If I get the "jist" of what you're saying, hardin, it's that the "contenginist" (spelling?) player will press his bets when he is on a winning streak and will decrease his bets when he is on a losing streak. Also, after having several losses, he will press his bet to "make up" for those previous losses because he will win (?) those future bets.

Unfortunately, that's not quite how it works. First of all, one cannot "hope" for positive variance to hit them at certain times. Well, kind of, kind of not. A player can (and should) always hope for positive variance (duh) -- BUT, hoping for, or "relying" on positive variance at certain SPECIFIC times, is not a fundamental way of actually winning. While the 'voodoo' player is hoping and relying on variance -- the counter relies on his edge [while also hoping for variance....but hoping for variance is not the point in counting].

You simply can't "bet with the streak" as the previous hands don't effect future hands. In order to "bet with a streak", there must be some sort of knowledge, intelligence, connection, or "relation" between the cards. And....there isn't one.

How many times have you seen a player win several hands in a row with a flat-bet...then he jumps his bet way up, loses, and people are like, "Wow, you're so greedy for jumping your bet up like that!" while other times, the player will continue to flat-bet and will continue winning, and you'll hear people say, "You should increase your bet, you're on a roll!"

You only know of the streak afterwards -- you don't know if you're going to keep winning hand after hand after hand....or if you're going to start losing every hand. Hindsight. You can't bet into a situation where you don't really know what's going to happen, hoping that the things that just happened in the past will continue.

As I was reading your post, I saw a few times where you mentioned things in a fact-like way, like "The voodoo player loses 5 hands of 1 unit, then wins 4 hands of 2 units" [or something like that]. I don't think you were trying to state them as actual facts, but you were simply giving an example (which is fine). But then took what was (falsely) created from the example and continued with the argument(?) with that premise.


No matter how you cut it, when you make a bet with a disadvantage, in the long run, you will lose that certain % of money.



Think of this -- you have a "the price is right" wheel, with 1,000 numbers on it. 499 of the numbers are "winning" numbers, in which case you get paid even money (1:1). The other 501 numbers are "losing" numbers, in which case you lose your bet. In this game, 499/1000 are winners, and 501/1000 are losers. In the long run, we expect an even distribution of each number (not even distribution of wins and losses, but even distribution of each number!) -- we can calculate that the EV for this game is:

+499/1000 -501/1000 = -2/1000 = -0.002 = -0.2%.

The game also has very low variance, since you'll be winning/losing even money every time and you win/lose approximately 50%.

A high-variance game would be one where there are 1,000 numbers, and 999 of them are losing numbers, but 1 of them is a winning number and pays out 997:1. The EV for this game is the same:

+(1/1000)*997 - (999/1000)*1 = -0.002 = -0.2%.


Or, you can create a similar game with different pay-tables (5 numbers pay 198:1, 995 numbers lose), but this has a -0.5% edge. This game has much higher variance than game one, but not quite as much as variance as game 2 [although still has quite a bit of variance].


Regardless, I ask this.

If you can (or you have) come up with a betting system that will win in blackjack, where you are betting into a disadvantage, I would find it reasonable that a similar system could be created to beat this "price is right" game. Actually, I'll throw in one more added bonus, making it more relatable to blackjack -- the amount of winning numbers isn't "fixed", meaning that (say for game #1), it isn't always 499 winners and 501 losers, but there is a chance to have 501 or 502 winners with 499 or 498 losers. Something like this:

95%: 499 winners, 501 losers
5%: 502 winners, 498 losers
Edge: ((499/1000) - (501/1000))*0.95 + ((502/1000) - (499/1000))*0.05 = -0.00175



What kind of betting system [similar to a BJ system] would you use for a 'the price is right' game? If you want to change around the edge, the factor of the payouts, the % of edge change, etc etc., by all means.