petroglyph
Joined: Jan 3, 2013
• Posts: 3360
April 12th, 2016 at 10:13:04 AM permalink
Quote: mustangsally

this does not at all result in average bets (and total action) being equal.

the no odds has an average of \$50

\$10 with \$40 looks like \$50
looks like \$50

buts only 2/3 of the time IS \$50
the other 1/3 = \$10
1/3*10 + 2/3* 50 = 10/3+100/3 = 110/3 = \$36.67

I say \$10 with \$60 odds has the same average bet as a \$50 flat
or even
\$30 flat with \$30 odds = \$50 average bet (even)

yes/no
no/yes
Sally

of course
the expected loss from the average \$50 bet is different
with different flat bets - as expected.

deleted
Last edited by: petroglyph on Apr 12, 2016
odiousgambit
Joined: Nov 9, 2009
• Posts: 8798
April 12th, 2016 at 10:36:20 AM permalink
Quote: MathExtremist

Sure, but once you've bet the \$50, taking odds doesn't change your expected loss at all

sure, and I for one do not say to anyone to think so

Quote:

... The focus on house edge as a percentage is misplaced in craps when it comes to the odds bets because that house edge percentage is reduced not by decreasing the numerator (the expected loss), but by increasing the denominator (the total wager). It's a qualitatively different issue than, say, doubling down vs. hitting on 11 vs. dealer 6.

I don't see how this foils the suggestion to the \$50 line bettor who never takes odds to instead make the \$10 + \$40 odds bets instead. Yes, he is likely to misunderstand and bet \$50 + \$200 in odds, see it all the time. But some people are too dumb to help.

And I don't know why Steen pounced on me instead of you LOL.
the next time Dame Fortune toys with your heart, your soul and your wallet, raise your glass and praise her thus: “Thanks for nothing, you cold-hearted, evil, damnable, nefarious, low-life, malicious monster from Hell!” She is, after all, stone deaf. ... Arnold Snyder
odiousgambit
Joined: Nov 9, 2009
• Posts: 8798
April 12th, 2016 at 10:37:24 AM permalink
Quote: mustangsally

this does not at all result in average bets (and total action) being equal.

the no odds has an average of \$50

\$10 with \$40 looks like \$50
looks like \$50

buts only 2/3 of the time IS \$50

that's a fact but is splitting hairs as an argument.

PS: I say you are helping that player even more because of your point. Can you see why?
the next time Dame Fortune toys with your heart, your soul and your wallet, raise your glass and praise her thus: “Thanks for nothing, you cold-hearted, evil, damnable, nefarious, low-life, malicious monster from Hell!” She is, after all, stone deaf. ... Arnold Snyder
MrGoldenSun
Joined: Apr 1, 2016
• Posts: 252
April 12th, 2016 at 11:27:43 AM permalink
Quote: odiousgambit

Can you see why?

Oh, failed to stick the landing. Should have gone with "do you see why?" for the Sklansky finish.
MathExtremist
Joined: Aug 31, 2010
• Posts: 6526
April 12th, 2016 at 1:32:14 PM permalink
Quote: odiousgambit

I don't see how this foils the suggestion to the \$50 line bettor who never takes odds to instead make the \$10 + \$40 odds bets instead.

It doesn't foil the suggestion per se, but those are two very different bets. \$50 on the line wins even money 49.3% of the time and loses the rest. \$10 on the line with the intention of betting \$40 odds has a totally different outcome distribution. Plot it out as an exercise -- how often do you win \$10, lose \$10, lose \$50, win \$90, etc. The fact that one has a lower expected loss is almost irrelevant. It's like comparing Pai Gow Poker to a 6-spot keno ticket.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice." -- Girolamo Cardano, 1563
odiousgambit
Joined: Nov 9, 2009
• Posts: 8798
April 12th, 2016 at 2:04:14 PM permalink
Quote: MathExtremist

Quote: odiousgambit

I don't see how this foils the suggestion to the \$50 line bettor who never takes odds to instead make the \$10 + \$40 odds bets instead.

It doesn't foil the suggestion per se, but those are two very different bets. \$50 on the line wins even money 49.3% of the time and loses the rest. \$10 on the line with the intention of betting \$40 odds has a totally different outcome distribution. Plot it out as an exercise -- how often do you win \$10, lose \$10, lose \$50, win \$90, etc. The fact that one has a lower expected loss is almost irrelevant. It's like comparing Pai Gow Poker to a 6-spot keno ticket.

well, I'll give you "very different bets" - in fact I will say that one of the dumbest things about the line bet is the mind-numbing low variance - the SD being around "one".

Let me strike out what I said upthread,

the only thing indisputable is that the guy who puts \$50 on the pass line and never adds odds could lower his EV and HE both by putting \$10 on the line and \$40 on free odds when it comes to EV and HE in Craps, there is absolutely nothing that is indisputable
the next time Dame Fortune toys with your heart, your soul and your wallet, raise your glass and praise her thus: “Thanks for nothing, you cold-hearted, evil, damnable, nefarious, low-life, malicious monster from Hell!” She is, after all, stone deaf. ... Arnold Snyder
MathExtremist
Joined: Aug 31, 2010
• Posts: 6526
April 12th, 2016 at 2:43:20 PM permalink
Quote: odiousgambit

well, I'll give you "very different bets" - in fact I will say that one of the dumbest things about the line bet is the mind-numbing low variance - the SD being around "one".

Yet baccarat is far and away the biggest money maker in the table game pit, especially on a per-table basis. Nothing else comes close, and the SD of the banker bet is even smaller than the craps passline. Go figure.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice." -- Girolamo Cardano, 1563
Steen
Joined: Apr 7, 2014
• Posts: 126
April 12th, 2016 at 3:19:18 PM permalink
Quote: RS

Someone else said it before me -- but it's true: Saying increasing (or working) your odds decreases the house edge is like saying adding water to a eggs & flour reduces the calories.

Adding water to eggs & flour most definitely DOES reduce the calories! You're just not looking at it properly. Yes, of course the total calories are not reduced if you force a person to consume the entire quantity but the calories per unit volume are decreased and so for a person consuming the same quantity as before, the calories will most definitely be reduced.

For example:
Let's say you can eat 1 cup of eggs & flour which a contain a total of X calories. If you add 1 cup of water to the mix, does that mean you're suddenly capable of consuming 2 cups? No. If you continue to consume 1 cup as before then the you'll now be consuming half the calories.

Quote:

Want to eat 3 Big Macs and be healthy? Just drink a gallon of water! Water away those calories..

Do you think a person who ate 3 Big Macs could also consume a gallon of water? Have him drink the gallon first and then see how many Big Macs he can eat! Can't eat three? Well hey, that reduces his calories!!

Quote:

As said by me: To say having odds on a come or line bet decreases the house edge is to say that the odds bet has a house edge. Or in other words, water by itself has no calories, but if you pour it into a coke - now water has calories.

Again this is looking at it wrong. Nobody is saying (or should be saying) that the water now has calories but rather the COMBINATION of food and water has fewer calories per volume unit than the food alone.

Each and every bet on the table has an expected value - some negative, some zero. Are you saying that you can't compute a combined expected loss if one of the included bets has an individual EV of zero? That's nonsense. You most definitely CAN compute a combined expected loss and it would rightfully be a weighted average of the included bets.

In the case of a line bet with odds, your combined average has a lower loss per dollar of action than the flat bet alone but a greater loss per dollar of action than the odds bet alone. So of course you're not lowering the house edge of the line bet itself. You're lowering the house edge per unit of action when compared to flat betting alone.

Therefore, just like the guy who can only consume X amount of food, a gambler who bets X number of dollars utilizing flats + odds will lose less than if he had bet X number of dollars on flat bets alone.

Steen
RS
Joined: Feb 11, 2014
• Posts: 8623
April 12th, 2016 at 4:20:10 PM permalink
Quote: MathExtremist

Yet baccarat is far and away the biggest money maker in the table game pit, especially on a per-table basis. Nothing else comes close, and the SD of the banker bet is even smaller than the craps passline. Go figure.

If I remember correctly, it's also one of the most volatile for the casino....since people tend to bet together with large spreads....they bet \$25 min for a long time then they all jump up to \$1k...
Steen
Joined: Apr 7, 2014
• Posts: 126
April 12th, 2016 at 6:27:32 PM permalink
Quote: odiousgambit

Let me strike out what I said upthread,

the only thing indisputable is that the guy who puts \$50 on the pass line and never adds odds could lower his EV and HE both by putting \$10 on the line and \$40 on free odds when it comes to EV and HE in Craps, there is absolutely nothing that is indisputable

Why did you backtrack on this? There's absolutely nothing wrong with what you wrote. A gambler betting \$10/\$40 will most definitely lose less on average that one betting \$50 flat.

Now, it's also true what Sally wrote. You should compare something like \$10/\$60 or \$30/\$30 to \$50 flat because they all produce the same average amount of action. When you compare \$50 flat to \$10/\$40, part of the reduction in average loss is due to the shifting of action from the flat to the odds (which I believe was your point) but part is due to the lower average action. Nevertheless I think the gist of your argument was correct.

Now, about this pouncing. How did I pounce on you? I actually agreed with your basic formula but just showed that it needed to be adorned with units and made into a correct percentage so it could be seen that the difference in interpretation was due to a mix-up of EV expressed in dollars versus one expressed as a percentage.

Steen