PetaMeta

Wednesday, 13 August, Year 6 d.Tr. | Author: Mircea Popescu

This I couldn't possibly pass up. So, you know how Bitcoin hashing was huge a short while ago and then farms doing MEGA!!! hashes showed up but then that didn't last long because gigahashes kicked in and then the network beat a petahash and on and on ? Well, same thing here : we're going to be doing so much meta it'll be PetaMeta. Which is a whole lot of never heard before meta.

To wit : Mike Cirello, who runs one of the few good Bitcoin magazines up and reviewed the work of one Mike Cirello, who made a Bitcoin betting game, in partnership with me, who will be your host tonight reviewing the review. Win yet ? Toldja I couldn't possibly have passed it up.

Says teh author :

There is a famous quote by H. L. Mencken:

No one in this world, so far as I know - and I have researched the records for years, and employed agents to help me - has ever lost money by underestimating the intelligence of the great masses of the plain people.

Fair enough. I have a better quote.

Democracy gives the beatification of mediocrity a certain appearance of objective and demonstrable truth.i The mob man, functioning as citizen, gets a feeling that he is really important to the world—that he is genuinely running things. Out of his maudlin herding after rogues and mountebanks there comes to him a sense of vast and mysterious power—which is what makes archbishops, police sergeants, the grand goblins of the Ku Klux and other such magnificoes happy. And out of it there comes, too, a conviction that he is somehow wise, that his views are taken seriously by his betters—which is what makes United States Senators, fortune tellers and Young Intellectuals happy. Finally, there comes out of it a glowing consciousness of a high duty triumphantly done which is what makes hangmen and husbands happy.

How does it feel for nothing substantial to have changed in roughly speaking a century of avowed, scienfitical "progress" ? But let's not entirely digress, gotta keep the meta in the meta. For I heard you like meta. So therefore :

The "Game of the Week" on War of Life is a parimutuel gambling game where players wager on the ebb and flow of a cellular automaton simulation. It meets the requirements of provably fair, secure, and trustworthy.

Other than the excision of that extra a, this really requires an in-depth write-up, because while I know why and wherefore the statement is exactly correct, nevertheless I may be the only one.

A drawback to parimutuel wagering is that it is very difficult to predict the odds you will get on your bet. If you bet on a horse at the track, or a wager on BitBet, a swarm of money following you could drop your odds below what you were hoping to achieve.

In the general form this is stated, this is not actually true. The proper formulation requires some foundation work : let there be a bet B, let there be implied odds Oi by the money already bet, let there be "real" odds somehow known by the almighty all-knower Or. So with an example : if the bet is for the ideal duck having three legs, and there's a Bitcoin for and ten against, then the Oi would be 90% and the Or would be 100%. Now in this context, whenever you put money down on a bet where Oi != Or, if the money you put fails to enact the equality of Oi to Or you can reasonably expect further bets in sum no less than what would suffice to enact that equality, dampened by how obscure the venue is and other factors to do with the cost of capital (basically, "or as reasonably close as anyone can be arsed to make it").

This property works for you or against you depending on whether you are a liquidity taker or a liquidity provider. Specifically, if your bet makes the Oi to Or spread wider, you can expect later money to narrow the spread. Conversely, if your bet makes the Oi to Or spread narrower, you can expect later money to... also narrow the spread. So in this sense : if you can't afford to close the Oi to Or gap, you have no grounds to complain when someone else does, nor are you entitled to a larger share than them. If the space that was to be bridged is worth twenty Bitcoin, of which you provide the first and others provide the final nineteen, you have no grounds to hold yourself in better esteem than any other of those nineteen, and will roughly get the same they do (minus some time value, as it works on BitBet). Meanwhile if you throw the spread out too far, you also have little grounds to complain : nobody forced you to bring a billion share block to the auction and spook the whole trading floor, take it easier next time.

In short : approaching parimutuel betting in terms of line betting ("the odds you will get") is fundamentally inadequate. Obviously in general you have no way of knowing what Or is, not really any way of knowing what the world thinks the Or will be. This is not really much of your concern, in any case : since you can see the Oi plainly and can compute your very own version of the Or, simply close that gap and that's that. You can't predict how much money you will make in the classical, Newtonian sense of the term, "go to school and you'll get three pies" sort of thing. Nevertheless you can quite well predict how much money you will make in the statistical sense : your capital times the positive risk differential you average (which is to say that if you misjudge Or and align the Oi the wrong way, you'll make a loss).

The Game of the Week functions differently. The teams split the pot - so if money swarms in on one side it makes shares of the other team worth more as well. This gives an economic incentive for what shares are worth to move towards what they cost. Therefore, if you bet on a team when they control 23% of the board, and at the end of the game they control 46%, there is a strong economic interest pushing your shares towards doubling their value.

This is very true, and a very clever use of the fact that the gambler's fallacy is actually a fallacy. The fact that the dice fell eight times on three does exactly nothing for the next throw, and similarly the fact that either side in a random contest prevailed so far does not justify much of a premium. Ironically, this is the case with conflicts in human history as well, if one can be bothered to study that subject matter (fortunately, as a general rule one can not be so bothered - hence Mencken's quote above).

It is impossible to lock in a profit at the track or on BitBet due to the payout structure.

This, again, is not actually true. Because you can always cancel out your BitBet bet, you can logically lock in profit - actual profit - by the same measure. Let's take a trivial example : suppose there's a bet open on whether a coin toss yields sides. For some ungodly reason there's a thousand Bitcoins bet on No right out of the gate and nothing on the Yes. You notice this, and risk your only Bitcent in the world on such a great wager. Somehow this bet does not get noticed by anyone else for a while, and eventually there's a thousand Bitcoin wagered at half weight. Your Bitcent is pretty much guaranteed two profit in case your side should win, which means you actually have, statistically, a 50% return.

What if the practicality of a 50% return implemented as a 50% shot at a 200% return does not actually work to your satisfaction ? Why, bet half a Bitcent on each side! That way, should the bet go unnoticed for a while, until eventually a thousand Bitcoin make it in at half weight, you are in fact guaranteed that whichever side wins, you get your Bitcent back and maybe something extra on the side.ii

Say you bought Death early at 0.5 and Death went on to do well. It now controls 75% of the board and you buy the same amount of Life at 0.25. That's it. As long as share worth stays close to cost (which as previously discussed, is economically incented) then you will end up profitable.

As the table (omitted here) illustrates, the situation is quite similar to the above discussed BitBet : you either make 0.x if there's an adverse wipeout, or 2ish if there's total win.

But all that aside, there is one matter I'm curious about : given that the random bombs are statistical noise and thus negligible, would it be possible for a betting algorithm to be devised that'd exploit the decade period between paid bets, specifically by seeing the start position, calculating the end position (which is deterministic, of course) then taking bets according to the probable evolution on years 0 and 9 of each decade and in so doing syphon a little (or a lot ?) of value out of the pots ? And, if you're less into slow and steady and more into coups, let me reiterate that the quadrupling reward for whoever manages to double their stake is yet to be claimed.

———
  1. This is quite true, and quite importantly so. Nothing wets the panties of everyman mediocrity more these days than this pretense to truthiness, demonstrated, "scienfitical".

    Completely ignoring everything that's substantive and proper to science, the nobody of today follows around unfunny comedians that pompously explain what the "scientific consensus" is : 97.3% of bureaucrats living on the government dole agree that whatever policy that government wishes to implement has sound scienfitical basis. And they're scientists, these people. Why wouldn't they be ? They have the labcoat, right ? A government body gave them a stamped piece of paper (no, not the food stamps, the other one). So that's it, right ? Gerrymandered, ballot stuffed scienfitical science to the rescue.

    Completely ignoring anything that could remotely be construed as science by the people - the individual people - that actually made science matter in the first place (because not so long ago "science" wasn't a +EV conceit for your average politico), Joe Schnook and his online girlfriend Yutzi Klip will tell you all you can take about anything anyone happens to mention, each word of it well backed with scienfitical proofs and verities such as are in common use today : "Internet research" understood as adding together the random numbers idiots published under the strict supervision of selected idiots, "Internet research" understood as "moot is Time's most influential man of the year" according to a buggy, self-selecting online questionnaire and finally "Internet research" understood as Leah Goodman : "here's some random shit I dreamed up, and if you laugh at me well just your saying so doesn't make me stupid". []

  2. This, together with the Qi Qr discussion above, is in fact the rationale behind the F.MPIF's 4th profit center. It's true that it happens to be underwater for now, but I would propose that has a lot more to do with the Q's than with the part discussed here. []
Category: S.WOL
Comments feed : RSS 2.0. Leave your own comment below, or send a trackback.
Add your cents! »
    If this is your first comment, it will wait to be approved. This usually takes a few hours. Subsequent comments are not delayed.