Craps Odds

Having read most of the WizzardOfOdds site everything makes sense to me except for the answer to one of the questions that went something like this: If player A bets only the pass line and player B bets pass line and takes full odds, don't they loose the same amount of money? The answer was 'Yes' but you decrease the house edge by a large amount. So my question is this, if both players loose the same amount of money what is the benefit, if any, in taking odds? If the house edge decreases but you loose the same amount of money what difference does it make? And why is taking full odds universally recommended if it doesn't pay off in your chip stack?

Thanks

Yeah, I've found that somewhat confusing too. I think the mistake is viewing it before you bet and viewing it at the end of the session.

Obviously with this 0.06 percent house edge at 100x odds the casino couldn't pay the electric bill, unless it was very lucky. The casino makes its money on the contract bet's house edge. The odds bet is a sweetener that allows the casino to exhaust the player's bankroll in a faster manner.

If during the short term the player wins, then odds bets have paid off for him. In the longer term, the player has lost more flat bets than he has won and therefore lost more odds bet than he has won also.

It's called gambling. If the house offers 100-1 full odds, the shooter can make $500 bets at actual odds. Basically ignoring the long term results of the $5 bet while trying to get lucky with the bigger bets. Makes sense if you believe in dice setting and can achieve better than expected results. I believe THE WIZ did some sort of trial years ago but not sure of the results. I would not trust any trial
in which a magician was not among the witnesses. Preferably without the shooter's knowledge.

Quote: mpetey3

Having read most of the WizzardOfOdds site everything makes sense to me except for the answer to one of the questions that went something like this: If player A bets only the pass line and player B bets pass line and takes full odds, don't they loose the same amount of money? The answer was 'Yes' but you decrease the house edge by a large amount. So my question is this, if both players loose the same amount of money what is the benefit, if any, in taking odds? If the house edge decreases but you loose the same amount of money what difference does it make? And why is taking full odds universally recommended if it doesn't pay off in your chip stack?Thanks

Good question. However, they are both "EXPECTED" to lose the same amount of money; this is not the same as actually losing it. The confusion is due to the misleading way the house advantage (HA) is calculated by most people for line bets with odds. The odds bet is a different bet from the line bet, so why should you lump the two bets together? You don't lump place bets on six together with place bets on 5 or 10.

pass bet: ev is -0.01414 * bet amount
odds bet: ev is zero

So, when you combine them, the expected value is still the same, but the total bet amount is higher, thereby "lowering" the HA.

So, why take odds? Free variance! Variance is the only thing that gives the individual player a chance to win.

The pass line has low variance, because it's an even-money bet with very close to a .5 win probability (.4929). After 5000 bets, you only have about a 16% chance of being even or ahead. If you want to add variance, you can make place bets or come bets, but all these also add expected loss. By taking or laying odds, you add variance without increasing expected loss. If the player takes 3, 4, 5X odds, then after 5000 bets (s)he still has about a 42% chance to be even or ahead.

Variance takes better advantage of good luck - you don't have to be as lucky to be ahead - but it also increases the damage done by bad luck.

Taking full odds is not "universally recommended". The decision whether to take (lay) odds and in what multiples should take into consideration the player's bankroll, loss tolerance and goal(s).

The only way to reduce expected loss is to lower the amount of your line bet. If you're a $10 pass bettor and cut that to $5, you cut your expected loss in half, but you also decrease your variance. If you bet $5 and take double odds, you still cut your expected loss in half, but you also increase your variance; your average bet would be $11.67.
Cheers,
Alan Shank
Woodland, CA

Alan is 100% correct as always.
Odds gives you a better chance to come out ahead in any number of trials. We call that variance.

Another way to look at pass line betting is if you always bet $25 on the pass. STOP!
Make it $5 with $20 odds, or $10 with $15 odds or $15 with $10 odds.

try to have more of your total wager on the odds part of the bet.
I think Alan has posts or a blog on that also.

I think I understand the difference now. If there are 3 players at a table, player A bets $5 pass line only, player B bets $5 pass line line and $5 odds, and player C bets $10 pass line only. After 5000 hands A and B will have lost the same amount and player C will have lost more. However in the short term B has a greater chance of being up/down than does A.

Is that correct?

Thanks for your time.

Quote: mpetey3

I think I understand the difference now. If there are 3 players at a table, player A bets $5 pass line only, player B bets $5 pass line line and $5 odds, and player C bets $10 pass line only. After 5000 hands A and B will have lost the same amount and player C will have lost more. However in the short term B has a greater chance of being up/down than does A.

Is that correct?

Except that you are still confusing expectation with actual results, at least in your wording. We don't know how any of them WILL do in 5000 decisions, but C has twice the expected loss as A and B. A and C have the same probabilities of being even or ahead, since C has twice the variance as well as twice the expected loss. B has the best chance of being ahead, with more variance than A and less expected loss than C. If they all have bad luck, C will lose the most, then B, then A.

For 5000 bets:


If they are each lucky to the extent of one standard deviation better than expectation (p ~ .16), A and C will be about even, while B will be ahead by about $330. If they are each equally unlucky (-1 SD), C will be down over $1400, B about $1050 and A only about $710.
Cheers,
Alan Shank
Woodland, CA

That makes sense. If given perfect probability over 1 million trials A and B would be equal. Since nothing is perfect B has the best chance to ahead.

Thanks for you time.