Re: Robertie - 501 Essential BG Problems - Problem #75 (errata)

Dear Mr Raccoon,

you wrote a long letter to this newsgroup, full of facts, most of them true. But
i still try to keep this discussion as simple as possible. I will only answer to
one fact within your contribution where i believe that you drew a wrong
conclusion.

On 9 Nov 2005 08:32:06 -0800, "Raccoon" <racgammon@xxxxxxxxx> wrote:

>Pete Lederer wrote:
>> The question is a different one. How would a backgammon site gain from
>> manipulating dice? Imagine there are some very good players (sharks) and some
>> very bad ones (fish). The sharks win more often and thus accumulate the money.
>> In fact, they money will be withdrawn soon and is no more available to create
>> rakes. Thus a backgammon site will profit from players who are all of the same
>> strength (biased luck can compensate for different grades of skill).
>
>Let me propose an alternative (long but information dense, let's hope)
>view.
>
>1. A backgammon site is populated with players whose skill ranges from
>very bad to very good. Most players are neither very bad nor very good.
>
>2. Money players are not equally motivated by profit nor equally averse
>to loss. Some are 100% motivated by profit, some are 100% motivated by
>the entertainment value of their wager, and most players are motivated
>by some combination of desire for profit and entertainment.
>
>3. A backgammon site populated by players with wide ranging skill and
>disparate motivations will profit by enabling bad, average and good
>players to find their own profit/entertainment equilibrium through
>their individual selections of stake and opponent.
>
>Excellent players highly motivated by profit can choose to play for
>high stakes. Less skilled players but similarly motivated players who
>are outclassed at higher stakes may nevertheless find profitable
>opposition at lesser stakes. Poor players may through opponent and
>stake selection avoid losing more than their entertainment budget
>allows, whether that means losing at small stakes or large ones.
>
>Players "find their own level." Manipulation of dice is neither more
>necessary nor desirable online than it is at a live backgammon
>tournament with Beginner, Intermediate and Open flights for small,
>medium, and large entry fees. Possibly, most players are happy enough
>with approximately even or somewhat better opposition while
>"professional" players naturally are happiest with worse opponents at,
>for their skill level, the most profitable stakes.
>
>4. All players with the slightest interest in evaluating their own play
>and in comparing results with their expected return can do so with
>whatever degree of accuracy and quantity of evidence they find
>sufficient.
>
>If as some allege a server is cheating _all_ of its customers by
>rigging the dice according to skill, surely one -- at least one! -- of
>the thousands of online players would have the evidence to prove it.
>And yet not a single one does (which does not mean it is impossible
>that a server _could_ cheat _all_ of its customers or -- as would be
>much harder to detect and publish proof of -- could cheat _some_ of its
>customers).
>
>How accurately bot analysis correlates with real life results is an
>open question. Surely with fair dice some will win more often and some
>will lose more often than their bot analysis predicts. However, if
>_everyone_ were being cheated, then wouldn't _everyone's_ results be
>skewed?
>
>> In fact, the money will be withdrawn soon and is no more available to create
>> rakes.
>
>Really? Common sense tells us that having made a profit, winners are
>motivated to continue playing. Reality tells us that on TMG or
>GamesGrid or GammonEmpire the same winning players can be found day
>after day playing with whoever does not mind losing to them more often
>than not. Meanwhile, money pocketed by winners is not disappearing from
>a finite amount of money that losers bring to the table. The amount
>available for extraction by winners and the house is replenished by (1)
>rebuys by the existing pool of losers and (2) deposits by new
>customers.

While winners are motivated to continue playing they usually only keep a base
amount of money on the server. The money that they withdraw is usually lost for
the server's owners.

You tell of money being taken away by winners that will immediately replenished
by new players or rebuys. Of course, that is not true. If the income of the
losers was unlimited and their willingness to bring the money to a backgammon
server you were right.

>Think Vegas. Local pros, local grinders, local losers, and a
>continuously replenishing, revolving pool of tourists who keep coming
>back regardless of results.

You really compare a small site like TMG who has about 1000 players
simultaneously in peek times with Vegas? Not to forget that most of these
players don't deposit much if at all. Lol, be more serious, please.

>> Did you think about the fact that Gammonempire rakes the higher rated player in
>> matches between players of different skills very high? They try to avoid the
>> effect i described above.
>
>Even if that is GE's motivation, does it work? The much lesser ranked
>player pays double rake regardless of who wins. The higher ranked
>player pays double rake (or worse) only if he wins. His win rate
>against lesser competition may be high enough to show a profit even at
>the higher rake (which, let's note, might even be lower than the usual
>rake at a competing server).
>
>Furthermore, GE's 1700-rating requirement for half-rake still leaves
>plenty of room for uneven competition among the over-1700s -- we're
>talking 1700 to 2000+. And for the under 1700's, double-plus rake
>applies only when the ratings difference is over 100. If the player
>pool is large, there should be a large number of opponents with
>disparate skill within any given 100 point range.
>
> For those reasons I hypothesize that the effect of GE's multi-tiered
>rake structure on opponent selection is not great, and regardless of
>its effect is motivated not so much by a desire to protect fish from
>sharks, but because (1) among money players skill correlates positively
>with sensitivity to profit-reducing rake and (2) skill correlates
>positively with stake.
>
>(1) means that profit-seeking players will take their action to
>whichever server is most profitable to them, and rake is one
>consideration (opponent pool is the other), while break-even players,
>losing players, and "cheap entertainment" players are either oblivious
>to the rake or lack the skill or bankroll necessary to gain rating and
>play higher stakes and thus pay lower rake.
>
>(2) means that a server can make as much if not more off of highly
>skilled players paying low rake at high stakes than off low skilled
>players paying high rake at low stakes. In this the GammonEmpire rake
>structure is similar to that of any poker server; experts and whales
>pay more reasonable rake at higher stakes, while the less skilled, the
>less affluent, the riffraff and the low-rent grinders pay higher rake
>for the dubious privilege of having a place to play at all.
.

Relevant Pages

  • Re: What is $81,405?
    ... rake. ... PS a very nice profit. ... rakes more reasonable and keep more money in players pockets. ... If you mean there is a lack of choice, ie., all the sites charge the same ...
    (rec.gambling.poker)
  • Re: Robertie - 501 Essential BG Problems - Problem #75 (errata)
    ... Thus a backgammon site will profit from players who are all of the same ... are outclassed at higher stakes may nevertheless find profitable ... player pays double rake only if he wins. ...
    (rec.games.backgammon)
  • Re: Why Is Everyone So Fixated on The Rake??
    ... why so mnay players lose. ... When I first started playing poker I was aware of the rake, ... did speculate that it would lessen my profit potential. ... $60 pots, but you give away plenty of pots that size too - the money is ...
    (rec.gambling.poker)
  • Re: If youve never really thought about the rake
    ... game and the rake and jackpot drop are $4 a hand. ... If players tip the dealer, the amount of money leaving the ... most low limit poker players don't even think about the rake. ...
    (rec.gambling.poker)
  • Re: If youve never really thought about the rake
    ... game and the rake and jackpot drop are $4 a hand. ... If players tip the dealer, the amount of money leaving the ... most low limit poker players don't even think about the rake. ...
    (rec.gambling.poker)