Seeking match equity tips
JMarttila
seem to have a heck of a time remembering many of the equities at the
scores, so I am using 'Neil's Numbers'. That helps me quite a bit as I can
calculate, in my head, the odds I am getting or the odds I would be
offering.
However, other numbers such as the early/late cost ratio, for both sides,
and the gammon price, also seem very helpful to me to make decisions
depending on the type of position I am facing. But to calculate everything,
over the board, just becomes overwhelming. So I wonder how other players
learn, calculate, or keep track of these things? I wonder if maybe there are
some 'key scores' that I should concentrate on (for example -3 -4 or -3 -8,
etc. in gammonish positions) and then just adjust from there?
I recently played quite a few matches on fibs using the aid of having the
take and drop points for both sides, early/late ratio's for both sides,
gammon prices, (and similar values for recubes). My rating began to shoot up
(sorry to anyone I beat using those artificial aids, but I wanted the
information), but is dropping back toward my 'stasis rating' now that I am
playing without them and trying to do all that stuff in my head without
taking too much time. I am going to put in the effort to get some sort of
procedure to use in up to 11 point matches since the value of having a clear
idea of what is going on with match equity has become apparent to me.
Help!
Thanks,
John M.
Douglas Zare
JMarttila wrote:
> I am working on doing a better job at using match equity calculations. I
> seem to have a heck of a time remembering many of the equities at the
> scores, so I am using 'Neil's Numbers'. That helps me quite a bit as I can
> calculate, in my head, the odds I am getting or the odds I would be
> offering.
One problem with using a match equity table is that you don't see the effect of
recubes. Of course, the exact match equity matters if you have a large cube, but
most of the time you will have problems with round-off errors, if not
miscalculations.
> However, other numbers such as the early/late cost ratio, for both sides,
> and the gammon price, also seem very helpful to me to make decisions
> depending on the type of position I am facing. But to calculate everything,
> over the board, just becomes overwhelming. So I wonder how other players
> learn, calculate, or keep track of these things? I wonder if maybe there are
> some 'key scores' that I should concentrate on (for example -3 -4 or -3 -8,
> etc. in gammonish positions) and then just adjust from there?
I played a lot of 3-point matches for a while. The numbers don't help much
unless you know what it means over the board, e.g., you should have a feel for
when to double in the first few moves trailing 2-away 3-away. After a while, my
winning percentage in 3-point matches was the same as for 5-point matches
(against inferior opposition).
In a 5-point match, there are some real surprises, and my intuition was wrong.
In most match scores in a 5-point match outside a 3-point match, the racing take
point is higher for the trailer than for the leader. The gammon price is
remarkably high at many match scores. These call for much different doubling
decisions than for money play. Luckily, if you play against opponents who do not
know match play well, many of the take/pass decisions you get will be easy.
I've memorized the stats for a few common scores outside the 5-point match
(e.g., 7-away 7-away, and 2-away 6-away), but there are declining returns. For
low cubes the scores rarely have their own distinctive characters, unlike the
scores within a 5-point match.
> I recently played quite a few matches on fibs using the aid of having the
> take and drop points for both sides, early/late ratio's for both sides,
> gammon prices, (and similar values for recubes).
I can't condone that. Play now, analyze later. If you want to know how well you
play with artificial aids, play against a bot, offline.
Douglas Zare
Hank Youngerman
to 5 points stands me in good stead. Many matches I play are 5points
or shorter, many longer matches come down to that, and at scores
further from the end of the match, initial cubes are not very
score-dependent anyway. One important thing to realize is that in
lopsided matches, gammonish cubes favor the trailer but non-gammonish
cubes favor the leader.
On Tue, 18 Jun 2002 17:53:21 -0500, "JMarttila" <mar...@bellsouth.net>
wrote:
JMarttila
"Douglas Zare" <za...@math.columbia.edu> wrote in message
news:3D0FDFA1...@math.columbia.edu...
>
>
>
> > I recently played quite a few matches on fibs using the aid of having
the
> > take and drop points for both sides, early/late ratio's for both sides,
> > gammon prices, (and similar values for recubes).
>
> I can't condone that. Play now, analyze later. If you want to know how
well you
> play with artificial aids, play against a bot, offline.
>
> Douglas Zare
>
I can't condone it either, actually, but I went ethically astray. My method
was to figure out the numbers in my head or as best I remembered, then check
my sheets to see if I was correct. Probably played 7-8 matches this way.
Then, not only did I not like being unethical in the manner of using
artificial aids, it did not seem to be helping me learn/memorize the numbers
anyway. So I ended the practice. (I think I will do that against an offline
bot though, for the practice.) Thus this plea for help in how to remember
all this stuff! :-)
JMarttila
"Hank Youngerman" <Red...@RedTopBG.com> wrote in message
news:3d0feedb....@news.charter.net...
> I have found that knowing all the relevant situations for matches up
> to 5 points stands me in good stead.
I think I am getting a decent feel for the 5 point match, but have been
playing more matches of 7-9 points to try to improve my knowledge of the
scores, since in club play we usually play 7 or 9.
One important thing to realize is that in
> lopsided matches, gammonish cubes favor the trailer but non-gammonish
> cubes favor the leader.
>
>
>
Hmm. I hadn't really thought of it in that manner except at certain scores.
It seems to me that at some scores the gammon doesn't matter as much to the
trailor as the leader. I'll have to look at the overall picture of gammons
with the cube involved to understand this idea. Thanks for the tip.
JMarttila
wrong.
> In most match scores in a 5-point match outside a 3-point match, the
racing take
> point is higher for the trailer than for the leader. The gammon price is
> remarkably high at many match scores. These call for much different
doubling
> decisions than for money play. Luckily, if you play against opponents who
do not
> know match play well, many of the take/pass decisions you get will be
easy.
>
> I've memorized the stats for a few common scores outside the 5-point match
> (e.g., 7-away 7-away, and 2-away 6-away), but there are declining returns.
For
> low cubes the scores rarely have their own distinctive characters, unlike
the
> scores within a 5-point match.
>
Thanks Douglas. I think that's what I'm going to have to do. Just buckle
down and pretty much memorize what I want within the 5 point match and then
some scores outside of that for longer matches. I suppose the key is to
memorize the ones where my own intuition steers me wrong.
Ric
definitely help you remember the numbers. Any numbers.
Ric
"JMarttila" <mar...@bellsouth.net> wrote in message
news:vrTP8.8011$jT.6...@e3500-atl1.usenetserver.com...
Adam Stocks
formula to keep in your head, which calculates the gammonless (racing)
takepoint at a given score:
racing takepoint = (Pass - TL) / (TW - TL)
where
Pass = match equity if you pass,
TW = match equity if you take and win single game
TL = match equity if you take and lose single game
e.g. for trailing -4,-3 (using the Woolsey match equity table),
racing takepoint = (32 - 17) / (60 -17) = 15/43 = approx 35%
This is higher than the normal money takepoint of 25% (dead cube), so if you
think it is a borderline money take, you can drop at a -4,-3 score with
confidence. This will save you from learning a whole heap of individual
takepoints, as you can work them out over the board very easily. However,
as Doug said, simple formulas like this don't allow for cube ownership after
a take, so you need to make an adjustment for that (usually by subtracting a
few percentage points off the takepoint). Having said all that, I agree with
Hank that it isn't worth spending undue effort learning reams of table
entries, at the expense of studying other aspects of the game which probably
have a more immediate impact on your overall playing strength. Numbers are
an added bonus, but most of the game hinges on getting a feel for different
classes of positions, and making rough adjustments from positions and scores
for which you know a reasonably accurate value.
Adam
"JMarttila" <mar...@bellsouth.net> wrote in message
news:TEOP8.3443$jT.7...@e3500-atl1.usenetserver.com...
JMarttila
"Ric" <nos...@nospam.net> wrote in message
news:Pv%P8.1178$CZ4.4...@news1.news.adelphia.net...
> I strongly recommend you get Harry Lorayne's 'The Memory Book'. It will
> definitely help you remember the numbers. Any numbers.
>
> Ric
>
>
Hey, great idea Ric. Thanks. I remember that book from high school. At the
time I just sort of skimmed it and yet it helped me remember some stuff for
my classes. I'll get it from the library and read it again, more thoroughly.
Ric
Literally.
"JMarttila" <mar...@bellsouth.net> wrote in message
news:nkqQ8.7$HT1...@e3500-atl1.usenetserver.com...
Back4U2 BBL
But:
Harry Lorayne, The Memory Book....
is worth being read... and studied ;-)
Nardy
"LostVegan" <NO.JUNKl...@hotmail.com> wrote in message
news:2um4hugth0munn1i1...@4ax.com...
> On Thu, 20 Jun 2002 21:05:50 GMT, "Ric" <nos...@nospam.net> wrote:
>
> >When I was really into it I could memorize a full deck of cards in a
minute.
> >Literally.
>
> When I was really into it (before programmed speed dialers), I could
> memorize a phone number in a minute. Literally.
> --
> Marty (to respond via email, remove NO.JUNK from email address)
>
> "to be yourself, in a world that tries, night and day, to make
> you just like everybody else - is to fight the greatest battle
> there ever is to fight, and never stop fighting" -- e.e. cummings
Ric
Ric
"LostVegan" <NO.JUNKl...@hotmail.com> wrote in message
news:2um4hugth0munn1i1...@4ax.com...
> On Thu, 20 Jun 2002 21:05:50 GMT, "Ric" <nos...@nospam.net> wrote:
>
> >When I was really into it I could memorize a full deck of cards in a
minute.
> >Literally.
>
Ric
Ric
"Back4U2 BBL" <nardy.p...@skynet.be> wrote in message
news:3d12ea1f$0$6938$ba62...@news.skynet.be...
Back4U2 BBL
> There are seven days in the week? Oh oh.
>
> Ric
>
Actually there are 5.
5 weekdays.
And there are 2 in a weekend.
But: not every week has 5 weekdays, sometimes there is a holiday.
And the days of the weekend are from time to time loaded with a holiday.
Depending on your enthusiasm (and the country you live in), this cuts down
the 365 days in a year to 220 - 300.
Giving an average of 21.67 days per month.
5.42 days per week.
So I guess that is why I don't see (yet) any use in that 7th day.
( I do realize I have to adapt this for leap years. But they told me that is
to be taugth in the 2nd year.... of basic school )
Nardy
Ric
Ric
"Back4U2 BBL" <nardy.p...@skynet.be> wrote in message
news:3d148e31$0$8152$ba62...@news.skynet.be...
Ric
Ric
"Back4U2 BBL" <nardy.p...@skynet.be> wrote in message
news:3d148e31$0$8152$ba62...@news.skynet.be...