Re: A5s in blind battle.
 

Okay I’m wired so I thought I’d take a shot at revising Baltostar’s prose to make it more palatable for the masses.

I’m taking a break from finishing up a paper too, so this is kind of easy and fun.

[ QUOTE ]
I can't help it if I notice serious flaws in mechanisms of thought that have become de rigeur in the poker community.

[/ QUOTE ]

I’ve noticed glaring mistakes in the thought process of many players.

[ QUOTE ]
The common pattern of ignoring relative stack risk when deciding to play across an event has been bothering me for nearly all of the two years I've been studying poker.

[/ QUOTE ]

Players often ignore the amount of risk involved in their decisions.

Also:

How the hell do you “play across” an event? Why are you deciding to do this?

[ QUOTE ]
And yet players routinely use cost-to-call to calculate implied odds given across event risk, compare the result to implied odds required (typically also mis-calculated), and base their decisions on it.

[/ QUOTE ]

Here you’re being so needlessly verbose that one has to assume it’s intentional. I just can’t figure if it’s for comedy or if you’re just on shrooms.

I think you’re trying to say the following:

You can’t just take the cost of a call and compare it to your stack size to come up with an accurate picture of your real risk/reward.

This is true and has been pointed out many times on 2p2. The fact that players commonly overestimate their true implied odds is nothing new. You should probably drop the whole I'm-on-some-revolutionary-[censored] vibe.

[ QUOTE ]
For those decisions (only) where probability of achieving the most desirous outcome is *primarily* dependent on event risk, I recommend basing implied odds calculations on the total hand risk your stack is incurring. This too is an imperfect tool, but it's better than what most players are doing.

[/ QUOTE ]

Before you use terms like “total hand risk” you should define them! Why? Because no one knows what the hell you’re talking about.

Also, when you say “the total hand risk your stack is incurring” - is it your hand or your stack that’s incurring the risk? WTF?

[ QUOTE ]
If a decision criteria for playing across event risk is to be useful, it should not incur radical swings in validity when successively applied to similar scenarios.

[/ QUOTE ]

This is a one-way conversation that everyone here is a part of.

[ QUOTE ]
In my min-re-raises example, a player is offered a sequence of propositions

[/ QUOTE ]

No he’s not.

[ QUOTE ]
each of which is logical to accept according to his criteria for playing across event risk.

[/ QUOTE ]

Awesome.

Re: A5s in blind battle.
 

Hey robertjohn, thanks for this.

Baltostar, your idea of

"If a decision criteria for playing across event risk is to be useful, it should not incur radical swings in validity when successively applied to similar scenarios."

Doesn't actualy aply in your min-raising battle example.

You assume that our criteria is simply "cost-to-call" so that in your minraising example we would just stack off with a drawing hand. But thats not actually what we are saying.

We say, consider cost to call, if you are closing the action, that's all you need to think about.

If you aren't closing the action, make an educated guess (we are allowed those in poker, and most of us are quite good at them) as to what chance you have of actually seing a flop.

So in your min raisng example, the first time it would come around we'd go, hmm cheap call, I have the odds, chance of guy behind me re-raising is reasonably low, and I have good odds, lets call.

The second time around, we'd go, WTF, these guys are stuck in some kind of min-raising battle, I don't want to be in the middle of it, my immediate cost to call is low, but I'm not closing the action, and my chance of seeing a flop cheaply has just gone way down, I'll fold.

If you don't get it the second time, you'd definitely get it the third and only an absolute moron would stack off using our criteria here.

SO our criteria does work in your min-raising example, as people have already pointed out. As this seems to be the only thing you argument is relying on, you really need to respond to this specifically.

Re: A5s in blind battle.
 

I'm surprised noone has said this before but the most obvious counter point that I can come up with is a direct example.

Using the hand from the other thread as a base, you have a 10000 stack with blinds at 200/400 and raise 3bb from UTG with QQ. UTG+1 who you have a read only ever minreraises with AA and will commit postflop, minreraises you and has you covered. It's folded back to you and it costs 800 to call.

This is a trivally easy call based on your reads as you're risking 800 to win the 3800 already in the pot + 8000 more when you flop your set and double up, offering you combined implied and pot odds of ~15-1. However using baltostar's theory your risk is actually 10800/2000 or ~5-1 so it would be a fold.

This is obviously a very artificial example but it seems the simplest way of pointing out the major flaw in baltostar's thinking.

Re: A5s in blind battle.
 

baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

Re: A5s in blind battle.
 

[ QUOTE ]
baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

[/ QUOTE ]

On the avg hand risk distribution curve, if your strategy involves adding variance to move your mean to the right (because right-skew comes along with the variance), then better quality variance is an achievable shape that better improves your mean.

I think that the flood of similar-styled players is causing the tails to get too long and fat and is reducing the amount of skew. I think that if you avoid pursuing lines that tend to scale to allin in the most marginal perceived cEV+ scenarios that you can slim-down the tails and get back some of the skew.

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

[/ QUOTE ]

On the avg hand risk distribution curve, if your strategy involves adding variance to move your mean to the right (because right-skew comes along with the variance), then better quality variance is an achievable shape that better improves your mean.

I think that the flood of similar-styled players is causing the tails to get too long and fat and is reducing the amount of skew. I think that if you avoid pursuing lines that tend to scale to allin in the most marginal perceived cEV+ scenarios that you can slim-down the tails and get back some of the skew.

[/ QUOTE ]

LOL! Ask and you shall recieve!

Re: A5s in blind battle.
 

Whoa holy ish! Baltostar is STILL TALKING! cuz all I can see is

[ QUOTE ]
*** You are ignoring this user ***

[/ QUOTE ]

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

[/ QUOTE ]


On the avg hand risk distribution curve, if your strategy involves adding variance to move your mean to the right (because right-skew comes along with the variance), then better quality variance is an achievable shape that better improves your mean.

I think that the flood of similar-styled players is causing the tails to get too long and fat and is reducing the amount of skew. I think that if you avoid pursuing lines that tend to scale to allin in the most marginal perceived cEV+ scenarios that you can slim-down the tails and get back some of the skew.

[/ QUOTE ]





http://www.gamerevolution.com/oldsit...acy_record.jpg

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

[/ QUOTE ]

On the avg hand risk distribution curve, if your strategy involves adding variance to move your mean to the right (because right-skew comes along with the variance), then better quality variance is an achievable shape that better improves your mean.

I think that the flood of similar-styled players is causing the tails to get too long and fat and is reducing the amount of skew. I think that if you avoid pursuing lines that tend to scale to allin in the most marginal perceived cEV+ scenarios that you can slim-down the tails and get back some of the skew.

[/ QUOTE ]

Oh for ffs. Enough already.

Re: A5s in blind battle.
 

[ QUOTE ]
Baltostar, your idea of

"If a decision criteria for playing across event risk is to be useful, it should not incur radical swings in validity when successively applied to similar scenarios."

Doesn't actualy aply in your min-raising battle example.

You assume that our criteria is simply "cost-to-call" so that in your minraising example we would just stack off with a drawing hand. But thats not actually what we are saying.

We say, consider cost to call, if you are closing the action, that's all you need to think about.

If you aren't closing the action, make an educated guess (we are allowed those in poker, and most of us are quite good at them) as to what chance you have of actually seing a flop.

So in your min raisng example, the first time it would come around we'd go, hmm cheap call, I have the odds, chance of guy behind me re-raising is reasonably low, and I have good odds, lets call.

The second time around, we'd go, WTF, these guys are stuck in some kind of min-raising battle, I don't want to be in the middle of it, my immediate cost to call is low, but I'm not closing the action, and my chance of seeing a flop cheaply has just gone way down, I'll fold.

If you don't get it the second time, you'd definitely get it the third and only an absolute moron would stack off using our criteria here.

SO our criteria does work in your min-raising example, as people have already pointed out. As this seems to be the only thing you argument is relying on, you really need to respond to this specifically.

[/ QUOTE ]

Good analysis Jammy.

The point I'm trying to make is not that we should throw away cost-to-call in favor of total-cost when calculating implied odds, or that we shouldn't involve other considerations. Yeah, if they min re-raise each other it's a probably a good read that they're both trying to suck each other in, and we should bail.

What you want out of an implied odds calculation given is an decent idea of whether the relative reward/risk is worth it (relative to other opportunities).

When my goal is to either hit the flop hard or abandon, I find it useful to calculate implied odds given based on total cost.

It's not a strict rule, just a guideline, just as calculating based on cost-to-call should also be.

Someone else mentioned that if you know that his min-raise means AA, and you know he'll stack, then using total-cost to calculate implied odds is useless. That's true. But I don't recommend using total-cost if you can make that additional read.

I find calculating implied odds based on total-cost is useful to prevent you from whittling down your stack-utility on sub-par set/FD/SD-mining opportunities.

Many times in a tourney, I've calc'd (and re-calc'd if raised) implied odds on many hands and then I felt I had to make the call because it was correct according to the math. Then at some point I notice my stack-utility has been frittered away and I wish I'd only paid the minimum for each set/FD/SD-mining opportunity.

I'm just trying to figure out ways to avoid getting yourself in trouble, to protect yourself from yourself.

Re: A5s in blind battle.
 

[ QUOTE ]
I'm just trying to figure out ways to avoid getting yourself in trouble, to protect yourself from yourself.

[/ QUOTE ]

step 1: listen to people that have won money at poker
step 2: ?????
step 3: profit

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
baltostar, at one point you recommended that someone take an intro probability course
as I have taken an intro probability course, can you explain to me what 'better quality variance' is?

[/ QUOTE ]

On the avg hand risk distribution curve, if your strategy involves adding variance to move your mean to the right (because right-skew comes along with the variance), then better quality variance is an achievable shape that better improves your mean.

I think that the flood of similar-styled players is causing the tails to get too long and fat and is reducing the amount of skew. I think that if you avoid pursuing lines that tend to scale to allin in the most marginal perceived cEV+ scenarios that you can slim-down the tails and get back some of the skew.

[/ QUOTE ]

I dont know what the 'avg hand risk distribution curve' is, but ignoring that, I can't tell whether you think that adding variance to move your mean to the right (presumably you're talking about increasing profit here) is a good or bad strategy. To clarify, it is in fact a good strategy. I also don't know what 'an achievable shape' means, so I dont know whether you think added variance that adds to your expected profit counts as better quality variance or not.

basically I have no idea wtf you're talking about, shocker.

also adanthar, your steps suck. most people who win money at poker are terrible at giving advice (and terrible at poker)

Re: A5s in blind battle.
 

btw for a TON of people, drastically lowering variance, even if it reduces your profit by 10-30% will be hugely +EV in poker and in life. There are many reasons for this, but I strongly believe it to be true. I have no idea if this is what people are talking about, as I refuse to read through all the neverending posts, but maybe seems like it's sort of being discussed?

Re: A5s in blind battle.
 

that's actually a really good point, and should probably be a new thread, but no I'm almost positive that's not what's being discussed

Re: A5s in blind battle.
 

[ QUOTE ]
that's actually a really good point, and should probably be a new thread, but no I'm almost positive that's not what's being discussed

[/ QUOTE ]

Ok I can't figure it out, its too long [img]/images/graemlins/smile.gif[/img] Also I can never understand what baltostar is talking about.

Re: A5s in blind battle.
 

[ QUOTE ]
btw for a TON of people, drastically lowering variance, even if it reduces your profit by 10-30% will be hugely +EV in poker and in life.

[/ QUOTE ]

curtains, I don't know what exactly you mean by 10-30% drop (in ROI?), but I'm not sure at all that most winning mtt players (specifically online players who play tons of tourneys), will accept that offer (if winning would drop a lot less, sure I might consider it). I don't play massive volume and I'd not accept it.

This might sound absurd to you, but for me, and I'm sure that for many others, the ups and downs of poker and the struggle that come with them, are big part of what makes poker (and other forms of lets call it "intelligent gambling") so appealing.

I don't know, maybe I see it this way cause I had a v big cash recently, that was so amazingly satisfying. And it was worth going through a lot of variance, to "get there" (edit: no, it didn't change much, as I'm sure was thinking the same way before that).

Also there was a big thread about this subject in hsnl, maybe a year and a half ago, interesting stuff, but most of it has to do with people's psychology, not a lot more.

Re: A5s in blind battle.
 

[ QUOTE ]
It would be absolutely hilarious to watch baltostar play a limit holdem tournament.

He'd limp the button with JTo, the bb would raise and he'd fold because he's risking TWICE AS MUCH OF HIS STACK AS BEFORE.

[/ QUOTE ]

Don't be ridiculous, gobbo. The bound of expectation for JTo OTB is OBV 1.5 BB. Thus, he would RAISE JT on the button. Then, if the BB 3-bets, the pot is > 1.5 BB, outside the boundary, into the murky waters of OMG too much risk, so then you fold. Jeez.

Edit to add: And the good lord help you if the BB TRIPLE 3-BETS DUH DUH DUHHHHHHH

Re: A5s in blind battle.
 

[ QUOTE ]
I dont know what the 'avg hand risk distribution curve' is, but ignoring that, I can't tell whether you think that adding variance to move your mean to the right (presumably you're talking about increasing profit here) is a good or bad strategy. To clarify, it is in fact a good strategy. I also don't know what 'an achievable shape' means, so I dont know whether you think added variance that adds to your expected profit counts as better quality variance or not.

basically I have no idea wtf you're talking about, shocker.


[/ QUOTE ]

NHFunkii : if I had some pictures, I bet you'd find my argument easy to follow. It's not advanced stuff. My bad for not having pics.

"Avg hand risk distribution curve" is easy to picture: play 100,000 tournament hands and graph the magnitude of gain/loss (x axis) against the frequency each magnitude of gain/loss occurs (y axis).

We're in total agreement that the principle way aggressive players get such an advantage over donks and tight players is by adding variance to the curve in an advantageous manner: they scale-up risk when they perceive cEV+.

Compared to a tight player's avg hand risk distribution curve, the aggressive player's curve is squashed down and spread-out and skewed to the right (the mean is positive).

So, yes, I agree with you, adding cEV+ variance is good.

However, as a flood of similar-styled aggressive players enter the game, when the aggressive player picks a line that tends to scale risk, more and more often it's against another aggressive player, and so your FE doesn't work as hard for you -- the aggressive guys just don't lie down as much as the donks and the tight players.

This is fine in significant cEV+ scenarios. But in marginal perceived cEV+ scenarios, it's very easy to be a little off in your cEV+ perception and, without the FE working as hard for you, you tend to kill off a lot of stack-utility.

The flood of similar-style aggressive players flattens your curve, and because your FE is blunted it causes the curve to lose some of its rightwards skew.

How to combat this ? By pursuing a curve that has somewhat less variance, but better quality.

You still have an awful lot of variance compared to the tight players, but you have less than the avg aggressive player.

If you can avoid scaling risk in the significantly sub-par opportunities then, on avg, when you do scale risk it doesn't matter so much that your FE isn't as powerful a weapon as it once was. You don't mind getting it allin when you have good confidence that you're ahead.

So, your flat squashed curve perks up a bit, it transfers some fat from its tails to its midsection, and it adds-back some rightward skew. You have a "better shaped" curve than the avg aggressive player: it has less variance and a more positive mean.

Re: A5s in blind battle.
 

[ QUOTE ]
So, yes, I agree with you, adding cEV+ variance is good.

[/ QUOTE ]

Unreal. Well, done, NHFunkii. You made him say that.

Yeah, I know that then come 7 more paragraphs with same old imbecilic notions about "flood of similar-styled aggressive players", but who cares.

[ QUOTE ]
if I had some pictures, I bet you'd find my argument easy to follow. It's not advanced stuff. My bad for not having pics.


[/ QUOTE ]

You should really bring them pics, baltostar, this thread is worthless without pics.

Re: A5s in blind battle.
 

Im pretty sure we're just talking about this...

"I will say one other thing, and I think it may be the point you are getting at. For a long time, players in MTTS by and large were bad in a very specific way. They were weak tight, especially when faced with a decision for all their chips. Therefore playing in a manner which increased your own varience almost definitionally increased your EV. That in my opinion is no longer the case. That doesnt mean that you should turn down EV ever, it just means that increasingly the most EV strategy may not be the most aggresive one."


which, you know, i said a while ago.

Re: A5s in blind battle.
 

[ QUOTE ]
Im pretty sure we're just talking about this...

"I will say one other thing, and I think it may be the point you are getting at. For a long time, players in MTTS by and large were bad in a very specific way. They were weak tight, especially when faced with a decision for all their chips. Therefore playing in a manner which increased your own varience almost definitionally increased your EV. That in my opinion is no longer the case. That doesnt mean that you should turn down EV ever, it just means that increasingly the most EV strategy may not be the most aggresive one.

which, you know, i said a while ago.

[/ QUOTE ]

Ok, but I think for most highly successful tournament players, aggressive play is the preferred style, what they're good at, and what gets them the best $/hr, because it's the best way to take advantage of risk-averse players, who are still the big majority.

So, if you are an aggressive style player, I don't see how you can scale-back aggression without passing up some perceived cEV+.

What makes a cEV+ line in a marginal spot cEV+ is the aggressive use of FE : bluffs and semi-bluffs.

If you scale back the aggressive use of FE for those lines, then they're definitely not cEV+.

In the marginal perceived cEV+ spots, you either have to be willing to max scale the risk (allin), or not pursue them at all.

Avoiding the significantly sub-par cEV+ spots is the only way I can think of to scale-back back aggression and improve your results.

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
btw for a TON of people, drastically lowering variance, even if it reduces your profit by 10-30% will be hugely +EV in poker and in life.

[/ QUOTE ]

curtains, I don't know what exactly you mean by 10-30% drop (in ROI?), but I'm not sure at all that most winning mtt players (specifically online players who play tons of tourneys), will accept that offer (if winning would drop a lot less, sure I might consider it). I don't play massive volume and I'd not accept it.

This might sound absurd to you, but for me, and I'm sure that for many others, the ups and downs of poker and the struggle that come with them, are big part of what makes poker (and other forms of lets call it "intelligent gambling") so appealing.

I don't know, maybe I see it this way cause I had a v big cash recently, that was so amazingly satisfying. And it was worth going through a lot of variance, to "get there" (edit: no, it didn't change much, as I'm sure was thinking the same way before that).

Also there was a big thread about this subject in hsnl, maybe a year and a half ago, interesting stuff, but most of it has to do with people's psychology, not a lot more.

[/ QUOTE ]


btw has nothing to do with MTTs. Of course MTTs by nature are ridiculously swingly with huge variance. You can't avoid gigantic swings in MTTs no matter what you do.

It's more about cash games and stuff like playing lower limits, even though your win rate may be slightly less, but your quality of life and stamina to play will be higher. It's much easier to play more hours and better when you are winning, no matter who you are. There are very few exceptions to this rule.

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
Im pretty sure we're just talking about this...

"I will say one other thing, and I think it may be the point you are getting at. For a long time, players in MTTS by and large were bad in a very specific way. They were weak tight, especially when faced with a decision for all their chips. Therefore playing in a manner which increased your own varience almost definitionally increased your EV. That in my opinion is no longer the case. That doesnt mean that you should turn down EV ever, it just means that increasingly the most EV strategy may not be the most aggresive one.

which, you know, i said a while ago.

[/ QUOTE ]

Ok, but I think for most highly successful tournament players, aggressive play is the preferred style, what they're good at, and what gets them the best $/hr, because it's the best way to take advantage of risk-averse players, who are still the big majority.

So, if you are an aggressive style player, I don't see how you can scale-back aggression without passing up some perceived cEV+.

What makes a cEV+ line in a marginal spot cEV+ is the aggressive use of FE : bluffs and semi-bluffs.

If you scale back the aggressive use of FE for those lines, then they're definitely not cEV+.

In the marginal perceived cEV+ spots, you either have to be willing to max scale the risk (allin), or not pursue them at all.

Avoiding the significantly sub-par cEV+ spots is the only way I can think of to scale-back back aggression and improve your results.

[/ QUOTE ]


Dude I'm not retarded and nothing you say ever makes any sense, because it's written so poorly.

Re: A5s in blind battle.
 

[ QUOTE ]
Dude I'm not retarded and nothing you say ever makes any sense, because it's written so poorly.

[/ QUOTE ]

MLG figured it out. We just don't agree on the solution.

What am I supposed to do ? Risk professionals think in terms of distribution curves and how to achieve the most desirous shape.

When you add in that it's kind of hard to understand this stuff unless you've taken a bunch of classes in math, stat, and probability, it's really pretty challenging to explain it to a diverse audience, even a poker audience, no matter how smart they are, and you guys are probably the smartest.

I don't think I am explaining it that badly. Definitely improving. It's just hard to explain this stuff without graphs and equations. And if you go that route, then you can only explain it to a bunch of math-heads.

I give up. All the haters have it wrong. I have some real insight into why today's aggressive game is going downhill in profitability and I'm just trying to explain in best I can.

Re: A5s in blind battle.
 

[ QUOTE ]
I give up. All the haters have it wrong. I have some real insight into why today's aggressive game is going downhill in profitability and I'm just trying to explain in best I can.


[/ QUOTE ]

NOBODY IS A HATER. AND NO, YOU DO NOT HAVE SOME REAL INSIGHT INTO ANYTHING THAT HAS TO DO WITH POKER, YOU ARE AN IDIOT. YOU HAVE NO IDEA WTF YOU ARE TALKING ABOUT. YOU ARE LIKE AN INSANE HOMELESS PERSON WHO PREACHES TO PEOPLE ABOUT HOW THEY CAN MAKE MORE MONEY IN THEIR CAREERS. IT'S BEYOND ABSURD.

Also, there were some ridiculous self-contradictions in your last reply to MLG. I won't even bother to point them out because you'll reply with the same repetitive long winded BS.

Farewell.

Re: A5s in blind battle.
 

Serious question how are people still reading Balto's posts. Not in a LOLZ he is stupid ignore him way, but they are the most boring rambly incoherent rants I have ever seen on 2p2. Reading more than three sentences makes me want to smash my head against a wall.

Re: A5s in blind battle.
 

[ QUOTE ]
gobbo most villains will bet their midpair on the flop and rightfully so.

[/ QUOTE ]

I'm still having issues as to when this is anything BUT a c/c.

Re: A5s in blind battle.
 

[ QUOTE ]
[ QUOTE ]
Dude I'm not retarded and nothing you say ever makes any sense, because it's written so poorly.

[/ QUOTE ]

MLG figured it out. We just don't agree on the solution.

What am I supposed to do ? Risk professionals think in terms of distribution curves and how to achieve the most desirous shape.

When you add in that it's kind of hard to understand this stuff unless you've taken a bunch of classes in math, stat, and probability, it's really pretty challenging to explain it to a diverse audience, even a poker audience, no matter how smart they are, and you guys are probably the smartest.

I don't think I am explaining it that badly. Definitely improving. It's just hard to explain this stuff without graphs and equations. And if you go that route, then you can only explain it to a bunch of math-heads.

I give up. All the haters have it wrong. I have some real insight into why today's aggressive game is going downhill in profitability and I'm just trying to explain in best I can.

[/ QUOTE ]

I'm still an undergrad, but I'm majoring in stats at harvard and you are definitely explaining it terribly. Maybe you should go the 'math-head' approach and maybe a couple people will understand you, instead of no one