Best EV for Matchplay and Freeplay coupon?

[Three]

What would be the difference if I used my own $50 on roulette and the freeplay on a BJ hand? The outcome is going to be the same either way....right? So why am I throwing my own $50 at a -EV game, when I can use it for a single bet (or part of a bet) at the BJ table with +EV?

if you think about it betting the freeplay as a bet you would make anyway you only are better off if you lose it. If you win you lose the coupon and win what you would have anyway. If you lose instead of losing cash you lose a piece of paper. The other part is you wouldn't make defensive splits. You have nothing at risk. So splitting puts money out at a disadvantage. Then there is doubling. You double a hand and give up some EV to get a smaller edge with double at risk and reward. The situation is strong so the math adds up. Now you have nothing at risk but a piece of paper. If you hit you risk nothing to win the $50 on an 11 that you will be able to hit again if you like. If you double against a 7 or higher you would hit a stiff if you could. By doubling the coupon your risk goes from 0 to win $50 to risk $50 to win $100 and give up the chance to hit that stiff as you would like to. It makes sense at risking $50 to win $50 to sell your ability to hit in order to get $100 on the felt to win $100 but when you are risking a coupon that costs you nothing if lost is it wise to sell the ability to hit by going from nothing at risk that wins $50 to $50 at risk in order to win $100 rather than $50? Most say the value of the coupon is half is face value but once it has a matchup attached to it the value changes. You have similar but different issues with a match play. These are questions you want to know the answers to before you put that coupon on the felt. If you play the coupon at a different game that is now a +EV bet you wouldn't have made anyway none of these issues come into play.

[AP360]

if you think about it betting the freeplay as a bet you would make anyway you only are better off if you lose it. If you win you lose the coupon and win what you would have anyway. If you lose instead of losing cash you lose a piece of paper. The other part is you wouldn't make defensive splits. You have nothing at risk. So splitting puts money out at a disadvantage. Then there is doubling. You double a hand and give up some EV to get a smaller edge with double at risk and reward. The situation is strong so the math adds up. Now you have nothing at risk but a piece of paper. If you hit you risk nothing to win the $50 on an 11 that you will be able to hit again if you like. If you double against a 7 or higher you would hit a stiff if you could. By doubling the coupon your risk goes from 0 to win $50 to risk $50 to win $100 and give up the chance to hit that stiff as you would like to. It makes sense at risking $50 to win $50 to sell your ability to hit in order to get $100 on the felt to win $100 but when you are risking a coupon that costs you nothing if lost is it wise to sell the ability to hit by going from nothing at risk that wins $50 to $50 at risk in order to win $100 rather than $50? Most say the value of the coupon is half is face value but once it has a matchup attached to it the value changes. You have similar but different issues with a match play. These are questions you want to know the answers to before you put that coupon on the felt. If you play the coupon at a different game that is now a +EV bet you wouldn't have made anyway none of these issues come into play.

Well put Sir

[RS]

Ah, that makes a lot more sense Tthree. I figured it may have had to do with the doubling and splitting aspect of BJ. I'll check out that article you posted, mofungoo.

[blackjack avenger]

Or depends NOT

the rules of the coupon hardly matter

given a $10 coupon if you make a $1 a hand at bj and your going to play roulette?
I imagine you would miss approx 3 hands? Or $3 in EV!

So if you make a $1 a hand at bj. I JUST ADDED 30% TO THE COUPON VALUE BECAUSE YOUR PLAYING BJ and not playing a negative EV game. I am referring to your average EV per hand seen/played.

Like I said I don't think its never mentioned that bj is positive EV! : devilish:
You are standing at a table counting it!

The difference in these decisions isn't worth the worrying. Is half your play coupons?

Or stumble to another game and miss $3 in bj EV and play a negative EV game That's a $6 swing.

[Nyne]

Ive always heard this that when you have such coupons or as in back in the day during my online bonus days some bonuses we suggested that I play them on games with a high variance. I truly never understood as to why this would be the best case. Is it for the fact that you would want to go for a higher payout that a higher variance game can provide? This is what I had always assumed.

Like moo said, the matchplay is similar to a loss rebate. So essentially, you're making two bets. (Let's assume a $10 matchplay.) One, you are betting $20 on the actual game, whether it's BJ, roulette, etc. Two, you're getting a freeroll that pays you $10 if you lose your main bet. This is the loss rebate aspect. The net result is if you win the main bet, you get paid $20 times the payoff odds of that game, or if you lose, you lose $20, less the $10 rebate, for an actual loss of $10. It sounds complicated when written out like that, but thinking of it that way is key to understanding why variance is good.

Now, we all know that BJ has a lower house edge than roulette, right? So why is roulette better for a matchplay? The reason is because the roulette payout distribution is skewed. When betting a single number straight up, we lose 37/38 spins. The loss rebate portion of our combined bets has an EV that depends solely on the likelihood of losing, so the larger the chance that we lose the main bet, the better, at least for the loss rebate. We give up a little EV on the main bet to get a lot of EV on the loss rebate. When we say that variance is good, what we really mean is that a high chance of losing helps to maximize the loss rebate value, which helps maximize the overall value of the coupon. This is even more clearly illustrated when you compare the value of using the matchplay on different roulette bets. The Beyond Coupons article linked earlier in this thread should have the specific numbers. Each bet on roulette has the same house edge, yet betting a number straight up is the best use of the matchplay specifically because the chance of losing is higher, so the EV of the loss rebate is higher.

In the days of the online bonuses, sticky bonuses also were very much like a loss rebate. Deposit $100, get a $200 sticky bonus? That's the same as depositing $300 and having a $200 loss rebate if you lose it all. So again, the value of the loss rebate is maximized by increasing the likelihood of losing the $300 (and triggering the $200 rebate). That's why it was better to bet big and try to double up a couple of times. If you lose, you get the rebate. If you win enough to hit your target, then you can grind out the wagering requirements with small bets. Even on bonuses that weren't sticky, sometimes it made sense to bet a lot and either bust or win big, especially if the bonus had a very high wagering requirement. If you do several different bonuses, each with large wagering requirements, using this tactic will result in big wins at some sites and immediate losses at others, which brings down your overall average playthrough. If you have 8 bonuses, each with 100x playthrough requirements, but before starting, you try to double each starting stake 3 times (I'm assuming a fair coinflip, but actual games aren't much worse), on average you'll bust 7, but start the eighth one with 8x the money, and you'll only need to complete the 100x playthrough requirements for that one bonus, not all 8. For a $100 bonus, that means you have to wager $10,000 instead of $80,000, saving you a significant amount of -EV play. If you had to deposit $100 to get that $100 bonus, that means you will have lost your $100 at the other sites, but you'll have a $1600 stake to work with on the site where you didn't bust out, so you've still got your $800 in buy-ins and $800 in bonuses, but by betting big on several sites, you've effectively consolidated the money into one site while avoiding a lot of the required playthrough. Of course, the actual games you can use to try to double your money will have a house edge associated with them, so in truth you'll be a little worse off than this example suggests, but the -EV of making a few big bets on those games is less than the -EV of simply completing the playthrough requirements at each site, so betting big at other games to generate variance is a viable strategy for increasing the EV of bonuses. This can even make bonuses that look -EV playable, since this strategy effectively reduces the "required" playthrough requirements. 100x playthrough and the best game that counts has a 1.5% HE? We just turned 100x into something closer to 25x, so rather than an expected loss of 1.5 times the bonus, we now expect to lose 37.5% of the bonus and cash the rest, on average.

Hopefully this helps explain why variance can help with coupons or with bonuses. In general, we are trying to create a skewed payout distribution with a large number of small losses and a few large wins, so we can take advantage of the loss rebate more frequently.

[blackjack avenger]

Nyne

Please read and comment on my post

Bj is not negative EV for a counter. The whole premise is mistaken.

[Nyne]

given a $10 coupon if you make a $1 a hand at bj and your going to play roulette?
I imagine you would miss approx 3 hands? Or $3 in EV!

So if you make a $1 a hand at bj. I JUST ADDED 30% TO THE COUPON VALUE BECAUSE YOUR PLAYING BJ and not playing a negative EV game. I am referring to your average EV per hand seen/played.

The fact that I go play the coupon at roulette doesn't mean I'm going to play less blackjack. I'm going to play blackjack for my intended session length, no more, no less. If I intend to show my spread once, twice, whatever, then leave, I'm not going to get to play 3 more hands of BJ because I didn't play my coupon at roulette. Instead, maybe when the count tanks and I wong out, and there isn't a fresh shuffle available, maybe that's a good time to drop by the roulette table and play the coupon. Maybe I play the coupon while backcounting the adjacent BJ game. If I don't get a break this session, I can save it for next time. I think most APs would agree that there are a lot of short periods of time to fill when you're in a casino. Finding time to play one spin of roulette shouldn't make you miss out on any other available play.

Also, in reply to your last post, the EV of the base game isn't the main factor that determines the value of the coupon, it's the likelihood of loss. Since BJ is much lower variance, your likelihood of a loss in any given hand is never really far off of 50%, which means even if you play the coupon on a bet you are already making, the loss rebate value is never more than about 50% of the rebate, versus roulette where the rebate value is over 97% of the amount of the rebate (since the rebate pays of 37 times out of 38). You give up 5.26% of $20 to gain 97% of $10, or thereabouts. For BJ, the maximum gain is about 50% of that same $10 rebate. The EV of blackjack has nothing to do with why it's not the best choice for matchplays.

[blackjack avenger]

I think most sessions are time limited. A trip to Vegas or AC as an example.

Think your splitting hairs proves my point its much to do about nothing.

I will undercut some of your claims about waiting. One wants to play it quick in case of getting backed off. So one should not wait. Get to the bj table and START COUNTING!

I was going to state it depends on casino layout. The game you get to first! LOL. In order to save time.

However, the EV swing of positive expectation BJ beats everything. My example is easy to follow.

Should one play a coupon on a positive or negative EV game? I would choose positive EV

Another thought on dead time. A shoe half dealt has more value then a roulette table doesn't it? Stay where the money is.

[Three]

given a $10 coupon if you make a $1 a hand at bj and your going to play roulette?
I imagine you would miss approx 3 hands? Or $3 in EV!

When I want to wong out I go play my coupons. It is +EV on both ends.

[blackjack avenger]

Having thought of my position overnight.

I'M STILL RIGHT!
but it depends.

My basic premise is standing in a bj pit has value to a counter which is never considered in these discussions.

If the coupon is equal to a minimum bet your probably better off at bj because every hand of bj you watch has value. A value that can be worth more in a few hands then the value of the coupon. As the coupon approaches ones max bets then other plays increase in value.

As far as casino layout, speed of games, rules, etc. are all very open variables.
I will add this. Many have talked about when nothing is available in bj. Unless the pit is closed several half dealt shoes have value.

Waiting any length of time you may lose the ability to play it. The casino can revoke at any time.

[moo321]

Read this: http://www.beyondcounting.com/pdfs/b...ouponsbjfo.pdf

A bet on high variance games can be worth twice as much as other games.

Some may ask: if it's a $10 matchplay, why are we splitting hairs over a few dollars here and there?

Answer: not all matchplays are $10. I have had $500 matchplays. I know people who have had several thousand dollars in matchplays. Having $2000 in matchplay be worth $1800 vs. $900 is not splitting hairs. This theory also applies to sticky bonuses. It also applies to loss rebates. It could potentially apply to all sorts of situations in casinos. It applies to options valuations in finance, etc. etc. etc.

[RS]

I've been reading through it -- takes a little while because I'm not familiar with the whole coupon thing. I'm still having a really tough time seeing how there's a difference in playing $20 cash on roulette and $10 cash + $10 coupon in BJ....compared to $10 cash + $10 coupon on roulette and $20 cash on roulette. [Well....other than the times when "if you surrender, you lose half your bet AND the entire coupon" or the "if you double-down you're triping your risk ($10 -> $30) to win twice as much ($20 -> $40). Either way, the win/loss amount is going to be the same if we lose BJ and win roulette or win BJ and lose roulette or any other combination.] I'll start over from the beginning again....but I don't think my brain wants to see the difference! >

[Nyne]

I've been reading through it -- takes a little while because I'm not familiar with the whole coupon thing. I'm still having a really tough time seeing how there's a difference in playing $20 cash on roulette and $10 cash + $10 coupon in BJ....compared to $10 cash + $10 coupon on roulette and $20 cash on roulette. [Well....other than the times when "if you surrender, you lose half your bet AND the entire coupon" or the "if you double-down you're triping your risk ($10 -> $30) to win twice as much ($20 -> $40). Either way, the win/loss amount is going to be the same if we lose BJ and win roulette or win BJ and lose roulette or any other combination.] I'll start over from the beginning again....but I don't think my brain wants to see the difference! >

The difference is you are far more likely to lose in roulette, so you'd rather lose $10 cash and the coupon there instead of $20 cash. The value of the coupon as a loss rebate only kicks in on a loss, and that happens 37 out of 38 times on a number at roulette. Simplifying BJ to a fair coin flip for the sake of argument, you get the following for the two scenarios you mentioned:

Bet $20 on roulette and $10 cash + $10 matchplay on BJ
EV = 1/38 * 700 + 37/38 * (-20) + 0.5 * 20 + 0.5 * (-10) = $3.95

Bet $10 cash + $10 matchplay on roulette and $20 cash on BJ
EV = 1/38 * 700 + 37/38 * (-10) + 0.5 * 20 + 0.5 * (-20) = $8.68

It's worth $4.73 more with the matchplay on roulette because you get the $10 discount 37/38 times instead of 1/2 of the time. $10 * (37/38 - 1/2) = $4.73

To take advantage of a loss rebate, you should aim for a payout distribution that maximizes the frequency of the loss (but with a large infrequent win to make sure the actual EV of that payout distribution is still reasonable).