Two Plus Two Newer Archives - Updated satellite strategy post

Two Plus Two Newer Archives (http://archives1.twoplustwo.com/index.php)

- MTT Strategy (http://archives1.twoplustwo.com/forumdisplay.php?f=25)

- - Updated satellite strategy post (http://archives1.twoplustwo.com/showthread.php?t=147829)

Updated satellite strategy post

I promised myself that I'd post another big sat thread after I won my WSOP entry, because I'm pretty sure the first one cost me money after I wrote it [img]/images/graemlins/smile.gif[/img] Looks like I'm set for the year now, though, so here's an update:

One year later

By and large, sats today are softer than they were. Because of sites (especially Stars) sending hundreds of people in from sub-satellites into supersats, there's tons of complete dead money that wasn't there in such numbers a year ago. This isn't exclusive to bigger sites; the Bodog WSOP sats, which paid one spot but excluded repeat winners, were simply awesome, because on top of the overlay they all had, half of their 50-100 players had no shot to win.

That has some practical implications for the types of sats you should be playing. Stars DSO's are even worse than a year ago (comparatively; I'm sure they're much easier on a per person basis) because they still have pros that dominate the last table and play for W$. Compare a 160, 81 person DSO with the 125+10 Bodog sats with 45-50 entries and no repeat winners - there's no contest. I am now pretty sure that top heavy one seat satellites should now almost never be played (except Bodog, but they stopped running those last week.) The one thing that could overcome the advantage of having multiple qualifying spots is massive overlay, such as at Bodog or a few other smaller sites.

A footnote: Stars and Full Tilt now give you lots of money for final tabling/winning the WSOP. I don't know Stars' endorsement deal terms so I won't add them up, but FT's deal gives a player with a real shot at winning an extra 1K+ (much more if you are above average) in EV just for qualifying from their site.

Satellite bubbles and the Nash equilibrium

Anybody that plays SNG's knows the ins and outs of pushbotting have a lot to do with game theory. If you push 100% of your hands in a given situation on the bubble, your opponent is right to call with, say, the top 25%, but being called that often is -EV for you in the long run, so you only push the top 50% (so he can only call with the top 10%, but that's -EV for him so he calls more to make you adjust...etc.) Eventually, the situation stabilizes - you push exactly the top 79%, he calls the top 14%, and you are both playing optimally. This is called a Nash equilibrium.

But there is no practically attainable Nash equilibrium in a big satellite bubble. Let's say there are 30 players left, the sat pays 25 seats, and stacks range from 8 BB to 50 BB. If the big stacks play optimally, they will sit out. If the medium stacks play optimally, too, they will go all in (into other medium stacks) virtually blind until they're big rather than medium. But even an 8 BB stack is dangerous when he pushes into someone with 15 BB, and if he pushes into a 50 BB stack, the value of time is the only reason to call the push; certainly, $EV wise, the correct play is probably still a fold with any 2. Technically, then, everybody should take turns pushing two cards - never being called except possibly with aces - until the blinds are so high that the extra fraction of a BB lost from calling an all in is meaningless.

In practice, this is unattainable, because I)people suck at poker and II)not every medium stack realizes that he should never do anything other than go all in in this situation. In the sat I was in, every hand was minraised, and 75% of the time, everyone would fold, except for the occasional medium stack restealing all in with any two (knowing it would almost never be called.) The minraise almost meant the same thing as an open push, even though it was essentially a chip giveaway to anybody that had a stack 2/3+ of the raiser's. It is a very odd situation that only comes up in such satellites, and the solution for it is for medium stacks to be much more aggro on the bubble than they are [assuming nobody else ever does anything totally suicidal to their EV like call...of course, you make that assumption at your own risk]. What currently goes on in sats like the one at Stars is as exploitable as anything else I've ever seen in poker.

The final hand of the sat went like this: a medium to large stack raised 4x to 12K in EP. Another medium to large, but somewhat smaller stack called the raise in MP. The SB, with ~22K total (he would have been about 20'th of 26) thought and pushed in with kings, EP instafolded, and MP thought forever and called with jacks (losing would have put him about 20th as well), then sucked out. Hilariously, every single decision made by all 3 players in the hand - in a 650 dollar buyin - was mildly to extremely wrong and, in at least one case, a $3000+ mistake. Think about that and then figure out how profitable these things are compared to a DSO, especially if you're the short stack watching this from the other table with ~10K chips.

Deep stack poker

Tournament structure has generally become far slower over the last year and change, while dead money still busts out at a relatively rapid clip. In general, the bigger the buyin, the better the structure, so big event sats tend to be very deep; on Stars, the mean number of chips 3-4 hours into this sat was over 50 BB, and as one of the chip leaders, I had over 100. This doesn't have much to do with anything except that you now have to know how to play with cash game stacks in order to beat sats; it's no longer optional. Furthermore, later on (before the true bubble but after bubble FE becomes a factor) it's pretty important to learn to resteal and bluff properly [img]/images/graemlins/tongue.gif[/img]

If I had to qualify again, I would...

I'm definitely playing the 150 seat Stars sat for W$ as it's probably going to be in the top 10 most profitable tournaments this year. The Full Tilt 100 seat sat is gonna be up there, too, especially since pros that have already won seats at Party and Stars will not be playing in it. Party's 8 seat guaranteed nightly tournaments seem like a good value and can even have a little overlay (although Party pays 1K less for winning 2+ seats). The good old 73+7 sat to the Stars 650's is how I won my seat (although I bought in direct to the other 650 I played) and also gives practice on how to play a sat FT. Various turbo rebuys are great, too. Of course, if Bodog were an option, or some other site with overlays like that still is, toss this whole thing out the window and qualify from there.

I'll hopefully see you all at the ME.