Putting too much emphasis is a human flaw, knowing two extra cards and playing against them is math.
Maybe David would have to write the TOP against "two known" cards but its IMPOSSIBLE that this would be a disadvantage ever. Poker is a game of partial information, having more of that information has to increase the amount of +EV in a given situation and reduce the amount of -EV.
If we have the skills to utlize this advantge in EV potential is another matter, but that does not change the fact that it is there.

The huge edge is the fact that you will play FTP-correct against the guy to your right.

Why do we have to assume there is no human error and we will always use the information perfectly? David just says we are a good player, but that doesn't mean we're infallible. My points are still applicable to "one of the better players".

I think people rating this edge as a 30 - 100x improvement are way, way off. if you were 1 in 2000 to win without the edge, I can't see any way your edge improves you beyond 1 in 200. Even seeing someone's cards, think about how hard it would be to have a 90% edge in a single hand, let alone the entire tournament. You'd have to fold every single hand that wasn't HU between you and the guy next to you.

I might believe 10x, but I doubt it's more than 3.

Let's ignore the assumption that we're better than average for a second.

Let's say that 50% of the time you get heads-up without this ability, you win the tournament. And now let's say that 90% of the time you get heads-up with this ability, you win the tournament.

That already increases your chances of winning the tournament by 1.8 times, without even considering that you have the ability before heads-up.

Now, let's say you normally have a 2/3 chance at getting to heads-up once it gets three-handed without this ability. And with this ability, let's say you have a 1/3 + 1.8*1/3 chance to get to the heads-up. That's another 1.4 times the chance to win it.

From four handed to three handed you get another 1.26666... using the same method.

What these calculations are doing is saying with N players left, everyone has an 1/N chance of not busting next, except you and the guy to your right. Between the two of you, you guys have a 2/N chance to bust out. But instead of dividing it equally, the guy to your right gets 1.8 times his fair share of the division.

I think this approach is reasonable for final table play. If it is, and if the 1.8 number is about right, then just having this ability for final tables increases your chances of winning the tournament a little over 6 times.

I like this model. There's no inherent justification for it past N=2, but it "feels" right for some reason. 1.8 may be conservative, but that's OK. It would be interesting to work it back all the way to the full size of the tournament. I don't think the logic changes when you go to multiple tables.

And of course it's somewhat on the low side since you'll have 2 additional advantages on the field:
1) knowing 2 cards (which is small, but might be of use in some circumstances)
2) A large chip stack (your model doesn't take this into account)

In fact, the large stack is very critical. It will allow you to establish a long-run while others are subjected to variance. The more I think about it, if you bust out a couple of guys on your right early in the tournament, I'm not sure a good player would ever lose. It would be hard to construct a scenario where it happened as long as your stack was >> than the stack on the right, and you avoided confrontation with the rest of the players.

Bump. You said you'd give your opinion in a few days. Your fans are anxious, David.

Guaranteeing you win the heads-up doubles your chances to win the tournament, yes. But it doesn't double your EV.

Before I do that I would like to see more specific analysis about specific hands where you would change your strategy.

once you got down to short handed you would be a real force to reckon with. Heads up, unless start severely short stacked you should win close to every time - so the odds of youw inning should be at least the odds you coming in top 2 given normal circumstances.

Good q, might expan on thoughts tomorrow.

Wow, this is some edge. I think if you know you will get it every hand your chances of winning the tournament have to go up to about 1 in 50 or better (this will of course be affected by how your opponents play, loose being worst for you as it allows for more variance. I'm assuming a pretty loose field, especially from day 2 onwards. However you want the player currently sitting to your right to be super aggressive, as you want him to be involved in a lot of pots so you can maximize your advantage). Because of the way the prize is distributed, your EV doesn't go up proportionally, you get proportionally less EV, but it's still a lot more. I don't know how to calculate it.

If you don't know you'll keep getting the same advantage all over the tournament (but you will), your chances of winning probably drop to about 1 in 250 or so.

Just guessing at this. However off I might be, I know the edge in both scenarios is huge. In the second one you lose the ability to wait for the absolute best spots, but the advantage is still very large.

Close to right, but this will depend majorly on stack sizes and their relative size to the initial pot (M ratio). With huge blinds and antes relative to stack sizes (which is very normal at this stage, albeit perhaps not in this particular situation), you will only win a little over the percentage of chips you have. So if you have 25% chips and your opponent has 75%, and you have, say, 2.5 M versus his 7.5 M you'll have maybe 28% chances to win the tournament, facing a similarly skilled player.

Even if he's the short stacked one with 2.5M and you have 7.5M, you'll have maybe 83% to win.

Since when was this event winner takes all?

ToT

Well, obviously if right guy folds, then you use his hand info to help decide whether to play marginal hands. But, as I suggest below, you're going to want to be avoiding marginal hands anyway, so this won't have too much impact.

If right guy raises in early position, you can't do too much because of the chance of someone else waking up with a hand. You can still call more lightly than usual. Also, you can come over the top with something like 99 or AQ when appropriate.

If right guy raises in late position, then you can blow him away with a reraise if he's raising light. And if he calls the reraise, you can just push a flop that he has no piece of.

If he's raising late with a hand that can fade a reraise (he'll no doubt loosen up, especially since you'll have done this to him before), you can just flat call and outplay him postflop. All you have to do is be sure to put in the last bet when he's got a hand that can't call.

Also, make last-longer bets with him.

You'd want to cultivate a very tight image apart from your duels with this player (show a lot of good hands), because you want other people to get out of your way allowing you to play HU. And, since your edge is super high vs. this guy and not particularly high vs others, you'd be correct to play quite tightly vs. other people, so that should work out nicely for you.

First hand at a new table, you have this ability and know it will continue.

You are in the BB with 40k, average stack 20k, blinds 500/1000. Folded to SB who pushes for 10k.

How much equity do you need to call this? Win or lose, calling gets you 10k more chips in the stack to your right.

58%?

Doesn't work that way, because if he gets eliminated you'll still see the cards of a player to your right. This would apply if you could only see "that" players' hand, but not others'

Having right-hand guy as a short stack is the worst scenario, because if he's playing jam/fold you can't bluff him. It's still going to be hard to bluff him at 20 BBs though, so it's not a big help doubling him up. Busting him, and having him replaced by an average stack, would be a significant coup, however.

Still, your own stack isn't enormous, so you shouldn't be taking bad gambles here.

Tough to say. I think I'd want .66 EV in the hand to make the call, but I'd be willing to listen to other opinions.

I don't see how what you said contradicted what I said. If you win the hand, the next guy on your right will have >10k stack (average stack is 20k).

So the benefit of winning this hand (compared to folding) is more than just the chips you gain. You also gain by getting a 20k stack on your right (on average).

And the loss of losing this hand (compared to folding) is less than the chips you lose, because you lose them to the stack on your right.

This only makes a minimal difference if any, because if he keeps his short stack, he's more likely to go busted and you're still going to get a bigger stack to your right.

I didn't read all the responses but it is obvious that any of us would win 99.997 - 100% of the time if we knew we'd see those cards every time. Think of how many times you'd stack the obvious victim ... otherwise play as tight as necessary.

In fact, 100% is the right answer...

You obviously don't know much about math. Donk goes all in to your right with 23o. You hold AA. You obviously call. You have only a 86.2% chance to win. You can't always wait until you have the absolute nuts to put your money in, this would most likely result in your whole stack being eaten away.

Isn't this just a classic case of "it depends" though?

If you get a Jim McManus on your right (at least the way he claimed to play in PFS, where his range for parts of the tourney was AA,KK and he didn't fold to bad boards) you've gained no edge. If he raises you enough to deny implied odds, you'll have no way to exploit the fact that you know his hand. In fact, that big hand/calling station approach is a deadly weapon against bluffers with a good read (Which is what you would become in this case). The only edge you get against such a player is certain knowledge that you need to get out of the way of their big hands, which frequency analysis should give you anyways.

On the other hand, if you get a Negreanu on your right, you're golden. He raises, you call with any live 2, and then you out small ball him with ease since you know his hand and know he likes to pick up pots so you can easily induce and then snap off bluffs. Since by the rules he won't remember your success against him when he plays small ball, and the rest of the table won't realize by frequency analysis that your calls are air (and hence will be hesitant to overcall), you'll get his entire stack with ease. The more he tries to be table captain, the happier you are.

With players somewhere in between, you're going to get the same kinds of edges as against Negreanu, but not as frequently since they won't enter the pot as frequently, and when they do you won't have 2 live cards as frequently.

It's not at all certain that calling is the best strategy. If donk also plays some smaller pots, you might want to wait for the sure thing rather than taking an 80:20 shot or whatever.

Very interesting thread. Made me get off my ass and register.

I agree that a major part of the strategy will be to beat up on the guy to your right...

Remember, if you have the edge, you'll likely have a better than average stack (hopefully much better than average). Unless you're covered in chips by the guy on your right (the victim), you wouldn't pass up an 80:20 shot. If you lose, which only happens 1 in 5 times, you'll still have chips to play. Better yet, the guy who's cards you can see will have them, and you'll just get them back later.

In other words, opposite of proper poker theory, you'll play more loosely when you have a larger M (specifically, a larger stack than the victim's), and tighter when you have a smaller M (more specifically when your stack is smaller than the victim's).

You simply can't pass up on the pair-over-pair or domination situations because you still have to accumulate chips quickly. The other front contenders will be. The biggest mistake would be to tighten up and wait for premium hands at the cost of waisting too much time to build your stack, because the blinds are increasing. Once your stack is built up, you can play more marginal hands including most that are giving you simply better-than-proper-odds to play.

For instance, if you play against a non-victim and on the flop you have top pair top kicker and you suspect he flopped a flush draw, you can give him "proper odds" to call when in reality you know that two of his suit were held by the victim and he has 2 less outs than he thinks. Once your stack is built up, you can start to exploit all of the more minor informational advantages at minimal risk and continue to build your chip stack.

Stealing opportunities will abound, not limited to raises, reraises, uncallable all-ins (h/u against the victim). Squeeze plays will be frequent...reraising after an early caller and a raise from the victim.

****
I disagree with the logic that you are ~100% to win heads up. The fact is, the Kill Phil (or Sklansky's "System") would work very well against you. Even if you know his cards, you're likely to be no better than ~3:2 during most hands, and (depending on the size of the blinds/antes) if he continually goes all-in you will have to call at some point and will be at most 4:1 to win if you wait to pick your spot. If you have him outstacked, he may have to outdraw you twice...not impossible. Better yet, if he doesn't even look at his cards and goes all-in in the blind (doesn't give you a chance to see his cards), you'll likely be closer to ~3:2 (if you pick your spot) and will have a much tougher time. Granted, this is closer the optimum strategy against you, the "better player/cheater", but it shows you're not guaranteed anything.


So what's your odds of winning the whole thing? Well, since we're playing poker, nothing is for sure. If I had to guess, I'd say that if you play mostly 3:1 or better for all-ins (against smaller stacked victims) and limit your bluffs to one in every 3 confrontations with the victim, you go from a top 40% player to a top 4% player. That's 1 chance in 200 to win it. TO WIN IT. I think you'll likely get farther than top 200. But remember, we're playing poker. You might get put out when someone sucks out on you ...and once you're shorter stacked you start tightening up (which I think is proper) and the blinds start eating you up.

Bad beats...I remember Aaron Kanter pulling at least 6 river cards in 2005 on what appeared to be lock hands. Running up against someone like that would nullify your edge pretty quickly.

I had two 800s in the GRE with about 30 minutes to spare in 1986. My IQ test results exceed whatever range the test measures (higher than 4 standard deviations above norm).

Your chances of winning this tournament in the real world is 100%.

The question is so good it's worth the price of "Poker Essays 4."

I'll give you a hint, if you happen to have quad kings on the flop and someone goes all in on you, you fold--every time (unless your stack is big enough).

keywords & ideas: real world, survival, big blind