Stop-n-Ramble (long) (I\'m bored)
 

I made a post recently that got a lot of responses. Here's a copy of the hand:

[ QUOTE ]
Party Poker No-Limit Hold'em Tourney, Big Blind is t300 (4 handed) FTR converter on zerodivide.cx

saw flop|<font color="#C00000">saw showdown</font>

<font color="#C00000">Button (t3730)</font>
<font color="#C00000">Hero (t2670)</font>
BB (t1365)
UTG (t2235)

Preflop: Hero is SB with A[img]/images/graemlins/spade.gif[/img], Q[img]/images/graemlins/club.gif[/img].
<font color="#666666">1 fold</font>, <font color="#CC3333">Button raises to t1100</font>, Hero ??

[/ QUOTE ]

The responses were kinda split, but I'm not so worried about the particulars of that posts. I want to use it as an example of something I've been thinking about recently.

In the actual hand, I decided to 'sorta stop-n-go' (I say sorta because I don't close the action preflop. Anyways, I kinda hoped the BB might come along, and there were other factors in play, but for the sake of this post, let's pretend the BB is folding for sure, or that I'm in the BB. Basically, assume this is a true stop-n-go.

Anyways, I called.
Flop comes 9 [img]/images/graemlins/diamond.gif[/img]6 [img]/images/graemlins/diamond.gif[/img]4 [img]/images/graemlins/club.gif[/img]

I push. He calls with K [img]/images/graemlins/spade.gif[/img]9 [img]/images/graemlins/club.gif[/img]

I'm going to try and find out how much advantage, if any the stop-n-go here has over pushing preflop assuming he calls preflop everytime.

Okay, I've been sitting here for a little bit, and realizing that to do this justice, I'd have to spend about 25334221243435 hours analyzing this, so I'm gonna be simplistic. Please excuse me.

Let's make some rough assumptions:
1) I push every flop after calling.
2) He calls with a pair or better (unless it's the board that's paired)
3) He calls OESD or equally good or better draws.
4) He folds everything else.

Now I'm going to make up possible flops, and see if the stop-n-go has a better result than just pushing preflop.

Flop 9 [img]/images/graemlins/diamond.gif[/img] 6 [img]/images/graemlins/diamond.gif[/img] 4 [img]/images/graemlins/club.gif[/img]
No need for analysis. It's the same result either way for me. This is true for any time he will call a flop push.

Flop T [img]/images/graemlins/diamond.gif[/img] 6 [img]/images/graemlins/diamond.gif[/img] 4 [img]/images/graemlins/club.gif[/img]

SNG - I push. He folds. I win a 2500 chip pot. My $EV = .3373 (ICM Calc)

Push Preflop - We're already all in. 75.423% of the time, I win a 5640 chip pot. My $EV is .3938. 24.577% of the time, I bust, my $EV is 0. Total $EV = .2970.

Result Stop-n-Go wins.

However, you'll notice that in cEV, pushing preflop wins, netting an average 4254 chips vs. the sng's 4070. The reason the stop-n-go is better is because of the bubble. It's better to not take large gambles in many bubble situations. I hope you already knew that.

Flop Q [img]/images/graemlins/diamond.gif[/img] 7 [img]/images/graemlins/heart.gif[/img] 7 [img]/images/graemlins/club.gif[/img]

SNG - I push. He folds. I win a 2500 chip pot. My $EV = .3373


Push Preflop - We're already all in. 86.566% of the time, I win a 5640 chip pot. My $EV is .3938. 13.434% of the time, I bust, my $EV is 0. Total $EV = .3408.

Result Push preflop wins.

Suprisingly, to me at least, it barely wins. I actually came into this post thinking that the stop-n-go was overrated and might leave thinking the opposite.


Flop A [img]/images/graemlins/diamond.gif[/img] 5 [img]/images/graemlins/heart.gif[/img] 7 [img]/images/graemlins/club.gif[/img]

SNG - I push. He folds. I win a 2500 chip pot. My $EV = .3373


Push Preflop - We're already all in. 96.869% of the time, I win a 5640 chip pot. My $EV is .3938. 3.131% of the time, I bust, my $EV is 0. Total $EV = .3815.

Result Push preflop wins.

This is kinda obvious, but I wanted to run the numbers.

Conclusions:

I've taken one tiny example where I know the cards of the opponent, picked specific flops for testing EV. We haven't really learned much. Don't take the fact that pushing preflop was better in 2 of the 3 above cases to mean that it's the better play, as the first (second) case will occur much more often than the others. In addition, we can improve the stop-n-go if we alter it to check flops like the last one, inducing a push from the opponent and making the play more profitable.

One thing you may notice about the sng, that most of you already know, is that it becomes more effective when you are near the bubble, and when busting out would be especially bad. This hand is an example of that since there is a short stack, we're 4 handed, and have a healthy stack ourselves.

In cEV, the stop-n-go seems to often not be very profitable. I would like to say that against reasonable opposition, and not near the bubble, the stop-n-go is not THAT effective of a play. Often it's worse than just getting the money in preflop (very situation dependent though). And these are times when we assume no FE preflop. If you may have any FE preflop, and you aren't near the bubble, or at risk of busting or being crippled, opting to push preflop seems like it will often be the best play.

When the Stop-n-go really shines is when you can get the opponent to fold the best hand on the flop. This is not easy to do, but it certainly can happen. If, for instance, you have A [img]/images/graemlins/club.gif[/img] T [img]/images/graemlins/heart.gif[/img] and your opponent has 7 [img]/images/graemlins/spade.gif[/img] 7 [img]/images/graemlins/diamond.gif[/img], and the flop is: Q [img]/images/graemlins/club.gif[/img] K [img]/images/graemlins/heart.gif[/img] 8 [img]/images/graemlins/spade.gif[/img], the villain will often fold to a push, which would be HUGE for you, especially on the bubble.

So, don't read too much into this post. There are so many unique situations to look at that I couldn't figure out a way to quantify the play based on all them. I started this post and then 50% in realized I couldn't prove what I wanted to prove and almost just deleted it. I figure though, it's better to post it than not, even though it doesn't really make any conclusions. Worst case, someone will learn how to do ICM calcs who didn't already, or we'll get some discussion going.

Thoughts?

Re: Stop-n-Ramble (long) (I\'m bored)
 

Why are you getting different answers for pushing preflop? Isn't there one and only one answer to your EV for pushing preflop?

This is a calc I've wanted to do for a long time, but as we both discovered it's quite tricky.
eastbay

Re: Stop-n-Ramble (long) (I\'m bored)
 

[ QUOTE ]

Why are you getting different answers for pushing preflop? Isn't there one and only one answer to your EV for pushing preflop?

[/ QUOTE ]

I am doing the $EV after the flop assuming we already pushed preflop. It's basically (and by basically, I mean entirely) the same as saying he calls the flop push. I thought that would be a good way to see how the SNG fares on each individual flop.

So, I was taking the times that the stop and go yeilded different results than simply pushing preflop and seeing if those results were better or worse.

[ QUOTE ]

This is a calc I've wanted to do for a long time, but as we both discovered it's quite tricky.

[/ QUOTE ]

Understatement of the week.

Re: Stop-n-Ramble (long) (I\'m bored)
 

ok, here is my stab at getting the odds of him folding post flop (it's roughly 2-1 that he will fold using your assumptions)

now, i didn't do all the precise math (left out 1-gap, 2-gap,3-gap connectors but i also didn't say that if his 72 pairs the 2 on the flop that he'd fold, so hopefully it'll close to even itself out [img]/images/graemlins/wink.gif[/img] )


suited:

odds he has suited pf: .24
odds he hits 2 more of that suit: .109
odds he hits flush on flop: .0084

.24(.109)+.24(.0084)=.02616+.002016=.028176

so 2.8% of the time this occurs


pocket pair:

odds he has pp pf: .059

odds he improves to set or better on flop: .118

.059(.118)=.006962

so .7% of the time this occurs


let's also say that the (1-.118)% of the time that 99+ doesn't improve, he still calls your s-n-g

.027(1-.118)=.023814

so 2.4% of the time this occurs


connectors that can make straight fully in either direction (54 - jt):

odds he has this pf: .085

odds it improves to straight on flop: .013

odds it becomes oes: .096

.085(.013) + .085(.096) = .001105 + .00816 = .009265

so .9% of the time this occurs


the rest of the time:

odds of holding pf: .616

odds of improving to a pair on flop: .29

odds of improving to two pair: .02

odds of improving to trips: .0135

odds of improving to boat: .0009

odds of improving to quads: .0001

.616(.29) + .616(.02) + .616(.0135) + .616(.0009) + .616(.0001)= .17864+.01232+.08316+.0005544+.0000616=.274736


total odds of not folding on flop: .342953

odds of folding on flop: .657047


so roughly 2-1...

Re: Stop-n-Ramble (long) (I\'m bored)
 

Nice calcs.

We're still leaving a lot out though. Not that I want you to figure it in. Bet size v pot size matters a lot. He will often call with A high hands on the flop, which he is fairly likely to have. We can't assume he'll always fold overs like in my initial post (although I think that was a fair assumption in that hand).

Re: Stop-n-Ramble (long) (I\'m bored)
 

well let's do some analysis using 66%, even tho it is inaccurate...we'll assume it's decently close though


equity if you fold pf: .2682


stop-n-go:


66% of the time he folds to your push, so you gain 1550 chips over folding, ev= .3373

33% of the time he calls because he has something: let's say you win 33% of these and lose 66% of these:
you win: you now have doubled up this hand (plus 300 bb), so your ev = .3938
you lose: ev = .0

.66(.3373)+.33(.33)(.3938)=.222618+.04288482=.2655

so it's slightly negative ev to do this instead of just folding


so, how strong of a hand do u need to make this positive? well you need to win more than 1/3 of the times that your opp hits a calling hand (which could be a drawing hand, so pps should win more than 1/3...)


.2682-.222618=.045582= what you need ev*p(w|c) to be

.33(x)(.3938)=.129954x
.129954x=.045582
x=.35075

so you need a hand that is &gt;35% after the flop against the range of hands he's calling (oes, flush draws, pair, two pair, boat, etc)


so that's the next step...figuring out what pf holdings will be &gt;35% against a random hand after the flop, given that the random hand hits something that meets your criteria

Re: Stop-n-Ramble (long) (I\'m bored)
 

Nice work man.

[ QUOTE ]

so that's the next step...figuring out what pf holdings will be &gt;35% against a random hand after the flop, given that the random hand hits something that meets your criteria

[/ QUOTE ]

I think we have to be careful. Most raising hands aren't random.

Re: Stop-n-Ramble (long) (I\'m bored)
 

[ QUOTE ]
Nice work man.

[ QUOTE ]

so that's the next step...figuring out what pf holdings will be &gt;35% against a random hand after the flop, given that the random hand hits something that meets your criteria

[/ QUOTE ]

I think we have to be careful. Most raising hands aren't random.

[/ QUOTE ]

Essential point.

eastbay

Re: Stop-n-Ramble (long) (I\'m bored)
 

yep, it has to be a better than random hand to meet the criteria for a s-n-g, cause a s-n-g is when they raise and you cold call then push on flop...so only hands that they will raise with should be considered

wish we could simulate this...we know how a range of hands will do against us from pf to the river but we don't know how it will do against our hand after the river only...


but can we boil this down to only 2 broad scenarios in which you are &gt;35% after the flop?

you are &gt;35%:
you are ahead after the flop
you are have 9+ outs to beat him

Re: Stop-n-Ramble (long) (I\'m bored)
 

let's define a range of raising pf...then we can do the calcs on those hands only for improving, becoming drawing hands, etc

22+,ax,k7+,2 broadway?

what range should we use