Quote:Mission146When the house edge for the Pass Line is calculated, it is calculated based on all possibilities.

This is true for all bets - not just the Pass Line. However, the definition of "possibilities" can vary (such as "outcome per roll", "outcome per winning or losing decision", etc.) and folks quite often try to compare bets calculated using dissimilar definitions. This leads to erroneous conclusions. If this is what you were intimating then I agree. When comparing different bets, one should always strive to define them using similar terms.

Quote:Mission146Anyway, if you do this for the Pass Line bets and you do it for the Place Bets, then you will come to see that the Pass Line bet has a lower house edge assuming you play any Place Bet to resolution.

This is true but sounds a bit awkward. The Pass Line has a lower house edge than any Place bet whether figured per roll, per decision, or per come-out roll. You seem to imply that the statement is rendered true by virtue of resolving the Place bets and therefore the Pass Line bet might not have a lower house edge if the Place bets were not played to resolution.

Quote:Mission146The difference with Place Bets is some people look at the house edge on a, "Per roll," basis, which is also detailed on the WizardofOdds site. Pass Line bets are not looked at this way because you have no choice but to leave them up until they resolve. Most people make a Place Bet with the plan to leave it up until it resolves, though.

Yes, some people prefer "loss per roll" calculations with Place bets. It's therefore incumbent on them to similarly utilize "loss per roll" when comparing them to other bets.

Nothing prevents anyone from computing the edge of the Pass Line on a per roll basis.

Bet | Loss/dec | Loss/roll | Loss/come-out |
---|---|---|---|

Pass Line | 1.414% | 0.419% | 1.414% |

Don't Pass | 1.403% | 0.404% | 1.364% |

Place 6 | 1.515% | 0.463% | 1.563% |

Another common error I see is folks comparing Don't Pass bets computed on a "per come-out" basis (1.364%) to Place bets computed on a "per decision" basis (e.g. 1.515% for Place6).

Steen

Quote:pwcrabbPlace bets left at risk for only a single toss, then taken down immediately regardless of toss outcome, have very low House Advantages. Those H.A. are detailed by the Wizard in his Per Roll analyses.

I think you're misunderstanding something here. Each bet has a house advantage which is determined by the probabilities of winning and losing and the amounts won or lost. Changing the way that you describe the edge does not change the inherent edge. For example, a Place 6 bet has an advantage that can be described as 1.52% per winning or losing decision. The fact that you can describe the same edge as 0.463% per roll doesn't change the inherent probabilities of winning or losing or the payoffs.

Suppose that I observe a particular table for a number of hours and determine that on average the dice are thrown off the table once every 37 rolls. Could I re-define the Place 6 bet edge to be 0.45% per attempted roll? Sure I could! Would it change the inherent edge defined by the house rules? No.

Quote:pwcrabbPlace bets left at risk only until resolution, then taken down immediately if they have won, have House Advantages that are greater than the H.A. for Pass and Don't Pass.

The house edge doesn't change depending on how many rolls you let bets ride!

Steen

Quote:SteenThis is true for all bets - not just the Pass Line. However, the definition of "possibilities" can vary (such as "outcome per roll", "outcome per winning or losing decision", etc.) and folks quite often try to compare bets calculated using dissimilar definitions. This leads to erroneous conclusions. If this is what you were intimating then I agree. When comparing different bets, one should always strive to define them using similar terms.

I know that it is, the OP specifically asked if the Pass Line accounted for all possibilities is the only reason why I would even say that.

To the rest, that's exactly what I was getting at. Lots of Craps system players will define the House Edge erroneously by making a complicated series of bets while using terms like, "Expected loss," or, "House Edge," when they are referring to the house edge per roll rather than per resolved bet. I actually wish House Edge per roll was not a metric because the entire concept muddies the waters, I would just mention that you can pull your bet back (unresolved) and leave it at that. Naturally the whole thing can be pulled back if it wins, and most people are going to let the Place Bets resolve at least once.

Usually, they press it or leave the original out there when it wins, which is effectively just making a whole new Place Bet. That's another area where these snakeoil system salesmen will get you. They'll say stuff like, "Your winning potential, for only a $6 bet..." You're not making a $6 bet, though. You're making a $6 bet, then a $12 bet and so on...

Quote:This is true but sounds a bit awkward. The Pass Line has a lower house edge than any Place bet whether figured per roll, per decision, or per come-out roll. You seem to imply that the statement is rendered true by virtue of resolving the Place bets and therefore the Pass Line bet might not have a lower house edge if the Place bets were not played to resolution.

I left some room for technicality on that one because I like to be precise. If you were only going to make one craps bet your entire life, win, lose or unresolved, then the Place Bet would be a better bet than the Pass Line (because of House Edge per roll and total expected $$$ loss as a result) due to the fact that the Pass Line bet must resolve. This also assumes same or close to same dollar amount on both. So, if you were going to Place 6 for one roll and one roll only and never play Craps again as long as you live, it would be a better bet than the Pass Line.

Quote:Yes, some people prefer "loss per roll" calculations with Place bets. It's therefore incumbent on them to similarly utilize "loss per roll" when comparing them to other bets.

Nothing prevents anyone from computing the edge of the Pass Line on a per roll basis.

Bet Loss/dec Loss/roll Loss/come-out Pass Line 1.414% 0.419% 1.414% Don't Pass 1.403% 0.404% 1.364% Place 6 1.515% 0.463% 1.563%

Another common error I see is folks comparing Don't Pass bets computed on a "per come-out" basis (1.364%) to Place bets computed on a "per decision" basis (e.g. 1.515% for Place6).

Steen

Nicely done! I think Wizard has average number of rolls per Pass Line bet somewhere, but I didn't feel like going to look for it.

You should feel the same way about someone playing PL with no odds as you do about someone playing three quarters in VP for a 97% RTP vs if he put in five quarters a pull he’d have 99% RTP.

Quote:unJonI disagree with the sentiment that the odds bet in craps is an independent bet from the PL (or DP) in craps. I find it useful to use a video poker analogy. Playing the PL is like short playing VP and adding odds in full is like putting in max quarters to access the better pay table.

You should feel the same way about someone playing PL with no odds as you do about someone playing three quarters in VP for a 97% RTP vs if he put in five quarters a pull he’d have 99% RTP.

You're looking at it from a house edge and expected loss relative to your total action standpoint with the Craps, then, which is fine. It's not an invalid way of looking at it. I guess my question would be, on a table with 100x odds (and there are a few left) would you be there playing a $5 PL and $500 in odds? Most people I have seen play Craps, $500 is something more akin to their entire session buy-in, not what they plop down on one bet.

I disagree with the comparison to Video Poker because you can actually increase the expected loss on video poker by playing that way. It changes the house edge of the game whereas the house edge of the Pass Line bet itself is unchanged. At 3%, you lose $0.0225 per $0.75 whereas you lose $0.0125 per $1.25 bet. With that said, betting a single coin undeniably yields a lower expected loss per hand and expected loss per hour whilst increasing the house edge.

The Pass Line bet is just the Pass Line bet.

Anyway, I totally get where you're coming from with the effective house edge (or whatever you want to call it) of PL + Odds assuming that you are going to make a maximum Odds bet every time you have the opportunity. I again think that's a perfectly valid way to look at it (just not the way I look at it), but I don't think it's apples-for-apples with VP.

legally it is separate. You don't gotta make it and if you do make you can take your money back anytime.

the line bet is a contract bet and I wouldn't try to take your money back when there is already someone holding a stick.

originally it was a way for casinos to compete with each other, you know: free peanuts, free booze, free odds.

Quote:Mission146You're looking at it from a house edge and expected loss relative to your total action standpoint with the Craps, then, which is fine. It's not an invalid way of looking at it. I guess my question would be, on a table with 100x odds (and there are a few left) would you be there playing a $5 PL and $500 in odds? Most people I have seen play Craps, $500 is something more akin to their entire session buy-in, not what they plop down on one bet.

I disagree with the comparison to Video Poker because you can actually increase the expected loss on video poker by playing that way. It changes the house edge of the game whereas the house edge of the Pass Line bet itself is unchanged. At 3%, you lose $0.0225 per $0.75 whereas you lose $0.0125 per $1.25 bet. With that said, betting a single coin undeniably yields a lower expected loss per hand and expected loss per hour whilst increasing the house edge.

The Pass Line bet is just the Pass Line bet.

Anyway, I totally get where you're coming from with the effective house edge (or whatever you want to call it) of PL + Odds assuming that you are going to make a maximum Odds bet every time you have the opportunity. I again think that's a perfectly valid way to look at it (just not the way I look at it), but I don't think it's apples-for-apples with VP.

I hear that. And defer to you on VP as I don’t play it so see that it’s not a perfect analogy.

I have no problem with someone taking less than full odds so long as they aren’t putting money at risk elsewhere during the roll. I have an issue with mainly to types of betting:

1) Someone at the $25 craps table betting PL with no odds while there are spots at the $15 table next to it.

2) Someone not taking full odds but making place bets. I get that the person wants to hedge so he can win on more numbers but it gets to me.

Quote:Mission146I left some room for technicality on that one because I like to be precise. If you were only going to make one craps bet your entire life, win, lose or unresolved, then the Place Bet would be a better bet than the Pass Line (because of House Edge per roll and total expected $$$ loss as a result) due to the fact that the Pass Line bet must resolve. This also assumes same or close to same dollar amount on both. So, if you were going to Place 6 for one roll and one roll only and never play Craps again as long as you live, it would be a better bet than the Pass Line.

I agree but it's an awkward and misleading comparison. You're comparing one Place bet versus one Pass Line bet but restricting the Place bet to just one roll while waiting for the Pass Line to resolve which takes on average 3.375 rolls. I would maintain that comparing the loss per roll of a Place bet with the loss per decision of a Pass Line bet is unfair.

Ordinarily, it doesn't matter whether you pull the Place bet in one roll or not because on average it would lose more per roll (and more per decision) than the Pass Line bet. However, because you're making just one bet for your entire life then the expected value of the bet is irrelevant. In this case, the only thing that matters is the probability of winning or pushing in one roll. In other words, the probability of walking away with money (regardless of the amount) after one roll.

To make this a fair comparison we can either allow both bets to resolve (average 3.27 rolls for Place and 3.375 rolls for Pass Line) or we can restrict both bets to just one roll.

In the first case, the Pass Line has a higher probability of returning money:

Place 6

-- probability of winning = 5/11 = 45.45%

Pass Line (if bet on come-out roll)

-- probability of winning = 244/495 = 49.29%

In the second case, the Place bet has a higher probability of returning money because if the Pass Line doesn't win in one roll then it must be abandoned as a loss (in other words you can't walk away with it).

Place6

-- probability of winning = 5/36

-- probability of pushing = 25/36

-- probability of walking with money after one roll = 30/36 = 83.33%

Pass Line (if bet on come-out roll)

-- probability of winning = 8/36

-- probability of pushing = 0

-- probability of walking with money after one roll = 8/36 = 22.22%

So you have almost a 4x better chance of walking with money in one roll on the Place6 bet. But this is soley due to non-resolving pushes being included in the equation. If you were to ask what's the chance of just winning either bet in one roll, then the Pass Line would be better (8/36 Pass Line versus 5/36 Place). In my opinion, including pushes in craps confuses the issue and can lead to erroneous conclusions.

If pushes matter then what about the player who observes the action but never bets? His chance of walking with money is 100%. A craps genius!

Personally, I've never met anyone who wanted to make just one lifetime bet for just one roll, so the issue is somewhat moot.

Steen

Expected Return for a $6 wager on Place 6 after one projected roll:

( 5 / 36 ) ( $7 ) + ( 30 / 36 ) ( $6 ) = ( $5.9722 )

Expected Loss = ( $0.0278 )

House Advantage = ( 0.4630 % )

Expect Return for a $6 wager on Place 6 after two projected rolls:

( 5 / 36 ) ( $7 ) + ( 30 / 36 ) [ ( 5 / 36 ) ( $7 ) + ( 30 / 36 ) ( $6 ) ] = ( $5.9491 )

Expected Loss = ( $0.0509 )

House Advantage = ( 0.8488 % )

The number of projected rolls is chosen at leisure. For most players of Place bets, that number is infinity

Just for grins, extending the equation above, after how many projected rolls does the H.A. closely approximate the standard presentation, which implicitly assumes exposure until resolution followed by wager removal?

The Standard Presentation is ( 5 / 11 ) ( $7 + 6 ) = ( $5.9091 )

Expected Loss = ( $0.0909 )

House Advantage = ( 1.5152 % )

Quote:pwcrabbSteen suggests that "The house edge doesn't change depending on how many rolls you let bets ride!" However, Steen himself offers a calculation of per-roll loss for a Place bet on 6. Exposure for multiple rolls would seem to have consequences for the house edge, and of course it does so. His suggestion is patently silly. The theoretical probabilities of course never change, but duration of exposure can change and does so at player discretion, with attendant implications for house edge.

Expected Return for a $6 wager on Place 6 after one projected roll:

( 5 / 36 ) ( $7 ) + ( 30 / 36 ) ( $6 ) = ( $5.9722 )

Expected Loss = ( $0.0278 )

House Advantage = ( 0.4630 % )

Expect Return for a $6 wager on Place 6 after two projected rolls:

( 5 / 36 ) ( $7 ) + ( 30 / 36 ) [ ( 5 / 36 ) ( $7 ) + ( 30 / 36 ) ( $6 ) ] = ( $5.9491 )

Expected Loss = ( $0.0509 )

House Advantage = ( 0.8488 % )

The number of projected rolls is chosen at leisure. For most players of Place bets, that number is infinity

Just for grins, extending the equation above, after how many projected rolls does the H.A. closely approximate the standard presentation, which implicitly assumes exposure until resolution followed by wager removal?

The Standard Presentation is ( 5 / 11 ) ( $7 + 6 ) = ( $5.9091 )

Expected Loss = ( $0.0909 )

House Advantage = ( 1.5152 % )

Unfortunately, I have to catch a plane right now so I can't give you a full answer until tomorrow. In the meantime, I'll let you re-think your nonsense here and post a correction.

I'll give you a couple hints: your calculations and your definitions are off.

:)

Steen