Optimum Betting and Spread

[Finrod]

I was working on a spreadsheet to help me determine an optimum betting schedule (using Kelly) and came across the dilemma of using optimum betting and keeping the spread in check.

Of course, as your bank roll goes up, the size of your optimum max bet goes up. Assuming you want to use the optimum max bet when odds are in your favor and bet table minimum when the odds are against you, the spread will increase as your bankroll increases. At some point, the spread will be intolerable to the store.

Seems you have two choices: Decrease your optimized max bet or increase your min bet. In either case, you are eroding your EV. All the evidence you need to prove that heat due to spread decreases EV.

My question is, on which end do you make the adjustment?

Of course, it probably has something to do with what counts you plan to play in. If you only play in true counts of +2 or more, the answer is probably obvious. But of course that is not really practical. Lets assume for purposes of this discussion, I start every shoe and weather it out as long as the true count is -1 or higher (wong out at -2).

Appreciate your input as to how you have approached this dilemma.

 

[MangoJ]

With aggressive wonging you can flat bet, and still maintain an infinite spread.

 

[kewljason]

In my case , my max bet is set by tolerance threshold. I have determined an amount that I wish to stay under ($500) for heat purposes, and have set my max bet accordingly.

The minimum end of the spread is a little trickier. Obviously the most advantageous method would be table minimum to this predetermined max bet. In my case $5 or $10 to $400. That would be sweet, but you wont last long, as you are actually going through 2 levels, from red through green to black. Just too easy to spot. So I start at green. This gives my a more tolerated 16-1 on shoe games. Plus there are some benefits to a higher minimum wager. Often games at a certain level like $25 min or $50 min are superior in terms of rules and initial house advantage. Also the tables are less crowded and there are more tables available to you as you can still play the lower limit tables if an opportunity is there.

This is all pretty general terms though as in reality each location and situation needs to be evaluated independently. I play smaller, local stores that I can't possibly hope to get down my regular max wager even though it is permitted by rule. Betting black I would stand out as if I were painted fluorescent green. So you have to adjust to them.

 

[forwhat77]

Time for a BP KJ:whip::grin:

 

[bigplayer]

I was working on a spreadsheet to help me determine an optimum betting schedule (using Kelly) and came across the dilemma of using optimum betting and keeping the spread in check.

Of course, as your bank roll goes up, the size of your optimum max bet goes up. Assuming you want to use the optimum max bet when odds are in your favor and bet table minimum when the odds are against you, the spread will increase as your bankroll increases. At some point, the spread will be intolerable to the store.

Seems you have two choices: Decrease your optimized max bet or increase your min bet. In either case, you are eroding your EV. All the evidence you need to prove that heat due to spread decreases EV.

My question is, on which end do you make the adjustment?

Of course, it probably has something to do with what counts you plan to play in. If you only play in true counts of +2 or more, the answer is probably obvious. But of course that is not really practical. Lets assume for purposes of this discussion, I start every shoe and weather it out as long as the true count is -1 or higher (wong out at -2).

Appreciate your input as to how you have approached this dilemma.

If you want to use a fixed top bet you can ramp faster as your bankroll grows, it is sub-optimal but EV maximizing for your top bet as you keep your level of risk constant. I know many players who bump up to table maxes at $500 max casinos and simply make their max bet earlier in the count.

 

[blackjack avenger]

What Wicked This Way Comes

If you want to use a fixed top bet you can ramp faster as your bankroll grows, it is sub-optimal but EV maximizing for your top bet as you keep your level of risk constant. I know many players who bump up to table maxes at $500 max casinos and simply make their max bet earlier in the count.

I agree with the above, let's look deeper.

Putting out top bets at lower TCs; same ror, due to table max's has these effects:
Raises NO
Lowers SCORE
Lowers score
Lowers CE/WR
Lowers CE
Lowers DI
Total advantage drops
All the above is negative

Subjectively placing larger bets sooner draws heat sooner. If one raises a table max bet at TC 2 one is getting a lot of heat for a small return.

It lowers every game quality indicator I can think of. So as one's bank grows they are taking on more variance; same ror, which I believe is counter to investment strategy. One should want to be more conservative with larger sums of money. As one wins they are increasing variance and if a large loss one faces the possibility of cutting bets. Once positive variance returns they may repeat the unpleasant process.

A way around this? If the bank is very large; 1/4th to 1/8 kelly resizing, then the CE/WR ratio improves. The effect? The variance; hopefully, is not going to cause one to cut bets on losses. The bank is so large it can absorb the added varance.

Eventually, with a big enough bank one can bet max or maintain top bets at any advantage, gracefully.

 

[moo321]

In my case , my max bet is set by tolerance threshold. I have determined an amount that I wish to stay under ($500) for heat purposes, and have set my max bet accordingly.

The minimum end of the spread is a little trickier. Obviously the most advantageous method would be table minimum to this predetermined max bet. In my case $5 or $10 to $400. That would be sweet, but you wont last long, as you are actually going through 2 levels, from red through green to black. Just too easy to spot. So I start at green. This gives my a more tolerated 16-1 on shoe games. Plus there are some benefits to a higher minimum wager. Often games at a certain level like $25 min or $50 min are superior in terms of rules and initial house advantage. Also the tables are less crowded and there are more tables available to you as you can still play the lower limit tables if an opportunity is there.

This is all pretty general terms though as in reality each location and situation needs to be evaluated independently. I play smaller, local stores that I can't possibly hope to get down my regular max wager even though it is permitted by rule. Betting black I would stand out as if I were painted fluorescent green. So you have to adjust to them.

This is a very good post from an experienced player. Read this and understand it, especially if you want to try to play green-black.

 

[Gamblor]

My betting spreads are very non-optimal. At least mathematically.

 

[blackjack avenger]

NO Backwards, Backwards NO

I play smaller, local stores that I can't possibly hope to get down my regular max wager even though it is permitted by rule. Betting black I would stand out as if I were painted fluorescent green. So you have to adjust to them.

We do what we feel we must, the above raises your NO quite a bit.

Example:

Day 1 play $25 to $400 for 400 hands
Day 2 play $25 to $200 for 400 hands

You have basically 2 different NOs. They do blend but it's not the same as playing 800 hands at the same spread.

Anecdotally, I am sure anyone who does this has experienced winning using the smaller bets and losing using the bigger bets which adds up to losing overall!

 

[kewljason]

We do what we feel we must, the above raises your NO quite a bit.

Example:

Day 1 play $25 to $400 for 400 hands
Day 2 play $25 to $200 for 400 hands

You have basically 2 different NOs. They do blend but it's not the same as playing 800 hands at the same spread.

Anecdotally, I am sure anyone who does this has experienced winning using the smaller bets and losing using the bigger bets which adds up to losing overall!

Yes you will have different NO's That's the price you pay. Everything doesn't always fit neatly into the envelope. Unfortunately, if there is a game I want to play at say, Bighorn, I am going to have to adjust to them and what I feel I can get down.

Your anecdote is irrelivent. It could just as well work the other way, but, no matter, it's just short term results, and in the end will all work itself out if you play enough.

Keep in mind, that if I am willing to play that 'smaller' game, there must be something special that attracts me to it to begin with. Better rules, outstanding penetration, unusual side bet opportunity. I am not choosing to travel further and bet smaller just for the hell of it. I am not a big NO guy to begin with. I either play or come close to 6 figures in hands played each year. I have no worries about getting to the long run. I just strive to play as many hands in positive situations at the best games I can find, and still spread my play around. Everything else will work itself out.

 

[21forme]

It's N0 (Enn Zero), not NO (Enn Oh).

 

[blackjack avenger]

A Kewl NO errrr N0

Your anecdote is irrelivent. It could just as well work the other way, but, no matter, it's just short term results, and in the end will all work itself out if you play enough.

You agree there are different N0's, then say the anecdote in suport of the statement is irrelevant? I think it's very relevant, it shows the negative of 2 different bet ramps. The negative is what we are concerned about?

Keep in mind, that if I am willing to play that 'smaller' game, there must be something special that attracts me to it to begin with. Better rules, outstanding penetration, unusual side bet opportunity. I am not choosing to travel further and bet smaller just for the hell of it. I am not a big NO guy to begin with. I either play or come close to 6 figures in hands played each year. I have no worries about getting to the long run. I just strive to play as many hands in positive situations at the best games I can find, and still spread my play around. Everything else will work itself out.

N0 is more then just a long run number, it is also a measure of game quality.
800 hands at $25 to $400 is superior to:

400 hands at $25 to $400
400 hands at $25 to $200

The N0 is increased in the long and short run because of the different bet ramps.

Consistency of play is very important, there is even critic's of kelly resize betting over this issue because of the effect using different bet ramps; with wins and losses, has on the long run.

Yes, I agree with you, out in the world we can't always employ our optimal ramp.

This discussion make me think of another thread covering whether one should play a smaller EV but better quality game in terms of N0, SCORE, CE etc.

 

[blackjack avenger]

Not Very PC

It's N0 (Enn Zero), not NO (Enn Oh).

Are you teasing my fat NO's?:laugh:

Check the slimmed down N0 is more PC:joker:

 

[kewljason]

800 hands at $25 to $400 is superior to:

400 hands at $25 to $400
400 hands at $25 to $200

Yes but

400 hands at $25 to $400
400 hands at $25 to $200 (and welcome back to play each week)

is superior to

400 hands at $25 to $400
50 hands at $25 to $400 (and banned/backed off and at very least take this game out of rotation for a period of time, more likely being a smaller casino, a long period of time or maybe indefinately)

Simulations and math equations are all fine and dandy, but there is a matter of practicality. It's like a new counter looks at the single deck game at EC and says, geez, if I play this game 25 hours per week, spreading 1-8 in black I could make a real nice living without killing myself. :laugh: And if I had wings, I could save a lot on air travel. Gotta factor practicality and real world situations into these equations to make them work.

 

[blackriver]

Consistency of play is very important, there is even critic's of kelly resize betting over this issue because of the effect using different bet ramps; with wins and losses, has on the long run.

critic's of kelly
–noun
1. a person who doesnt understand kelly criterion
2. see gambler

 

[blackjack avenger]

Not So Much

critic's of kelly
–noun
1. a person who doesnt understand kelly criterion
2. see gambler

wiki
kelly criterion
see, reasons to bet less then kelly

Full kelly does not take into consideration preservation of capital. Also, Kelly does not consider human error!

Fully Kelly resizing:
10% chance of losing 90% of bank.
Long run increases 9 times vs flat betting

half Kelly resizing:
.1% chance of losing 90% of bank
Long run increases 1.87 times vs flat betting
Growth rate 75% that of full kelly

If playing full Kelly does losing 90% of bank stop one's playing due to table minimums or forces one to play inferior lower level crowded games? If so, then Kelly is not doable in the real world! Eventually, table max's force one to bet less then Kelly.

I believe most pros bet less then Kelly as I believe the MIT and Hyland teams did. As a bank grows or with larger banks wouldn't one want to be more conservative?

The number of hands needed to assure Kelly resizing overcomes flat betting is staggering, it's longer then many playing careers!