	18/13 4/1	eq: +1.538
	Player: Opponent: 88.68% (G:73.59% B:0.61%) 11.32% (G:1.96% B:0.14%)
2.	XG Roller++	13/8 4/1	eq: +1.506 (-0.031)
	Player: Opponent: 87.70% (G:71.72% B:0.59%) 12.30% (G:1.91% B:0.11%)
3.	3-ply	8/3 4/1	eq: +1.383 (-0.154)
	Player: Opponent: 82.50% (G:65.75% B:0.64%) 17.50% (G:4.38% B:0.27%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I moved 13/8, 4/1.  This covers the blot and brings another man around to complete the closeout.

Certainly a reasonable move, but a small technical error.  Better is 18/13, 4/1.  Julius counseled me on such positions.  He observes the key is whether the extra builder is diversified or not.  Here the extra man lands on top of my 8 point.  So even though this is an extra man, it's not a diversified or effective builder for the 3 point.  Therefore, it's slightly better to scramble the last man home as this is slightly better achievement for one roll.  The anti-joker 33 could be trouble down the line.

If the spare moved the 9 point (say with a 43) then bringing the builder down would be right. 

*************

3) I lead 7 away 5 away.  own the cube and have this 26 from the bar.

is Julius High score: 4 pip: 67

9 point match

pip: 119 score: 2 is Bill Calton

XGID=-BBABBB--A------a--cdcbBbA:1:1:1:62:2:4:0:9:10

to play 62

1.	XG Roller++	Bar/17	eq: -0.966
	Player: Opponent: 10.47% (G:0.00% B:0.00%) 89.53% (G:20.19% B:0.62%)
2.	XG Roller++	Bar/23 9/3	eq: -1.048 (-0.082)
	Player: Opponent: 7.74% (G:0.00% B:0.00%) 92.26% (G:21.28% B:0.61%)

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

I made the board.  Bar/23, 9/3.  I make the board and don't want to get hit.  The problem is I am under immediate pressure to roll a 5 or 6 next turn.  The problem is this happens on 16 total rolls!!

Note this happens more often than I get hit by popping out with Bar/17.  I get 11 times immediately versus the 16 times I bust so popping out is actually relatively safer!!

Clearly popping out is the DMP play but the gammons are only a tiny bit higher.  This is perhaps the big eye-opener of the position.  It looks scary to volunteer a direct shot but we only get gammoned 1% more often.

************

4) I trail 7 away 3 away and own the cube.  Cube action?

is Julius High score: 6 pip: 113

9 point match

pip: 168 score: 2 is Bill Calton

XGID=---BbaCBC---------AcccDc--:1:1:1:00:2:6:0:9:10

on roll, cube action?

Analyzed in XG Roller++	No redouble	Redouble/Take
Player Winning Chances:	50.43% (G:11.30% B:0.44%)	50.57% (G:11.36% B:0.47%)
Opponent Winning Chances:	49.57% (G:23.78% B:0.75%)	49.43% (G:23.88% B:0.72%)
Cubeless Equities	-0.024	+0.673
Cubeful Equities
No redouble:	+0.612 (-0.061)
Redouble/Take:	+0.673
Redouble/Pass:	+1.000 (+0.327)
Best Cube action: Redouble / Take

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

Julius has violated Stick's rule for 3 away 6 away or more. Namely "never double with contact." The proper doubling window for the leader is very difficult to find. You are better off trying to win an undoubled gammon or to simply double the opponent out from a winning position.

The reasons are:

1) leader cripples his own gammons (a gamon wins four points when only three is needed to win, thus creating overage). Leader's gammon value is only about .35 at this score.

2) trailer has tremendous cube leverage. Trailer should be looking any excuse to send a recube. reasons:

a) Leader's gammons are killed. A single loss or doubled loss loses the match either way.

b) Leader holds a dead cube. He can never redouble you out and must get all the way to 100%. Sometimes you turn around a hopeless game late that you would otherwise lose at a normal score since opponent can't double you out.

c) Trailer's gammons are now quite valuable. A lucky gammon now scores 8 points and suddenly wins you the match!!

Looking at the computer's Cube Information function, it appears the doubling window opens at about 40% and closes at 70% (Not adjusting for gammons). So Julius has an elevated take point of 30%. Once again, this doesn't factor in the gammons.

In this position I am already about 50/50 with 11% gammons and as you can see, this is a powerful recube. The lessons...

1) You can take deep as the Trailer at a lopsided score

2) If you take you MUST recube aggressively to get full value out of the take!

3) My own "rule of thumb" for these lopsided scores is to think about the recube nce you get to about 50/50 game. Too bad I didn't follow this advice on this game.

************

5) Next roll of the same game.  Cube action?

is Julius High score: 6 pip: 113

9 point match

pip: 164 score: 2 is Bill Calton

XGID=---BcBBBB---------AcccDba-:1:1:1:00:2:6:0:9:10

on roll, cube action?

Analyzed in XG Roller++	No redouble	Redouble/Take
Player Winning Chances:	60.54% (G:14.52% B:0.56%)	60.94% (G:14.85% B:0.63%)
Opponent Winning Chances:	39.46% (G:15.74% B:0.36%)	39.06% (G:16.22% B:0.34%)
Cubeless Equities	+0.275	+1.208
Cubeful Equities
No redouble:	+0.908 (-0.092)
Redouble/Take:	+1.208 (+0.208)
Redouble/Pass:	+1.000
Best Cube action: Redouble / Pass

eXtreme Gammon Version: 2.10, MET: Kazaross XG2

OK so there isn't too much drama on this one. Given my writing above we know this is a double. I blew it by not doubling.

I included the position to demonstrate that this position is already a drop according to the computer!!

If I understand the computer correctly, then with a four cube the trailers gammon value is 1.0 while the leaders takepoint is 30%

So from the leaders point of view, Julius would need 30% + 2 x my gammons. So Julius would need 30% + 28% of 58% to take and he is way short of that here. I wonder if I made a mistake on this since 58% seems an absurdly high takepoint and the take error, while large, is only .20. Oh well, math is not my bag, baby.