RIO and other stuff: the first levels and the late game of a turbo SNG

After playing 30-40 or so of these turbos in the last two days days (had been playing ring only for a few weeks) I'm baffled by the amount of people willing to give implied odds to me (and to others) at the table by playing like they would in the later stages of the SNG. I'm also baffled at the horrendous late stage play: calling with K7o for 3/4 of your stack. Bullying big stacks, going AI four handed when two people are already AI, etc.

Don't get me wrong, I'm not trying to re-invent the whell here and these concept are probably very evident to the "veteran" players, but this is intended much more for the beginners or the people who have trouble with the structure of turbos:

a) Not every hand worth playing at level 5 is worth playing at level 1, and especially not from any position.

Example: I am in MP and nobody has raised before me during the first level. I have AJo. I clearly should lay this down most of the time because:
1) I cannot limp this and must raise since AJo hates multiway pots.
2) At the looser table, I'm more likely to get called by something like K9s, and, guess what ? AJo is not at all a big favourite over K9s. 60/40 preflop and we're looking at three outs if other dude pairs.
3) C-betting these tiny pots is worthless since anybody with any piece will call and you need to hit hard to continue safely.

This can mean laying down AQo from the SB to a raise at level 1-2-3. Let the villain have your 10 chips and move on.

b) Chips are our babies. We must not let them go at all costs.

Since turbos are much more about doubling up and stealing blinds from level 3-4 and on than playing actual, real poker, the old adage that you can play loosely to draw and get payed off is not always true. Sure playing A3s from SB with 7 limpers is good, no doubt about it, but if there were a raise and only 3 people call, maybe not so much.

We must have a medium or big stack by 50/100 to make doubling up profitable since we will rarely see a lot of good hands. Also, having a medium to huge size stack makes stealing easier.

c) Stealing is our bread and butter, but we should not attempt it before 50/100 most of the time.

Since you start with 1500 chips, the huge jump in blind level is from 25/50 to 50/100 since your M, no matter how you played before, cuts down in half. This means that you should go from nit that craves the reverse implied odds people give them in the previous levels, to a filthy thief. This is because stealing 75 chips is not much, but stealing 150 is 10% of starting stacks.

This doesn't mean that stealing is profitable from any position however, or that it should be attempted the same way from any position.

For example, this is a hand that came up today.

I have 1700 chips, the blinds are 75/150, 7 handed. I pick up 88 UTG, a very nice hand. Or is it ? Against 6 randow hands, this is how 88 fares. 20% equity versus 13% for a random hand. 20%, with no info on others holdings, is crap. Now, if I have K9o against the blinds only in this very scenario, my equity is 36 against their 30 each. This is the same equity differencial, but I have less players to face. And if I have the same 88, my equity is at a huge 50%. I ended up folding, having two people with big PP go AI after me and win the thing later on. I that 88 was OTB, I'm instashoving. From EP 7 handed, it's meh.

What does this mean ? It means that since we will be AI if we decide to steal the blinds to max. FE, it is way better to do it with crap if no one is in late, than do it with a fair hand early.

d) Maximize your FE late

When the majority of our chips come from stealing the blinds, we must maximize our fold equity, this means that we will be going AI to give players crappy odds to defend. IE:

We have 2000 chips, the blinds are 100/200 and it is 3 handed. The other two guys are about equal in chips and have big stacks. Clearly, we could raise to 500 to try to pick up the blinds, but we really should not since from the BB, it'll only be 300 more to win 700 and it won't dent our villain's chipstack to do so. Then, we must see a flop, hit and play correctly. This is mostly hardwork. Stick it in son.

The stop n' go strategy uses the same principle, but in reverse. If you are sure you are going to get called if you raise AI (if the blinds were 400/800 and you are standing on 2000), calling, and pushing first into your opponent on the flop to give him a chance to get scared/see monsters under the bed might be best. Note that this only works when we won't have position on villain. We should also not be picky about the flops we push and push anything. If you have 49o and the flop comes AAK, push, opponent might not call with his T hi...

e) Just like with dates, being first in matters

It sucks to date girls that have already been "tested" by your friends, and it also sucks to try to steal pots that have already been raised. This means that the standard with wich we are willing to defend our blinds should be a lot higher than the one with wich we are willing to steal. Restealing (with crap) in these turbos is not something that should be attempted very often, especially late, since most raises will be pot committing. Plus, donks are willing to call you, so just don't. Restealing has its place in higher buy-ins games or MTTs where people have history between them and think on more than two levels. Here, it is just spew.

f) Recognize that flipping is bad/ recognize that flipping is good

If flipping were to be graphed, it would start to suck heavily at around 12-13 BB and start to get very good at 30BB. It is also always very good when you have a low M since you have no choice but to go AI with that 66 and hope A3 calls... You should also be more willing to gamble when you don't risk the chance of busting out or busting out a player for a much higher payout than if you have only the chips to gain.

Today, I layed down QQ on the bubble when two people went AI before me and I did not cover one of them. Both were donks, but we have to realize that even if we have an equity advantage a good portion of the time, busting now would be really disastrous. Just steal some more blinds to make up for it. The same goes for defending your blind with 33. Sure it's a pp, sure it's probably in front, but you are flipping 99% of the time while we could be not flipping by stealing someone else's blind next hand.

The less we flip without being in extreme necessity, imho, the better our chances.

g) Know thy ICM, know how to play HU.

Good, you are HU, but if you always finish second, it won't do much good to your winrate. The money is to be had in first place. Read up on shortstacked HU play: it is mostly preflop hand selection and ranges so it can be learned easily (unlike ring poker where the most valuable decisions come postflop where the errors are very costly).

To sum it up:
- Be aware that everytime you enter a pot with a marginal or easily dominated hand, you are giving away RIO. Defending your blind with A3 is not always right: you are always flipping or being dominated. KQ is much better since only two hands dominate you, you dominate a lot of hands, you have live outs again PP and you are 60/40 against small Aces. I should add that small A become a lot more valuable HU since the chances of your opponent having an Ace drops down, compared to 4-5 handed.

- Stealing is the heart and soul of this game. Know how, when and why to steal. The next time you raise J9s OTB in level one to steal, kick yourself in the sack so it passes: there's no value in those measly 30 chips, at least not enough to warrant opening yourself up for mistakes on the flop.

- Flipping is pretty much always bad unless you really need to. Yes, this means laying down 88-99 once in a while if you don't gain much by winning but lose lots by losing (3500 chips, villain has 2200 four handed, with another extremly shorstack: let the small guy bust and open up ITM).

- Multitabling those can be better, IMHO, than trying to get foolproof reads by single tablng. Very reasonable multitabling (2-4 tables) will cut down on FPS, will make you tighten up, incease your winrate, and, in some cases, lessen tilt (it does for me). It's always infuriating to have JJ not hold up against AK on the bubble since you have invested 45 minutes and have nothing to show for it. Had you had 2 other tables running, you shrug it off, register in a new game and continue your quest for donk and go glory on the two remaining tables.

GL and yes, you are right in saying tl;dr.

That's a nice post. I actually took the time to read it all as I was waiting for a vid to download. I must admit that my study time has been very little since I started playing this game, yet I have pretty much come to most of the same conclusions as you just from experience at the tables and a lot of observing without actually having a theoretical underpinning to point to, especially the first few points. Sometimes I don't even bother playing until the blinds get to 50/100, it seems pointless. I'm still trying to get the blind stealing bit right and know I could improve in that area because it gets me into trouble, I don't know if a book will help with that or whether I just need to keep playing more until I eventually work it out for myself. I think I learn the most from watching what good players do.

One area I think you overlooked though, I still think that playing players is just as important as playing cards. Although I don't use a PT and can't point to any stats, I do know that a very large percentage of the pots I win are with nothing, and that that is based on reading your opponents and situations and knowing when to fold!! I would rate that just as highly as the theoretical skills and I'd rather have a read that the guy I'm going to AI on is likely to fold before trying it against any player (could be where I'm going wrong though!).

Reading opponents is important, but it is often less important than knowing how to play preflop optimally since we will rarely see flops: I find myself c-betting maybe twice before it gets three handed and then, you have a pretty good read on your opponents from having played them during the donk and go. I often make notes on people who bust me before leaving if they had weird cards/play for future use. Same thing for people I bust with weird crap (they take a stand with 73o against my KK).

Also, a lot of our opponents will play the same (loose preflop, no conception of GAP, WA/WB, etc.) so a good/standard line can very often be found on the spot without many experience with that particular player. ABC poker wins in this in the long run. Also, knowing when to c-bet and when not to takes some hand guessing skills, but it's pretty much always easy to figure out where you stand in a hand after you get called or raised.

And the thing with going AI on a steal: if you do it with hands that have an equity advantage against a random hand (wich the blinds have) and you factor in fold equity, you are never wrong in the long run. But taking reads about who folds what and when is a necessity for any serious player who wants to steal lots and thus win lots.

EDIT: After rereading my original post, it sounds as if you should always go AI on a steal. This is obviously not the case if you have a huge stack and are willing to only try to steal with a hand ouyou do not wnat to take to SD (87o). These situations arise frequently when it gets shorthanded since you won't have the luxury of waiting to pick spots.