MJ: Question for Don: Compensation for player/investor

[MJ]

Would it make any sense for an investor to back a solo card counter who would play once a week for 6 months at roughly 5 hours/trip? The card counter's style is strictly WiWo and would play to around 0.5 kelly. The table minimum is $25. AC rules would apply.

26 weeks of play boils down to 130 hrs of backcounting at around 60 hands dealt per hour. Do you think that there is a reasonable chance for the wonger to be ahead after 6 months of play?

Once the bank is broken, is a 50/50 split for player/investor fair once expenses have been deducted? Take a case where the investor puts up $20k. Let us say that it takes 40 trips to the casino to win $10k (break bank). Expenses work out to be $2k. $10k - $2k = $8k. NOW, does it make sense for the player and investor to each receive $4k? If you think about it, the investor's ROI = 40%, not bad. If it required 80 trips to break bank, then the expenses would be doubled and the ROI for the investor would be 30%.

Some might say that the investor should be compensated based upon the Gross win rather then Net win; he should receive 50% of 10k rather than 50% of 8k. What do you think?

Is it better to break bank based upon a designated time period, after a goal has been reached, or whichever comes first?

Of course, all of this assumes the wonger will be ahead after 130 hours of wonging. I am uncertain what N0 would be for this particular situation. Is this even worthwhile or is it more or less gambling for the investor due to the limited time factor?

Thanks,
MJ

[Don Schlesinger]

> Would it make any sense for an investor to back a solo
> card counter who would play once a week for 6 months
> at roughly 5 hours/trip? The card counter's style is
> strictly WiWo and would play to around 0.5 Kelly. The
> table minimum is $25. AC rules would apply.

Sure. Why not? But AC rules are pretty shitty. Las Vegas, with LS and RSA, would be much better.

> 26 weeks of play boils down to 130 hrs of backcounting
> at around 60 hands dealt per hour.

Where does that come from? You aren't going to back-count full tables; that's a waste of time. So, the dealer is going to deal more than 60 rounds per hour. But, with AC rules, you aren't going to play more than 25-27 of them.

> Do you think that
> there is a reasonable chance for the wonger to be
> ahead after 6 months of play?

Well, talk about hours, not months, because that's all that matters. You're going to back-count for 130 hours. Using Table 10.43, from p. 236, Let's say you make $40 per hour with $20,000 of bank (you stipulated half-Kelly), and 16 times that, or $640 hourly s.d. For 130 hours, your e.v. is $5,200 and your s.d. is $7,300. So, your probability of being ahead after the 130 hours is the same as not underperforming by more than 5200/7300 = 0.71 s.d.s, which is 76%. So, the odds are 3 to 1 in your favor that you'll be ahead at the end of the six months.

> Once the bank is broken, is a 50/50 split for
> player/investor fair once expenses have been deducted?

No, probably not. In my book, on pp. 298-301, I discuss what I think would be a reasonable amount of play for a 50-50 split and how to make the division if fewer hours are played. The determination of the 50-50 number is subjective, but, if accepted, then the determination of the other values is purely mathematical and flows from the initial decision.

My choice corresponds to Brett Harris's N0 value, or the number of hours necessary for expectation to catch up to one s.d., such that e.v. - s.d. = 0. In the above example, it would be more like 256 hours, or double what you're proposing.

> Take a case where the investor puts up $20k. Let us
> say that it takes 40 trips to the casino to win $10k
> (break bank). Expenses work out to be $2k. $10k - $2k
> = $8k. NOW, does it make sense for the player and
> investor to each receive $4k?

See above. Are you sharing the expenses, or is the investor bearing 100% of them?

> If you think about it,
> the investor's ROI = 40%, not bad.

It's not an annualized 40% (if that's what you meant) in your example unless the player manages the 40 trips in six months. Earlier, you said one trip per week, which is only 26 trips in six months. To get to 40, you'd need 10 months. So, you can't annualize the 20% investor's share to become 40% the way you did.

> If it required 80
> trips to break bank, then the expenses would be
> doubled and the ROI for the investor would be 30%.

See above. I think you're forgetting that if you make $4,000 each on a $20,000 bank, that's 20% each, not 40%.

> Some might say that the investor should be compensated
> based upon the Gross win rather than Net win; he
> should receive 50% of 10k rather than 50% of 8k. What
> do you think?

Expenses have to be paid by someone. If the investor bears the entire burden, then you have to account for that some way. If 50% of the gross, then the ultimate split won't be 50-50, will it? Investor will get $5,000 and player $3,000.

> Is it better to break bank based upon a designated
> time period, after a goal has been reached, or
> whichever comes first?

Yes! :-) This has been debated for years. At the minimum, the player has to commit to a predetermined minimum number of hours of commitment to the project. It sucks to get way behind early and to have to dig out, knowing that you made a commitment to the investor. Trust me, I've been there -- on both sides. But, it isn't fair otherwise. On the flip side, if you win very quickly, and make the split, everyone is happy with the premature division. So, for example, you commit to a certain number of hours (250?) or a doubling of the bank, whichever comes first.

> Of course, all of this assumes the wonger will be
> ahead after 130 hours of wonging.

See above. You have absolutely no guarantee whatsoever that that will be the case. I've heard of teams playing for 1,000 hours and not being ahead! Losing streaks can be vicious.

> I am uncertain what
> N0 would be for this particular situation. Is this
> even worthwhile or is it more or less gambling for the
> investor due to the limited time factor?

Define "gambling." Yes, it is a gamble. But, it is a decent investment opportunity, and all investments bear risk. However 130 hours is not enough. It should be at least double that to be fair to the investor.

Don

[MJ]

> Sure. Why not? But AC rules are pretty shitty. Las
> Vegas, with LS and RSA, would be much better.

Too expensive to fly out to Vegas.

> Where does that come from? You aren't going to
> back-count full tables; that's a waste of time. So,
> the dealer is going to deal more than 60 rounds per
> hour. But, with AC rules, you aren't going to play
> more than 25-27 of them.

I am just being conservative in my estimate. The player would backcount any table where there is at least one spot available. Tables with more than one spot open will of course be given priority.

Just to clarify, I assume that 60 rounds/hr will be observed. Don't forget that even if 100 rounds are dealt per hour, that doesn't mean you will observe every last round. You must consider lag time once you wong out of a shoe and lag time for when you depart a shoe while observing.

> Well, talk about hours, not months, because that's all
> that matters. You're going to back-count for 130
> hours. Using Table 10.43, from p. 236,

Stop right there! Table 10.43 assumes the backcounter never departs the shoe and just stands behind the same table regardless of how negative the count becomes! The style of wonging I am talking about is WiWo. Two different approaches.

> Let's say you
> make $40 per hour with $20,000 of bank (you stipulated
> half-Kelly), and 16 times that, or $640 hourly s.d.

But you are using different BC assumptions. See above. With WiWo, the earnings and SD will be higher.

Additionally, if you factor in a $50 expense per trip, then that is $10/hr in trip expenses. Thus, we are losing $10/$40 = 25% of hourly earnings on expenses. Hence, while the bet schedule may reflect 0.5 kelly, the counter is really playing to 0.5/75% = 2/3 kelly once expenses are factored into the equation. Do you agree?

> For 130 hours, your e.v. is $5,200 and your s.d. is
> $7,300. So, your probability of being ahead after the
> 130 hours is the same as not underperforming by more
> than 5200/7300 = 0.71 s.d.s, which is 76%. So, the
> odds are 3 to 1 in your favor that you'll be ahead at
> the end of the six months.

Again, you are using pure BC assumptions with no departure points. I suppose your numbers are good for ballpark figures though. After 130 hours of play with WiWo, the % chance of being ahead would probably be in the range of 80%-85%.

> No, probably not. In my book, on pp. 298-301, I
> discuss what I think would be a reasonable amount of
> play for a 50-50 split and how to make the division if
> fewer hours are played. The determination of the 50-50
> number is subjective, but, if accepted, then the
> determination of the other values is purely
> mathematical and flows from the initial decision.

Assuming the split was not 50-50, but 30-70, would the p formula now equal (100 x .3bn)/xhw?

> My choice corresponds to Brett Harris's N0 value, or
> the number of hours necessary for expectation to catch
> up to one s.d., such that e.v. - s.d. = 0. In the
> above example, it would be more like 256 hours, or
> double what you're proposing.

256 hours is an entire year before breaking bank, assuming the target is not reached prematurely. This is too long.

> See above. Are you sharing the expenses, or is the
> investor bearing 100% of them?

I suppose the fair thing is for expenses to be split, 50-50.

> Yes! :-) This has been debated for years. At the
> minimum, the player has to commit to a predetermined
> minimum number of hours of commitment to the project.
> It sucks to get way behind early and to have to dig
> out, knowing that you made a commitment to the
> investor.

Generally speaking, I guess players want to jump ship if things take a downward spiral early on because they will not break bank anytime soon. So, their payday is a far way off.

From the investor's perspective, I can see why he would want the player to commit to a predetermined # hours. This way his investment is protected to a certain degree and he can be more confident in making money due to the long run.

> Trust me, I've been there -- on both sides.
> But, it isn't fair otherwise. On the flip side, if you
> win very quickly, and make the split, everyone is
> happy with the premature division. So, for example,
> you commit to a certain number of hours (250?) or a
> doubling of the bank, whichever comes first.

But if the bank takes a severe downward spiral in the beginning, what incentive is there for the player to continue to play?

Also, I read the article on pp. 298-301. The formula for w gives avg win, assuming there is a win. W is $150 and p = 33%. Does this mean the players receive 50 units after each 40 hr segment of play or that the player simply receives 33% of any win for a 40 hr segment of play?

The reader asked how profits should be split percentage wise for a given time period. What happens if there is no profit at the end of a 40 hour session? Players should be given some type of compensation in the short run irrespective of whether there is a profit or not.

> Define "gambling." Yes, it is a gamble. But,
> it is a decent investment opportunity, and all
> investments bear risk. However 130 hours is not
> enough. It should be at least double that to be fair
> to the investor.

I don't follow your logic. Why is 130 hours "unfair" to the investor? If the counter is competent he is generating a positive EV. You even said yourself that the odds are 3 to 1 of the player being ahead at the end of 130 hours.

MJ

[Don Schlesinger]

MJ,

I need to scold you a little first, then I'll get back to explaining everything for the second time, ostensibly saying the same things I said the first time! :-)

One of your problems seems to be that, often, after you solicit the advice of an expert, specifically designating him by name (in this case, me), and you get a well-reasoned and complete answer, you then take exception with what you're being told, in this case, actually bringing up sections of my own book to me, as if I'm not aware of what they say. What sense does that make? You're talking to the person who wrote the book, and you're quoting material as if I may not be familiar with it.

> Too expensive to fly out to Vegas.

Figure out the difference in e.v., and you might be penny-wise and pound foolish.

> I am just being conservative in my estimate. The player would backcount any table where there is at least one spot available. Tables with more than one spot open will of course be given priority.

Many more of these in Las Vegas than in, say, A.C.

> Just to clarify, I assume that 60 rounds/hr will be observed. Don't forget that even if 100 rounds are dealt per hour, that doesn't mean you will observe every last round. You must consider lag time once you wong out of a shoe and lag time for when you depart a shoe while observing.

That has a familiar ring to it. :-) Where did you learn that? :-) In any event, despite what you've written, it's much closer to 100 than 60. In a large casino, you can almost always find a new shuffle rather quickly, if you are aggressive.

Don: >Well, talk about hours, not months, because that's all that matters. You're going to back-count for 130
hours. Using Table 10.43, from p. 236,

> Stop right there! Table 10.43 assumes the backcounter never departs the shoe and just stands behind the same table regardless of how negative the count becomes! The style of wonging I am talking about is WiWo. Two different approaches.

Thank you for explaining that. :-) Now, go and read the middle paragraph of p. 356 very carefully and very slowly, and savor every word. Do you think I would have had Norm create the Chapter 10 charts as a curiosity, just for fun, if I didn't think they had practical, real-world applicability and represented, closely, what the average player, under real-world casino conditions, could win?

Don: > Let's say you make $40 per hour with $20,000 of bank (you stipulated half-Kelly), and 16 times that, or $640 hourly s.d.

> But you are using different BC assumptions. See above. With WiWo, the earnings and SD will be higher.

Read my lips: No they won't. And, since you're not a computer, the earnings will almost certainly be less.

> Additionally, if you factor in a $50 expense per trip, then that is $10/hr in trip expenses. Thus, we are losing $10/$40 = 25% of hourly earnings on expenses. Hence, while the bet schedule may reflect 0.5 kelly, the counter is really playing to 0.5/75% = 2/3 kelly once expenses are factored into the equation. Do you agree?

No. That's not the right way to figure it. To really net $40, you'd have to gross $50 per hour, to compensate for expenses. So, you'd need, for full Kelly, 25% more bank than $10,000, or $12,500. So, if you keep the bank at $20,000, then that's 12,500/20,000 = 5/8-Kelly. But, I don't think this is as important as you make it out to be.

Don: > For 130 hours, your e.v. is $5,200 and your s.d. is
> $7,300. So, your probability of being ahead after the
> 130 hours is the same as not underperforming by more
> than 5200/7300 = 0.71 s.d.s, which is 76%. So, the
> odds are 3 to 1 in your favor that you'll be ahead at
> the end of the six months.

> Again, you are using pure BC assumptions with no departure points. I suppose your numbers are good for ballpark figures though. After 130 hours of play with WiWo, the % chance of being ahead would probably be in the range of 80%-85%.

No, not at all. You're being much too optimistic.

Don: > No, probably not. In my book, on pp. 298-301, I
> discuss what I think would be a reasonable amount of
> play for a 50-50 split and how to make the division if
> fewer hours are played. The determination of the 50-50
> number is subjective, but, if accepted, then the
> determination of the other values is purely
> mathematical and flows from the initial decision.

>Assuming the split was not 50-50, but 30-70, would the p formula now equal (100 x .3bn)/xhw?

Not sure. I'll have to go back and reread the math. Who gets the 30%, the player?

Don: > My choice corresponds to Brett Harris's N0 value, or
> the number of hours necessary for expectation to catch
> up to one s.d., such that e.v. - s.d. = 0. In the
> above example, it would be more like 256 hours, or
> double what you're proposing.

> 256 hours is an entire year before breaking bank, assuming the target is not reached prematurely. This is too long.

It's your arrangement; do whatever you like. I'm simply pointing out that, were I the investor, I would not agree to just 50% of the profits for only 130 hours of play. I'd use the formula in my book and require a larger percentage for the split. I'd require 256 hours or more before I'd agree to getting only 50% of the profits.

Don: > See above. Are you sharing the expenses, or is the
> investor bearing 100% of them?

>I suppose the fair thing is for expenses to be split, 50-50.

Well, then, forget about them and make sure that you play an adequate number of hours to make them relatively unimportant with respect to expectation.

Don: > Yes! :-) This has been debated for years. At the
> minimum, the player has to commit to a predetermined
> minimum number of hours of commitment to the project.
> It sucks to get way behind early and to have to dig
> out, knowing that you made a commitment to the
> investor.

> Generally speaking, I guess players want to jump ship if things take a downward spiral early on because they will not break the bank anytime soon. So, their payday is a far way off.

Correct.

> From the investor's perspective, I can see why he would want the player to commit to a predetermined # hours. This way his investment is protected to a certain degree and he can be more confident in making money due to the long run.

Now, you're getting there! :-)

Don: > Trust me, I've been there -- on both sides.
> But, it isn't fair otherwise. On the flip side, if you
> win very quickly, and make the split, everyone is
> happy with the premature division. So, for example,
> you commit to a certain number of hours (250?) or a
> doubling of the bank, whichever comes first.

> But if the bank takes a severe downward spiral in the beginning, what incentive is there for the player to continue to play?

Uh, because he's, er, honorable (?) and because he made a financial agreement with the investor?

> Also, I read the article on pp. 298-301. The formula for w gives avg win, assuming there is a win. W is $150 and p = 33%. Does this mean the players receive 50 units after each 40 hr segment of play?

Certainly not. What if 40 hours produces a loss? Why would the players get paid? From where would the money come?

> or that the player simply receives 33% of any win for a 40 hr segment of play?

Correct.

> The reader asked how profits should be split percentage wise for a given time period. What happens if there is no profit at the end of a 40 hour session?

The investor bears the loss, and you start a new time period.

> Players should be given some type of compensation in the short run irrespective of whether there is a profit or not.

Then you'll have to give the investor an even larger percentage than what my formula stipulates. You think money grows on trees? On top of everything else, you now want the player to have a salary too? Who pays it? You want a guaranteed hourly wage, even if you earn nothing all year? Good luck to you finding an investor for that!

Don: > Define "gambling." Yes, it is a gamble. But,
> it is a decent investment opportunity, and all
> investments bear risk. However 130 hours is not
> enough. It should be at least double that to be fair
> to the investor.

> I don't follow your logic.

Yes you do. You just don't like the answer; that's not the same thing! :-)

> Why is 130 hours "unfair" to the investor?

Already asked, already answered. It's too short a time period to give the investor a fair shot at a reasonable expectation to make money rather than lose it. You really don't seem to understand how ridiculously short a time period 130 hours is. Or, did you miss my point about the team that played over 1,000 hours and was still behind, after expenses?

> If the counter is competent he is generating a positive EV.

Every hour. So, why not agree to split profits, 50-50, after every hour? You win, investor gets half; you lose, investor eats it. Fair? Oh. I see. The investor is surely ruined with that arrangement. So, we try again. And we increase the hours to a point where there is a very high probability of making money -- higher than 75%. In my example in the book, enough hours such that the expectation is to double the bank. Then, there is only 16% chance of losing.

> You even said yourself that the odds are 3 to 1 of the player being ahead at the end of 130 hours.

NOT enough for the investor to accept a 50-50 split. Unless he's a stupid investor. :-)

Don

[MJ]

Sorry for the delay, I have been quite busy.

> You're talking to the person who wrote
> the book, and you're quoting material as if I may not
> be familiar with it.

I was merely explaining why I felt your analysis might be more of a crude estimate based upon your ODP study from BJA3.

Who would expect a backcounter who stands behind the table all day jumping in and out of the shoe and never leaving to have the same SCORE as a WiWo player? Furthermore, although you point out the similar performance between the protypical backcounter and lag WiWo, you never really offer any explanation for it (p.356). Do you care to take a stab at it? :-)

Does this mean I can stand behind the table when my friends are playing, come in and out at will, never leave the table and my SCORE will be about the same as if I departed to look for a freshly shuffled shoe every time the TC dipped to -1?

> Figure out the difference in e.v., and you might be
> penny-wise and pound foolish.

I looked into the expenses associated with a flight to Vegas. A round trip flight from the Northeast plus 2 night stay in a hotel, car rental, and other expenses will easily run $700. My expectation would be about $60/Hr backcounting in a 6D game with good rules.

Over the course of 3 days, I could probably play no more then 6 hours a day without starting to make errors due to fatigue. That boils down to 18 Hrs x $60/Hr = $1080 expectation for a 3 day trip. That means expenses would consume $700/$1080 = 65% of my expected earnings, leaving only $380 net for a trip.

Ok, if we standardize the bank from $20k at 0.5 kelly to 10k at full kelly, then that means we need to increase the 10k bank by 165% to offset expenses. So, it would require $16,500 to play at full kelly given our expenses. Going back to the 20k bank, after taking expenses into account we go from 0.5 kelly to $16.5k/$20k = 0.825 kelly.

Is that correct?

With AC, we established that expectation was $40/Hr x 5 Hrs = $200 backcounting. Then there is $50 expenses for the trip, which leaves $150 net. I believe you said the adjusted kelly-factor after taking expenses into account is 0.625.

In summary:

Vegas: 0.825 kelly with $380 Net
AC: 0.625 kelly with $150 Net

Before we can fairly compare the two net earnings, we must make the aforementioned kelly factors equivalent. If we convert the AC kf to Vegas kf:

0.825/0.625 = 1.32 conversion factor. 1.32 x $150 = $198 AC net at 0.825 kelly.

So, I guess the Vegas game really is much better than AC. Please let me know if I did this correctly.

I have more to say on this post, but will write it after I get a response.

Thanks,
MJ

[MJ]

> Who would expect a backcounter who stands behind the
> table all day jumping in and out of the shoe and never
> leaving to have the same SCORE as a WiWo player?
> Furthermore, although you point out the similar
> performance between the protypical backcounter and lag
> WiWo, you never really offer any explanation for it
> (p.356). Do you care to take a stab at it? :-)

> Does this mean I can stand behind the table when my
> friends are playing, come in and out at will, never
> leave the table and my SCORE will be about the same as
> if I departed to look for a freshly shuffled shoe
> every time the TC dipped to -1?

One other point for now. If the performance is that identical between the prototypical backcounter and lag WiWo, why even bother with departure points for playing and observing the shoe? It would seem like lag WiWo is just wasting his energy going from table to table when he can make the same $$$ just standing behind one table.

Also, I believe one of the assumptions of the CVCX backcounting charts is that every time the backcounter enters the game, another player leaves so as not to throw off the TC frequencies. I am not sure how much of an affect this would have on the backcounter's SCORE, but I would think it must inflate it just a bit.

There will be some cases when the count is high and the wonger enters the game and as a result of the extra player leaving, the wonger will get an extra round due to the player leaving which would save around 2.7 cards from coming out of the shoe.

Was there ever any type of comparison done between CVData and the CVCX charts for wonging? I believe CVData would yield a more accurate SCORE due to all the players remaining at the table once the wonger enters.

MJ