What odds?
Murat Kalinyaprak
Murat Kalinyaprak wrote <7qvh87$po...@taisp3.in-tch.com>...
>>Assuming I'm rated 200 points below JF level 7/1000,
>>FIBS formula gives me about 24% chances of winning a
>>25-point match. With a 300-point gap in ratings that
>>number drops to about 15%.
>>What I would like to know is the chances of me or JF
>>winning a 25-point match with a score of 25-0. Could
>>anybody offer any suggestions as to how to go about
>>calculating such odds...? Thanks.
>..... p = 0.9 and the chance of winning 25-0 = 0.0717.
>The above calculations assume no cubes and no gammons.
>If your chance of winning a gammon in any single.....
>....You can build up and up from each possible score
>....If you want to factor in the cube you would .....
Ian, thanks for the effort but cubeless/gammonless
figures aren't useful for what I had in mind and I
was hoping that somebody would actually take a few
steps forward into at least calculating some rough
estimates...
What I was after was to solicit what odds anybody
out there might offer me if I were to play against
JF 25-point matches with the intention of winning
with a score of 25-0. Depending on how slim may be
my "estimated" chances of accomplishing it, I may
have to play perhaps several thousand matches or
the odds could be adjusted down to a reasonably
large number like 1000 matches. This may seem like
an enormous task but could actually go quite fast
since even 1 point lost would mean the whole match
lost and I think I could play 50-100 matches a day
on a real fast PC.
Anybody willing to come up with some estimates and
try to explain how they would justify them...?
>Of course, if you are ONLY interested in winning 25-0,
>then your tactics would change. You could double
>immediatley and you would only have to win 13 games in
>a row.
Yes and I would, of course, take every double from JF.
I love this... :) If you guys won't buy an argument
just because MuratK makes it instead of some famous
world-class player (who probably made lots of money
also selling you folks books on "cube-skills"), then
maybe you'll buy it if I succeed in having somebody
else among you make that same argument for me... I
have a feeling that if we pursue this subject, such
a thing is likely to happen pretty quicly...
> Winning the match 25-1 effectively counts as a loss;
> you have not achieved your goal.
Yes.
> To play against JF in this manner you would need to
> set up a match score of 24-0 to JF before beginning play.
Unless I'm missing it completely, this doesn't make
any sense at all in relation to the subject on hand.
> otherwise, JF won't realise what you are trying
> to achieve and wont play the best tactics to try to
> stop you.
The fact is that there is nothing JF (or any other
world-class player) could do about it even if it knew
what my goal was... :) What would it do...? Not take
my immediate double at the beginning of each game and
give me points one after another endlessly...? Not
double when it would be the right cube action for it
to take just because it would know that I'd accept
each and every one of its doubles...? Please try to
explain what JF/SW/the-human-world-champ/etc. could
do and you'll see what I mean.
BTW, I would also be interested in hearing what odds
anybody would offer me with additional conditions such
as requiring that a match should at least last N-games.
So if JF and I together jack up the cube too fast and
JF happens to lose in just 1 game for example, that
wouldn't count but would count if I won in 3 games, or
5 games, etc. I just want to know how low will anybody go
in rating JF's (or any other highly-cube-skilled player's
for that matter) inability to know what to do with that
little device which is useless for anything other than
jacking up the stakes and saving gamblers time anyway...
MK
Douglas Zare
Murat Kalinyaprak wrote:
> Ian Shaw wrote:
> [...]
> > To play against JF in this manner you would need to
> > set up a match score of 24-0 to JF before beginning play.
>
> Unless I'm missing it completely, this doesn't make
> any sense at all in relation to the subject on hand.
>
> > otherwise, JF won't realise what you are trying
> > to achieve and wont play the best tactics to try to
> > stop you.
>
> The fact is that there is nothing JF (or any other
> world-class player) could do about it even if it knew
> what my goal was... :) What would it do...? Not take
> my immediate double at the beginning of each game and
> give me points one after another endlessly...?
That would be dumb.
> Not
> double when it would be the right cube action for it
> to take just because it would know that I'd accept
> each and every one of its doubles...?
That's the one. Suppose you wanted to win 400-0. Get yourself into a bad
position where you have a few percent chance of winning, double and get
beavered, beaver the return, and repeat. This might not work all at
once, depending on how well the match equity tables are set up (whether
they account for the disparity of skill), but you can sacrifice expected
value for variance.
This is why getting a huge bonus if the stocks you pick go up and
getting fired if they go down is a good deal even if you have no idea
what you are doing.
If you set the match score appropriately, your opponent will realize the
cube is dead and it will make life more difficult for you.
This sort of thing makes it difficult to practice spotting people a
handicap without actually doing it; winning most matches by at least
25-20 does not mean you can offer them 5 points.
> [...]
Douglas Zare