Help On Pip Counting Methods
Stephen Turner
NYCGuy wrote:
>
> [how to count pips]
> I would like to know, from experienced tournament players, which method(s) they
> employ, which are easiest to learn and use, and any other thoughts and/or
> observations they have on this subject. TIA.
>
Well, I'm not sure whether I count as an "experienced tournament player", but
for what it's worth (and I don't think you're going to like this), I vary.
Out of preference I slide the pieces around in my head, adding and subtracting
the number of pips moved, until the pieces all match up. (Is that what Magriel
calls "mental shift"?). It's important when using this method to make sure that
you don't add when you mean to subtract or vice versa!
There are circumstances in which I just count them, however. These are broadly
1) When the positions are sufficiently dissimilar that "mental shift" turns
into "mental contortion";
2) When the cube decision is sufficiently close that you need to check the
length of the race (you then need to count at least the leader's side);
3) When I want to do a Thorpe count (let's leave that for another day).
Top tips for doing a direct count.
* Start at the high end; it's easier to do 3 times 23 and then add on 4
times 4 than do 4 times 4 and then add on 3 times 23.
* Count one side, and then store the answer on your fingers in your lap to
save having to remember it while counting the other side. It's really easy
to do this. Record the 10s on your left hand, one finger down for 10, 2 for
20 etc.; one down curled up for 60, 2 for 70, etc. Same for the units on
your right hand. Ignore the 100s; you know them.
[Computer scientists might prefer to store up to 1024 on their fingers by
representing the number in binary].
Hope at least some of that helps.
--
Stephen Turner sr...@cam.ac.uk http://www.statslab.cam.ac.uk/~sret1/
Stochastic Networks Group, Statistical Laboratory,
16 Mill Lane, Cambridge, CB2 1SB, England Tel.: +44 1223 337955
"Collection of rent is subject to Compulsive Competitive Tendering" Cam. City
Jim Wallace
> nyc...@worldnet.att.net (NYCGuy) writes:
> I have decided to mount an attack against my pip-counting phobia.
> In spite of cold sweats, clammy hands and uncontrollable shivers at the mere
> thought of attempting to tackle this faceless beast, I am determined to force a
> confrontation.
> I am aware of the Direct, Comparison and Mental Shift methods as illustrated in
> Magriel's Backgammon, although I have not mastered them (OK ... the above
> symptoms prevented me from even _attempting_ them) to the point that I really
> cannot get accurate counts in complicated positions.
> I would like to know, from experienced tournament players, which method(s) they
> employ, which are easiest to learn and use, and any other thoughts and/or
> observations they have on this subject. TIA.
>
Nobody can count a position faster than jack kissane. (jkissane on fibs) Jack illustrated his pip
count method in an article in the Chicago Point about 2 yr. ago and it provides a terrific method
of grouping reference positions as a basis for the pip count. (examples. closed board with 2
checkers on every point = 42, Two checkers directly opposite each other on a board = 25, 10 times
the centre checker point in a five prime counts those 10 checkers and so on.)
I modified my counting methods after this and it improved my skills but basically you still have
to do the work and practice. The worst way to learn is use the pip count on FIBS or FIBS/W
software. You get the right total but you will be useless once you play someone face-to-face. Do
the count in your head first then let fibs check to see if you are right.
If you know someone with back copies of the Point maybe ask to see it. Perhaps Bill Davis might
repost the article here or at his website (Bill?)
Jim Wallace
Mark Denihan
Back in 1981, while struggling with cumbersome pipcounting methods, I
thought up a method that allows me to get an accurate count quickly and
easily. Jack Kissane is still a faster counter than me, but most other
players aren't. I asked several prominent players about this method and
none had ever heard of it except Magriel; so I guess I am a co-inventor
with someone.
I call it Casting Out Crossovers.
All you need to learn are a few multiples of six and how to recognize
some patterns of ten.
The fundamental idea is that to move a checker from one quandrant to the
same position in the next quadrant takes six pips. So if you count all
the crossovers needed to bring all the men into your home board and
multiply this number by 6 you have reduced each quadrant into a home
board position with pip count values of 1-6. Then you can learn to
recognize patterns of 10; such as 2 men on the 2point and 2 men on the
3point, and just keep adding these tens onto the first number you
obtained when you multiplied by 6. Then add on the few extra pips that
did not fit conveniently into patterns of ten. Thats it. Gone is the
problem of trying to multiply and add or subtract different groups of
numbers or keep mental shifts in your memory. It's easy for us humans to
increment by tens.
I wrote this explanation off-the-cuff so to speak so if anyone wants a
clearer explanation with examples let me know. I find that I am faster
and more accurate at counting than most good players even though I have
no natural talent for counting. It may take you a few days to build up
speed.
Mar3k
Robert
Is it necessary to know the exact count such as 112 to 95 or is it ok to
know only the relative count of +17 in this case. I am new to the game and
if the relative difference is all that is needed let me know. I have a
great idea on this. Thanks
Hadler
had...@richnet.net
Kit Woolsey
Robert (Had...@richnet.net) wrote:
: Is it necessary to know the exact count such as 112 to 95 or is it ok to
: Hadler
: had...@richnet.net
For the most part, the relative difference is all that is necessary. If
you have a great idea, send it on in.
Kit
Stephen Turner
Robert wrote:
>
> Is it necessary to know the exact count such as 112 to 95 or is it ok to
> know only the relative count of +17 in this case.
Good question. You really need to know the exact count. If the game has become
a straight race, then the lead you need to double is expressed in terms of a
percentage of the distance left to go. Suppose that the leader has L pips to go,
and the trailer T (L=95 and T=112 in your example). As a guide, we reckon that
the leader can double if the difference (17) is more than 8% of L, redouble if
it is more than 9%, and that the trailer can take if the difference is less
than 12% of L. So you need to know L.
Having said which, it's often easier to count the difference by comparing the
positions on both sides of the board without actually calculating L and T;
also, once you've counted the difference, you can keep a running tally. And
sometimes the difference is sufficiently large or small that it tells you all
you need to know. In your example, it would be obvious that it was a double/
pass, because you could see with a bit of experience that the race wasn't long
enough to take with a deficit of 17, without having to count the race length.
Similarly, if the pipcounts were 95 and 99, it would be clear that it wasn't
yet a double without counting exactly. So in practice you only need to know L
in the close cases; otherwise the difference suffices.
HADLER
To evaluate my method of counting a position I need some benchmarks from
the pros out there:
1. What is the average time and best time to count a random position with
men spread all over the board?
2. What is the average time and best time to count a position with all
the men in the home boards?
Thanks for your expertise.
Simon Woodhead - Uniq Professional Services
Many of the top players I know actually keep a running count, right
from the starting position of 167. It doesn't take much practise to
do this, and gives you an instant pip count all the time.
Anyone who's prepared to work at counting pips will actually find
this easier than all these fancy methods of counting crossovers
and relative positions.
-Simon (simonw on fibs)
Simon Woodhead si...@uniq.com.au | If you're not
Uniq Professional Services Pty Ltd ACN 056 279 335 | confused...
PO Box 70, Paddington, NSW 2021, (Sydney) Australia |
Phone: +61-2-9380-6360 Fax: +61-2-9380-6416 | you're not
URL: http://www.uniq.com.au | well informed