Correlation

Definition: Correlating - Attempting to maintain a lead over an opponent by betting an amount equal or similar to his bet.

The idea behind correlation is to obtain similar results as your opponents. When the dealer busts the entire table wins and conversely when the dealers makes a good hand the table loses.

In a perfect world by correlating bets most players at the table would have very similar results heading into an elimination hand (EH1 or EH2). Given the normal variance, along with double down and splitting opportunities the player results should be close (with ¼ max bet separating BR1 from BRL) this assumes that the entire table correlates with the initial betting that was placed with the button.

So my question is this: Why don’t more people correlate their bets leading up to EH’s?

 

Some thoughts...

I WISH everyone correlated perfectly. Then I would do completely the opposite of what the whole table is doing, but also factoring in risk of ruin. I'll try to explain my reasoning and you can tell me if you agree. If everyone has similar results, then the dealer can either help or hurt everyone pretty much equally when bets are the same. Think about this. If everyone is betting low and the table wins, everyone is about in the same postion. Ok so there are still 6 other people who have equal chips as you if you correlated.

Youre at no advantage or disadvantage by raw odds. However, if you go against them and bet high, you are at an advantage compared to the correlating rest of the table. They all won low bets and you won big. In the same situation, if the table lost, they lost minimum bets and youre at a big disadvantage to them because you lost big. But so what, all other things being equal, you probably weren't going to win anyway. There are 2 ways the table can go and both are just about equal. Dealer dumps the chips out or takes them away. If you go against everyone else, you have about a 50% chance of standing alone and being ahead of all others and a 50% chance of standing alone and being behind all others. If you correlate, you have a 50% chance of being ahead with all others and a 50% chance of being behind with all others. I prefer to go against the pack and if the table goes MY way (50% of the time) I will be ahead of everyone. If the table goes the table way-if youre betting low youre hoping dealer is hot (50% of the time) I will be behind. I think your overall odds would be better to go against the rest of the table.

My favorite situation is where everyone bets big such as the freerolls on UBT. I bet minimum and win about 1 in 3 tourneys I think. The raw odds are 1 in 7. If dealer beats up on everyone, I stand alone and the tourney is over quickly. If everyone wins, I am behind, but still have a slight chance of catching up. I let them all play until I feel like I have to take chase. If its a house game, I have no one to battle it out with in the end if I bet low. If its a player game, I have everyone to battle it out with if I bet high (which I dont its not worth it). I'm not sure if what I'm saying makes sense, but I would like others' opinions on this.

My 2 cents,
Rounder21

ps, also my question would be at what point does ror become important? By my logic, at the B&M casinos where I play everyone bets pretty low and at times I have went all in on the first hand. Table win and I'm way ahead of everyone. Table loss and I'm out. I have not played this way in a while because I think its just too risky with a 50% ROR even though I'm betting contrary to the pack, I dont like that huge ror in the beginning but I'm still not sure what approach to use, so now I just bet according to the count until hand 20 unless someone has jumped way ahead which is very rare.

 

Good points...

Yea there are a few big bettors at the B&M casino I play, but most bet pretty low. At least at the tables I've played at.

Good Cards,
Rounder

 

rounder21,

I don't want to discuss your whole post but just wanted to point out a few things. BJ is not a 50-50 game. Although the house edge is only 0.5% when perfect basic strategy is applied, that takes into account the "extra" money a player wins on BJs, double downs or some splits and the reduced losses for a player on other splits. Depending on the game, the numbers generally run like this:
Player wins = 44%
House wins = 48%
Pushes = 8%

Now the question is: On hands where a decision is made (no push), how often will I win and how often will I lose?
Player win percentage if we exclude pushes = 44 / (44 + 48) = 48%
House win percentage if we exclude pushes = 48 / (44 + 48) = 52%



The next point I'd like to make is the fallacy that "most of the time everyone wins or everyone loses on a given hand". I play a lot more live BJ than the average player and I can tell you that when you have 7 players it's a rarity when all 7 players win or lose on the same hand. When you have 7 players that is of course a odd number so you almost always have a situation where the majority will win or lose (pushes make for an exception) but rarely will they all have the same results. That rarity does not mean it is an anomaly but rather "not the norm".

Your methods may work against the "Freebie Ploppies" (I think I just coined a new phrase ) but in B&M casinos were people pay money to play, well all I can say is good luck.

 

new words

toolman - I think your new BJ term could be shortened to a single word, either "froppies" or "pleebies". What do you think?

The use of correlation in TBJ is typically a tactic to maintain position with 1 or several players. Most people associate it with maintaining a lead but I often use it when behind to hold my position until a more appropriate time to make a move for 1st. I read (over a LVA) that Blair Rodman uses correlation to catch up cheaply and from the other posts in the thread I'm guessing that this is another accepted tactic. So it seems that correlation betting can have many uses and is not just restricted to maintaining a lead.

Fgk, for correlation as you describe it to be used three conditions must exist. First, you must have the lead (or at least not be last). Second, you must act after those with whom you must correlate. Lastly, they must make a bet that it is wise to correlate with.

Here's a question, mid-game, TTBJ, you are BR1 by smallish amount, BR3 (slightly behind BR2) acts before you and bets max, BR2 bets 1/2 max, what do you do?

Cheers

Reachy

 

Too clumsy and hard to remember. Actually I'm working on creating a series of 2 word phrases to describe various things a ploppy is capable of doing. The new one you referred to just came to mind while I was typing my response and it will be added to the list. More to come at a later time.

 

Correction is in O

rder!

The definition of correlation is from the terms at playubt.com! So ask THEM! I just do a lot of plagerism!

Freebee ploppies = Freeppies (kinda sounds like a drink at Starbucks)

As far as tactics - I want to hear from other people. I'm just a simple ploppy that needs to learn what the good players are doing!

As far as your scenario is concerned - good question. I mean really good question. I hope other people reply with what they would do:

Quantifiers:

Not enough information for me to answer. Additional information:

BR at this time. If a Max bet = BR then I play differently than if max bet is 1/10 of BR
How many rounds to go before final 3 hands?
Are there any OTHER players or is it just use 3 playing?
Is this a live or online tourney - this relates to playing a shoe versus a RNG - yes I play the two differently
betting styles of my opponents. Do I have profiles on these players and therefore think I can anticipate their likely actions
Is surrender available?

WOW - and to think that in the past I would just answer XXXX. NOW look what's happened! You all have got me thinking about so many OTHER factors that I hear in the background "10 seconds...9....8....7...6...5...4...3...2...1.. time" - Oops I guess it turned into a min bet

 

General

fgk - Assume Global rules, typical Global players and BRs around about starting BR. I'm not after specifics really more a general idea about your thoughts on what correlation actually is. Lets add another player BR4 into the mix who minimum bets, now what do you do? There is less than a 1/2 max bet between BR1 and BR4.

Cheers

Reachy

Ps. What does rder mean?

 

I left the O on the t

itle line - see?

In the strictest definition of correlation that as BR1 you should max bet also.

Would I?

Probably not, especially when the max bet = BR. I mean why give up now? (assuming this is one of the 48% of the hands the dealer wins?)

According to Wong, page 41, With 10 hands to go and you are BR1 then bet with the flow.

This is an area that is SEVERELY lacking in good "rules". Most all of what is written applies to the last 1-3 hands. In this case, middle of the tourney, BR2 bets max, BR3 bets 1/2 max and BRL bets min. So what is the "flow"?

I would take an average of the table's bets, in this case 1/2 max and bet that. It's not a correlation but once again it gives me options.

At a live table a lot would depend on where we were in the shoe and the general feel of the cards at that time. (can't reveal all my secrets!) So at a LIVE event I would likely be either max or min.

Online probably 1/2 max.

 

I would s

ay that a 1/2 max bet is a correlation with the table. Incidentally starting BR at Global is 2X max bet so.....

You could also consider betting somewhere between BR2 and BR3s bet, maybe taking the low against BR3 which would also give you the low against BR1. OK lets put some numbers to this and see what comes out (Betting 500-10, Surrender etc...)

BR2 1100 Bets 500
BR4 980 Bets 10
BR5 970 Bets 10
BR3 1080 Bets 250
BR1 1130 Bets ????

Correlate that lot

Cheers big ears

Reachy

 

Using Reachy's numbers

AND assuming that I decided correlating with BR2 & BR3 is my play on the hand, then I would bet it this way:

In the middle hands, say 10 or more to go in the round, I would bet around 200. This gives me the option to surrender and still be NO WORSE than BR3 assuming a swing by both BR2 & 3. I also have the possibility of splitting or DD and being BR2 on an all win. In other words, this bet keeps me in the game under a lot of different scenarios.

In the late hands, 5 or so to go, my bet would be closer to 300. Any successful DD or split leaves me in the lead and BJ leaves me very close. Even on a surrender, I may be BRL, but I'm still in the game. Same as before, this bet won't cripple me, win or lose, under a lot of possible outcomes. But, with fewer hands left to play, a little more risk is called for.

Last 3 hands. Max bet or very close to it. Time to fire my bullets.

 

Guys...I'm a newbie at this stuff but always reading and (most times) learning from the posts.

Using Reachy's numbers above BR1 cannot correlate all players.Would it not make sense to give a little more weight to correlating BR3's betting pattern as he is the player who is betting immediately before you 4 out of 5 hands?

If so,I'd bet $240 which allows the following....

a double down to stay ahead of BR2
surrender to remain BR3 or better
the high and low on BR3(see above)

I know I'm probably missing something so if someone could post the negatives of the $240 bet,I'd appreciate it.

Thanks

 

In my experience, correlating is not the Godsend that some believe it to be. I've been at many tables with the lead then started correlating only to lose the lead and the table with it. It happend to me most recently at the semi-final table at the Palms UBT quailfier. I'm not saying correlating is bad, just be aware that it can easily backfire.

Food for thought.

 

Question with a question

fgk asked a question in his first post; why don't more people correlate. My response, I suppose, is that it can be difficult to do for many reasons and I used the scenario I outlined above to demonstrate that. How can one possibly correlate a table with a wide rnage of different sized bets? This is not an unusual scenario and my answer would be, act in a way that maximises your chances of maintaining your postion. So in this case I would examine each of my opponents in turn and look for a bet that is flexible enough to deal with as many eventualities as possible

The following bets by BR1 will cover different scenorios for each of the other players:

BR2 - 500-480 High low; 260-240 DD high low; 20 Push low; 50 Surr push low; 270 Surr lock-out; 500 Surr Surr lock-out

BR3 - 290-210 High low; 150-100 DD high low; 40 push low; 90 Surr push low; 160 Surr lock-out; 320 Surr surr lock-out

BR4 - 130-10 High low; 60-10 DD high low; 130 Push low; 270 Surr push low; 140 Surr lock-out; 280 Surr surr lock-out

BR5 - Same as BR4 for the sake of arguement

So if we look for overlaps, 260-240 gives you the high low over BR1 with a DD as well as the straight high low over BR3. 270 gives you the surrender push low over BR4 and BR5 and therefore so would 260-240. You can also lock-out various other outcomes with surrender. My bet would therefore fall in that range, probably around the upper end.

Cheers

Reachy

 

I got my knuckles rapped a while back for using 'correlation' too loosely - using it to mean any bet that was based on a relationship with another player's bet - it was decided that 'coordination' was a better term for betting in relationship to another player's bet - and 'correlation' was to be used only when you were matching another bet (or a near match) -

one problem with the freerolls, cheapo games, and such - is that the ploppies bet all over the place - without reason - such as in Reachy's example - and you can't make a rational bet that covers everyone - and you know that one player will always get a result that will screw you - you just don't know which one -

my approach is to try to find a bet that gives you something with everyone - it may correlate with BR2, give you a 1/2 of BR3's bet - be 50% of your lead with BR4, etc. - I look for the 1/2 of lead, either way, - correlation, - full lead-1 min bet - etc. - something that makes some kind of sense vs every player -

or - I just pick a player to play against - and that's it - if ebj - then pick a player that will bet before you on the next elim hand - and - if possible - who has used his secret bet -

or - just ignore the extreme betters - trust to luck that the max bettor will lose his bet - or the min bettor will see everyone else win -

or - construct a synthetic high/low - bet so you have the second high if everyone wins, and the second low if everyone loses - this is the best approach when possible

 

Wondering about this...

Why is this best RK?

Rounder21

 

huh

rounder - I am not sure -

I guess I was assuming that this would be a two advance table - but the way the info was given - it wasn't even stated if this was a final hand - or what - just - how would you correlate?

so what you are trying to achieve would depend on the situation - last hand? next to last? - two away from last hand? - two advance? - single advance?

in many situations - having the second high if everyone wins and the second low if everyone loses - can give you a good result if it is two advance - and also may set you up for the next hand - leaving you BR2 no matter what happens - and may give you options for surrender to take the absolute low - or double/split for the absolute high -

even stronger if you can set up a synthetic high/low -

but define the hand/situation better - then the objective becomes clearer

 

I see...

Wong also says middle is best on last hand when 2 advance and 3 people are left. 2nd high and 2nd low like you suggested. I just thought you were suggesting this being the best approach in other situations thats why I was curious. Thanks for the clarification.

Good Cards,
Rounder21