this is a simulation on back-betting lousy players. but why should you care about this simulation if you don't even back-bet yourself? well, i want to know if a certain kind of conservative, lousy player can bring in a profit for the back-counting back-bettor. non-back-bettors may be interested in my results to see just how bad a player has to be to lose when the count is high (assuming the player correctly varies his bets with the count). in other words, how much can bet-variation overcome lousy play?
the impetus behind this simulation comes from Wong's pp.45 and 50 of PB, where he notes that "Playing hands according to basic strategy and using the hi-lo for bet variation wins at the rate of $12 per hour. So BET VARIATION ALONE [my emphasis] improves on flat-betting by $27 per hour, or 87% of the $31 gain from both bet variation and strategy variation."
ok, i'm sure you all knew that already. but what i want to find out with my simulation is whether a lousy player who also deviates from basic strategy can maintain an advantage by 1) only betting in hi-lo TCs of +2 or higher, 2) using a betting variation corresponding to the TC, 3) correctly taking insurance.
So my simulation compares three players:
1) a hi-lo counter using indices up to +7,
2) a basic strategy player (ENHC basic strategy),
3) a lousy player (described below).
each player uses the following betting scheme at respective positive TCs of 2, 3, 4, 5, 6, 7...: $20, $40, $60, $100, $150, $150....(yes, there's not heat at this casino).
the three players essentially wong into the game at +2 TC or higher. they wong into a full table--so whenever the count is +2 or higher, 7 players are at the table. (this is supposed to simulate a counter back-betting three different players at a full table.)
all players correctly take insurance at TC of +3. in european casinos, the back-bettor has the option of taking insurance regardless of what the sitting-player chooses to do.
i could not simulate a feature of the european game that benefits the back-bettor. when the sitting player incorrectly splits 22 against 2 in a no DAS game, the backbettor has the option of adding more money on the table, or he can split his original wager in half. this will be of minor benefit to the back-bettor since this casino does not offer DAS. but it is an advantage nonetheless.
House Rules:
1) ENHC
2) 5 decks, 3.5 penetration (i get better than that at my casino--but i want to simulate a kind of worse-case scenario)
3) no DAS, no resplits, one card to split aces
4) no soft doubling (ace doesn't swing if you soft double)
5) only double on 9, 10, 11
6) normal insurance, normal BJ pay off
ok---what you've all been waiting for...the LOUSY PLAYER: how lousy is he? well, he's a she. she's the best non-counter at the casino i play at. she's not nearly as bad as the lousy player i programmed into the simulation, but her strategy served as a model for the lousy player i programmed. here are the following deviations from ENHC basic strategy that this lousy player consistently makes:
1) ALWAYS stands on 16 and 15 against any dealer up card.
2) stands on 12, 3 and 12, 2 (of course, both are correct moves at high TCs)
3) does not split 88 against 9 (stands)
4) does not split 99 against 9
5) does not hit A7 v 9/10/Ace
6) splits 33 and 22 v 2 and AA v A (this will hurt the lousy player in my simulation, but will help the back-bettor)
7) does not double 11 v 9
8) does not double 10 v 9
I hope this player is sufficiently lousy for you to fear back-betting her. nonetheless, backbetting this player will bring you a profit. and if you had to choose between back-betting this player or letting a high count pass you by because you can't get a seat, then you would be losing money by not back-betting this player.
all players start with $1000 bankrolls.
here are the results after 7,429,386 rounds played.
dollar winnings percent advantage
1) card-counter $7,142,665 1.33%
2) basic strategy $5,837,810 1.11%
3) lousy player $5,009,010 .95%
basic strategy player wins 82 % as much as a counter.
the lousy player wins 70% as much as a counter.
i should note that the expectation would be higher if one did not back-bet the basic strategy player nor the lousy player at TC of +2. according to my simulation, at +2 both the BS and the lousy player had a slight negative expectation. in my own experiences, i only backbet other players at TC of +3 or higher, mainly because the table is always full and a TC of +2 just doesn't seem like enough. but the counter (in the simulation) had a slight positive expectation at TC of +2.
although this simulation is useless for most of you, i hope you can appreciate the power of bet variation. it can even overcome lousy playing when the TC is high enough.
by the way, i used Sage Blackjack to get these results. i let my Pentium III 750 computer run all day long using simulation mode. i invite criticism and comments on my simulation.
the cheeze
