Sunday, March 02, 2008

Micro Stack SNG Bubble Play - Part #3 - Folding??!?

Time for part #3 of my series on the options we have to play a 'micro-stack' of less than 2BBs at (or close to) the SNG bubble.

If your opponents undertsand the mathematics behind SNG bubble play (ICM and when not to use it!) then you are in trouble with a micro stack and will need some good fortune to get back into the game.

However, opponents in the lower limits (or even mid-limits away from the biggest poker sites) are not always rational, mathematically minded players for whom every move is designed to maximise their $ev... the reality is far from this, especially in the smallest buy-ins


The question I'd like to pose today is this:


Assuming opponents with no idea about SNG bubble strategy, what likelihood of them 'bursting the bubble for us' would be required in order to attempt to fold into the money??


You will notice that this question has an implied assumption that we are trying to cash only - not to win... we'll try and factor that in some other time.... this post is really to excercise the grey matter as to how to quantify the fact that others may 'donk you into the money' rather than being an excercise in mathematical precision...


So, here is our hypothetical situation *($ equity before posting in brackets of $100 prize pool)


UTG: 3000 ($30.89c)
Button (Us!): 500 ($7.32c)
SB: 3000 ($30.89c)
bb: 3000 ($30.89c)


With blinds at 150 / 300 (and not about to go up) we have 6 hands here before being forced all in with just 50 chips left for the 2nd BB.


Now, for each hand that passes there is a non-zero chance that someone busts and our equity increases from $7.32 to $20 for 3rd place, a healthy increase of $13.68c

There are a couple of ways to look at things from here:

If we take a percentage chance of the bubble bursting as X we can look at the $ev fold as actually being bigger by X for each hand, if you estimate a 20% chance per hand of the donks busting you into the money then $ev fold = $13.68 / 5 or +$2.74c... with a 10% chance we have + $1.37c.... so, if we plan to make a 'move' with our micro-stack we need to use a higher $ev fold than the 'pure ICM' would suggest (at least at the lower SNG buy-ins!)

The other way of looking at this is to calculate an approximate chance that you'll be 'gifted' 3rd place over 6 hands and use this as the basis for your ICM call / fold decision making. I make it a 66% chance of a 'gift 3rd' using 20% for each hand (probably high even at the $5 SNG level) and a 40% chance a 'gift 3rd' at 10% chance each hand (more realistic?).

Using these figures to increase our $ev fold we get:

- 20% / Hand 'Gift 3rd'.... $ev Fold base number = $12.15c
- 10% / Hand 'Gift 3rd'.... $ev Fold base number = $8.05c

Well, we reached the end of this post with a conclusion (that the fact that lower buy-in donks do not understand SNG Bubble strategy increases your expectation when folding in terms of prize pool equity!), and a way to roughly calculate this (!). What we are missing is a practical application in decision making... to illustrate this we need to find a hand that would have been a 'call' according to ICM that now becomes a 'fold' with the 'gift 3rd' chances factored in....

Will find one for you - but not today... duty calls.

Check out SNG Planet (banner top right) for an intro to ICM for SNGs if you are not quite sure about the terms used here.

GL at those tables, Mark

No comments: