Clarification on Murat Ali vs. bots bet.

[mu...@compuplus.net]

michae...@gmail.com wrote:

> mu...@compuplus.net wrote:

>> Why would I do that? When I feel I should instead eliminate
>> the first part, which I offered as a step to meet half way.

> Because by playing just one reference match you could play
> junk to get a very low reference PR. The correct way to do
> it is to play a series of matches and THEN compute your
> average ER. And AFTER THAT check if your total wins agree
> with the theoretical ones.

Michael... must be new in RGB, because the details of my
bet based on ER has been discussed in detail many times
over.

Let me clarify and illustrate one more time for newcomers
and seniles of weak menory, at al...

I will play a "long series" (i.e. 100!) "long matches" (i.e.
25 pointers), against the strongest settings of a bot (i.e.
the latest pre-release version of XGR).

In the past, I had estimated that I could accomlish this in
about three months or so.

I will bet $5,000 on even odds for those 100 matches (i.e.
$50 per match against the bot backers' $50 per match).

I will bet $5,000 on odds to be determined after the fact,
based on my winning chances derived from my ER from those
100 matches, against a proportional amount by the backers'
of the bot.

EXAMPLE 1: I win 40 of 100 matches. Bot is 80%-20% favorite.

For the 50x50 portion of the bet. I lose 60x$50=$3,000.
Bot loses 40x$50=$2,000. Bot's net gain=$1,000.

For the ER adjusted portion of the bet. I lose 60x$50=$3,000.
Bot loses 40x$200=$8,000. My net gain=$5,000.

Overall result: I win $5,000-$1,000=$4,000.

EXAMPLE 2: I win 60 of 100 matches. Bot is 80%-20% favorite.

For the 50x50 portion of the bet. I lose 40x$50=$2,000.
Bot loses 60x$50=$3,000. My net gain=$1,000.

For the ER adjusted portion of the bet. I lose 40x$50=$2,000.
Bot loses 60x$200=$12,000. My net gain=$10,000.

Overall result: I win $1,000+$10,000=$11,000.

-------------------------------

This is not only a more than fair bet but an excellent
opportunity for all those "BG gorillas" to prove that
their 2.5 ER is really meaningful...

They constantly brag about how big sums they bet on even
a single match/session. So, a few thousand bucks should
be a bag of peanuts for the chimps... :))

The only thing that worries me, is that some of them may
get suicidal after such a bet and I may feel guilty for
the rest of my life... :(

MK

[mu...@compuplus.net]

mu...@compuplus.net wrote:

Oops, I forgot the last example:

EXAMPLE 3: I win 20 of 100 matches. Bot is 80%-20% favorite.

For the 50x50 portion of the bet. I lose 80x$50=$4,000.
Bot loses 20x$50=$1,000. Bot's net gain=$3,000.

For the ER adjusted portion of the bet. I lose 80x$50=$4,000.
Bot loses 20x$200=$4,000. As expected, the result is "zero sum" :)

Hah hahh... I love that term "zero sum". It's sooo math PHD... ;)

Overall result: Bot wins $3,000-$0=$3,000.

So, if the ER bullshit is what is cracked up to be, bot backers
should make an easy $3,000 without lifting a finger...

How can you "professional+scientific gamblers" miss out on such
an opportunity...? :)

MK