Craps system when 5 is the point... please show me my error

Was up late last night thinking about craps and per roll probabilities after the come out roll when specific points are made. Found an interesting scenario for 5's. Didn't see this elsewhere and wanted your thoughts on the math. Also let me know if this is in the wrong section.

Assumptions: 2 pays 2x and 12 pays 3x in Field, 3x/4x/5x max odds, rolls that result in no win/loss are ignored

Strategy 1 (baseline)
Bet $5 Pass, point of 5 is made
*Max odds bet of 4x ($20)
House edge after the point is made: (((4/10)*35)+((6/10)*-25)))/25 = -4%

Strategy 2 (proposed strategy)
Bet $5 Pass, point of 5 is made
*Max odds bet of 4x ($20)
*Place bet on 6 for $12
*Place bet on 8 for $12
*Field bet for $5 (anytime this bet loses on a 6 or 8, replenish with earnings)
House edge after the point is made: (((1/36)*10)+((2/36)*5)+((3/36)*5)+((4/36)*(35-5))+((5/36)*(14-5))+((6/36)*-54)+((5/36)*(14-5))+((4/36)*5)+((3/36)*5)+((2/36)*5)+((1/36)*15))/54 = -0.977%

Was originally thinking of a cheap way to be invested in every roll and was surprised when I did out the math. I know systems are junk, so what's my mathematical error in Strategy 2 that gives a lower house edge?

OK, here is your error.

Your strategy 2 is the same as strategy 1, except with several negative-expectation bets added on. Therefore your strategy 2 must perform worse than strategy 1.

Quote: sodawater

OK, here is your error.

Your strategy 2 is the same as strategy 1, except with several negative-expectation bets added on. Therefore your strategy 2 must perform worse than strategy 1.

So why when I worked out the values did I end up with a lower house edge in Strategy 2? Looking to see if there's an error in the math when I calculated the probability.

Quote: ziggynagy

So why when I worked out the values did I end up with a lower house edge in Strategy 2? Looking to see if there's an error in the math when I calculated the probability.

Because your first calculation is per resolution, and your second calculation is per roll. If you only resolve the bet on a 5 or a 7, the house edge per roll will be less.

That and 5-6-8 lose on the field-bet. Never hedge a (presumed) winner. No faith, no $$$$.

Quote: ChampagneFireball

Because your first calculation is per resolution, and your second calculation is per roll. If you only resolve the bet on a 5 or a 7, the house edge per roll will be less.

This, plus other problems. You are comparing apples to oranges, because the house edge you are measuring is calculated against different values, over different length of times. You should just look at the expectation and be done with it. Expectation in the 1st case: -$1 per resolution. Expectation in the 2nd case: worse.

Quote: ChampagneFireball

Because your first calculation is per resolution, and your second calculation is per roll. If you only resolve the bet on a 5 or a 7, the house edge per roll will be less.

Oh, SNAP!

Quote: ziggynagy

Was up late last night thinking about craps and per roll probabilities after the come out roll when specific points are made. Found an interesting scenario for 5's. Didn't see this elsewhere and wanted your thoughts on the math. Also let me know if this is in the wrong section.

Assumptions: 2 pays 2x and 12 pays 3x in Field, 3x/4x/5x max odds, rolls that result in no win/loss are ignored

Strategy 1 (baseline)
Bet $5 Pass, point of 5 is made
*Max odds bet of 4x ($20)
House edge after the point is made: (((4/10)*35)+((6/10)*-25)))/25 = -4%

Strategy 2 (proposed strategy)
Bet $5 Pass, point of 5 is made
*Max odds bet of 4x ($20)
*Place bet on 6 for $12
*Place bet on 8 for $12
*Field bet for $5 (anytime this bet loses on a 6 or 8, replenish with earnings)
House edge after the point is made: (((1/36)*10)+((2/36)*5)+((3/36)*5)+((4/36)*(35-5))+((5/36)*(14-5))+((6/36)*-54)+((5/36)*(14-5))+((4/36)*5)+((3/36)*5)+((2/36)*5)+((1/36)*15))/54 = -0.977%

Was originally thinking of a cheap way to be invested in every roll and was surprised when I did out the math. I know systems are junk, so what's my mathematical error in Strategy 2 that gives a lower house edge?

You're dividing by too much. If the point is 5, you make your extra bets, and then you roll a 6, you're only resolving the field and place 6 bets. That's not 54 in action, just 11.

To test this, ignore the line and field bets and just repeat your calculations for the place 6 bet alone, then just the place 6 + 8 bets. You know the edge is the same for both of them, but what do your calculations show?

Quote: ziggynagy

Was up late last night thinking about craps and per roll probabilities after the come out roll when specific points are made. Found an interesting scenario for 5's.

cool, thinking about craps and per roll probabilities. Ziggy played guitar

Quote: ziggynagy

Was originally thinking of a cheap way to be invested in every roll and was surprised when I did out the math.

you are a lucky one. I am never surprised by the math any more. Ziggy really sang

Quote: ziggynagy

I know systems are junk...

"One man's junk is another man's treasure"

here be my results

per roll for P5
ev: -$0.277777778
edge: -0.011111111
avg bet: $25

your S2 Iron Cross
ev: -$0.527777778
edge: -0.027377522
avg bet: $19.27777778

Good Luck in your late night thinking sessions