Adjusting with Aces by Nate Melvis

[jfk]

[ QUOTE ]
I viewed MP as potentially good or excellent but more likely mediocre-to-good

[/ QUOTE ]

How did you reach this assessment?

[ QUOTE ]
I had picked up both a preflop betting pattern and two physical tells on him. When he opened for exactly 3x the big blind, he had shown down a steal-type hand, or at least one happy to win uncontested, but when he opened for 3.5x or 4x, it had been a stronger preflop hand. When he said "I'm going to raise" or "raise," his hands had been very strong, and silent bets had accompanied weaker hands. Finally, he handled his chips much more casually, and displayed much more casual body language in general, with a good hand than with a bad one.

I had only played with MP for 35 or so hands but had observed him carefully in the pots he had contested. None of the observed factors are infallible, but they agreed and were thus mutually reinforcing. When you play live tournaments, you will often face unfamiliar opponents and play thirty hands per hour. I've found it's far better to search for information despite small sample sizes than to hide behind standard lines.

[/ QUOTE ]

For those of us who find focusing on each hand of play for many hours to be among the greatest challenges of play, how do you go about do this? Did you focus on this player specifically for a reason or were you able to give a similar bit of attention to each at the table? Is it common for you to find a tell of this (apparent) reliability so quickly or was this an aberration?

[ QUOTE ]
I had just processed this information when the dealer burned and turned

[/ QUOTE ]

That seems like like pretty quick thinking at the table. I would imagine that you'd anticipated this range beforehand. Is this a safe assumption?

[ QUOTE ]
45% of the time he would have QT or JT. He would usually call and usually check behind with those hands, so my bet would gain roughly 600 chips relative to checking, for a weighted gain of 280 chips.

30% of the time he would have T9 and J9 and betting and checking would be equivalent.

20% of the time he would have T7. Sometimes he would check behind and save me my stack. I estimate my weighted loss at 85 chips.

5% of the time he would have QJ and occasionally bluff with it. My weighted loss by not inducing this bluff is small, 15 chips or so.

[/ QUOTE ]

To be frank, I would have a hard time with these calculations at the table (one reason why I don't play high end MTTs). From your article it seems like you had these figures worked out quickly. At what stage of your MTT career did you get to this point. What sort of time elapsed for you to make your river decision. In online tournies where the decision time is more finite can you reach the same comprehensive conclusion?

Also, I'm curious as to the degree to which stack preservation figures into your equation. You mention it but don't quantify its consideration. How crippled would you have felt with a 725 chip stack and blinds of 25/50?

[ QUOTE ]
Even if you occasionally land in a tough spot -- I hope you would play better than I did on this river, for example -- I hope you are be willing to trust your reads and abilities instead of automatically making pot-sized bets and never slowplaying preflop. When you have hundreds of thousands of chips to collect, dogmatic autopilot is among the biggest leaks you can develop.

[/ QUOTE ]

Please forgive my confusion, but I'm unclear with this last paragraph whether or not you're content with the decision (if not the outcome). Apparently your range considerations were on target. In hindsight, would you've made a different river play or were you destined to stack off? Might you've put more turn pressure on if you felt stack committed?

Thanks again for the article.

[Nate.]

[ QUOTE ]
[ QUOTE ]
I viewed MP as potentially good or excellent but more likely mediocre-to-good

[/ QUOTE ]

How did you reach this assessment?

[/ QUOTE ]

Hm. Hard to remember. First off, I had found those tells/betting patterns, and better-than-good players would be less likely to have those. If memory serves he'd also missed a clear river value-bet or paid off too light or something. I want to say that he also did some talking at the table and showed that he was at least somewhat conversant with some basic concepts.

[ QUOTE ]
[ QUOTE ]
I had picked up both a preflop betting pattern and two physical tells on him. When he opened for exactly 3x the big blind, he had shown down a steal-type hand, or at least one happy to win uncontested, but when he opened for 3.5x or 4x, it had been a stronger preflop hand. When he said "I'm going to raise" or "raise," his hands had been very strong, and silent bets had accompanied weaker hands. Finally, he handled his chips much more casually, and displayed much more casual body language in general, with a good hand than with a bad one.

I had only played with MP for 35 or so hands but had observed him carefully in the pots he had contested. None of the observed factors are infallible, but they agreed and were thus mutually reinforcing. When you play live tournaments, you will often face unfamiliar opponents and play thirty hands per hour. I've found it's far better to search for information despite small sample sizes than to hide behind standard lines.

[/ QUOTE ]

For those of us who find focusing on each hand of play for many hours to be among the greatest challenges of play, how do you go about do this?

[/ QUOTE ]

Hm. Well, after a while you find common things to look for. Speech while raising and raise multipliers are two very common ones. As far as willingness to focus... I was at the World Series and I love playing poker. Getting sleep helps. Realizing how much this sort of information works in one's favor is also good motivation to look for it. But really, it's sort of like driving. There is a vastness of information but you learn how to sift through it, and to keep sifting.

[ QUOTE ]
Did you focus on this player specifically for a reason or were you able to give a similar bit of attention to each at the table?

[/ QUOTE ]

This guy had played a handful of pots already, and he was sitting directly across from me, so I'd had plenty of time to look at him. He was also a talker. So he got more attention than most, though to this day I can remember stuff about a couple other guys at the table (even though I busted quickly). Definitely pay attention to everyone, but obviously the ones making more decisions will give you more to look at. (Also, you'll be more likely to play pots with looser ones.)

[ QUOTE ]
Is it common for you to find a tell of this (apparent) reliability so quickly or was this an aberration?

[/ QUOTE ]

It's common, especially at the World Series, where (a) people are cowed and (b) mediocre players are often trying so hard to make a perfect decision that they end up giving away a ton of information. (That is, part of what keeps weak players from giving away information is laziness, and that laziness is absent at the WSOP.)

[ QUOTE ]
[ QUOTE ]
I had just processed this information when the dealer burned and turned

[/ QUOTE ]

That seems like like pretty quick thinking at the table. I would imagine that you'd anticipated this range beforehand. Is this a safe assumption?

[/ QUOTE ]

Well, obviously the range after the turn is very much related to the flop range I'd been thinking about already. Another thing I didn't mention is that when he called my initial thought was, "that looks like a guy who just made top pair."

So when I began to process the hand logically, the first thing I was considering was that he had a ten. So I didn't go down any logical dead-ends before hitting the likeliest result. (Logic and the flop range dictate that a ten is likely, but without the physical read it might have taken longer to come up with that.)


[ QUOTE ]
[ QUOTE ]
45% of the time he would have QT or JT. He would usually call and usually check behind with those hands, so my bet would gain roughly 600 chips relative to checking, for a weighted gain of 280 chips.

30% of the time he would have T9 and J9 and betting and checking would be equivalent.

20% of the time he would have T7. Sometimes he would check behind and save me my stack. I estimate my weighted loss at 85 chips.

5% of the time he would have QJ and occasionally bluff with it. My weighted loss by not inducing this bluff is small, 15 chips or so.

[/ QUOTE ]

To be frank, I would have a hard time with these calculations at the table (one reason why I don't play high end MTTs). From your article it seems like you had these figures worked out quickly.

[/ QUOTE ]

Well, over the table it was more like, "he has one pair too often, and he's not totally predictable, so I can't fold for this price. Also he will likely feel tied to the pot if I bet. All-in!"

It takes a while to type, but given the hands you post I bet you go through these types of considerations pretty quickly.

[ QUOTE ]
At what stage of your MTT career did you get to this point.

[/ QUOTE ]

Wow. Tough. Last summer I lived with some world-class players. I hadn't played too many live MTTs before that. So yeah, probably sometime last summer.

[ QUOTE ]
What sort of time elapsed for you to make your river decision.

[/ QUOTE ]

My chips were in the pot within about a second and a half.

[ QUOTE ]
In online tournies where the decision time is more finite can you reach the same comprehensive conclusion?

[/ QUOTE ]

My pause would have fit well within any site's allowed decision time. That said, I wouldn't have been working with as much information online, so I wouldn't have had the same base of good information going into the river, which would have made any tough river decision take longer. There are reads to be had online but rarely do I have this much opponent-specific information to work with unless Villain is a regular.

[ QUOTE ]
Also, I'm curious as to the degree to which stack preservation figures into your equation. You mention it but don't quantify its consideration. How crippled would you have felt with a 725 chip stack and blinds of 25/50?

[/ QUOTE ]

It was so early that I was working as if $EV roughly equaled cEV. I had an eye toward accumulation, because chips beget chips in these mammoth weak fields, but I wasn't going to take bad bets to try to build my stack.

750 is enough to work with at 25/50, especially because live opponents handle push/fold situations (offensively and defensively) so poorly.

[ QUOTE ]
[ QUOTE ]
Even if you occasionally land in a tough spot -- I hope you would play better than I did on this river, for example -- I hope you are be willing to trust your reads and abilities instead of automatically making pot-sized bets and never slowplaying preflop. When you have hundreds of thousands of chips to collect, dogmatic autopilot is among the biggest leaks you can develop.

[/ QUOTE ]

Please forgive my confusion, but I'm unclear with this last paragraph whether or not you're content with the decision (if not the outcome). Apparently your range considerations were on target. In hindsight, would you've made a different river play or were you destined to stack off? Might you've put more turn pressure on if you felt stack committed?

[/ QUOTE ]

In hindsight I think he bluffs rarely enough, and my physical reads on him were good enough, that checking and evaluating (possibly folding) is slightly better than pushing.

I don't regret my turn play, because he's so likely to have so few outs, and to pay off the river with one pair (so the effective odds aren't just working against me). If I knew he had top pair I would have bet more, but I was out of position and I was still working with the broader hand range.

[ QUOTE ]
Thanks again for the article.

[/ QUOTE ]

Thanks for the very good questions. I hope these answers are coherent. Please ask anything else you wish.

--Nate

[creedofhubris]

[ QUOTE ]
45% of the time he would have QT or JT. He would usually call and usually check behind with those hands, so my bet would gain roughly 600 chips relative to checking, for a weighted gain of 280 chips.

<...>

5% of the time he would have QJ and occasionally bluff with it. My weighted loss by not inducing this bluff is small, 15 chips or so.

[/ QUOTE ]

Your hand weighting made no sense to me, because you put in a decent-sized checkraise on the flop, and these hands, which you give him a 50% chance of holding based on his turn action, are no pair no draw on the flop.

Do you really expect people to call you with nothing but random overs, not even ace high?

[Nate.]

[ QUOTE ]
[ QUOTE ]
45% of the time he would have QT or JT. He would usually call and usually check behind with those hands, so my bet would gain roughly 600 chips relative to checking, for a weighted gain of 280 chips.

<...>

5% of the time he would have QJ and occasionally bluff with it. My weighted loss by not inducing this bluff is small, 15 chips or so.

[/ QUOTE ]

Your hand weighting made no sense to me, because you put in a decent-sized checkraise on the flop, and these hands, which you give him a 50% chance of holding based on his turn action, are no pair no draw on the flop.

Do you really expect people to call you with nothing but random overs, not even ace high?

[/ QUOTE ]

Creed --

Yeah. I only made it a few hundred more, and I think he would have called with all sorts of garbage. As it turned out he called with two overs and a gutshot. He seemed to be the can't-pass-up-a-price type.

--Nate

[novel20]

Nate, if you were doing it again, do you have any alternate line?

[Nate.]

Novel --

I like my line through the turn. He was willing enough to play with garbage in the face of apparently good pot odds, and so overwhelmingly likely to have no pair, that trying to get him to make a mistake by confronting him with an attractive price was the best idea, I think. As I mentioned earlier, given my good physical read on him I think checking might have been a little better than betting the river. (But when I discussed the hand with a world-class player right afterward and got to the river, he immediately said "blah, just shove.")

--Nate

[UpstateMatt]

Nice article Nate. I hope you do more writing for 2+2 (and other poker outlets). Your strengths as a writer are a great compliment to your strengths as a poker player.

One question: this hand came from early on in a preliminary event, and as such, you were working with an effective stack of about 40 big blinds. For this particular situation - i.e. trying to take advantage of probably steal hand / average opponent - how does your strategy change over a variety of depths. For instance, would you have played it the same PF at a depth of 200bb? how about 20bb?

It's not obvious to me what the ideal depth(s) to do this at is , but it struck me that 40bb might be non-ideal, particularly out of position. You are likely going to check-raise the flop for value most of the time, thus creating something like a 28bb pot going into the turn, with somewhat less than that left behind. That strikes me as likely to produce a lot of tough situations if you still just have one pair.

Thoughts?

cheers
matt

[Nate.]

Matt --

Thanks again.

First off, I think that given my reads, calling is better than reraising no matter the depth, because I think he's folding if I reraise, and he will make enough mistakes at any depth (that isn't trivially short) to be willing to take a flop. Against a stronger player, or one who I thought would be more apt to call a reraise with a weak hand, raising would be better. But here I would have slowplayed with 25BB, 125BB, 1250BB...

I think 40BB is actually one of the better depths for this. My quick-and-dirty over-the-table way of deciding hinges on the basic NL principle of shooting for the other guy's stack and primarily weighs the probability that slowplaying will create an extra chance at stacking someone compared to the chance that it costs me someone's stack.

So many of these guys seem to have laminated cards reminding them not to call reraises with AQ -- a good reason to reraise light in these events -- but have no trouble going broke with top pair. With 37 BB behind and a 6 BB pot it shouldn't be too much trouble to get a guy's stack in (you don't even need to pot it on every street).

By this reasoning one of the worse depths is the one where normal betting after a reraise would have just barely gotten a guy's stack in. If the standard reraise is to 9BB, something like 9 + 12 + 30 + 60 ~= 110 is a bad number of big blinds at which to try this, because you're giving up one of your best routes to getting his stack.

Anyway, everywhere the question hinges on his ability to outdraw one pair. If we simplify the situation to one where there's a pot-sized bet on the flop and one on the turn, or maybe as it played out, with likely a little money to be bet on the river, I'm pretty sure any model that involves his calling with one pair on the flop will leave me net-profitable. Something like 30BB might be even better. (And something like 65BB might also be better, if I can take advantage of him on the river, as I think I can.)

I'll try to give this question more thought when I'm not 6-tabling... and I suspect you have some good thoughts on the matter yourself.

--Nate

[UpstateMatt]

Nate:

Thanks for the response. My initial, gut-reaction thought regarding the depth went something like this:

----
In slowplaying these Aces out of the big-blind, we're adopting a strategy to maximize our EV. However, because the strategy seems destined to be "call preflop, checkraise flop, bet turn," when we have 40bb, it looks like we're going to put in something like 7% of our stack preflop and then about another 40-50% of it between the flop and turn, virtually automatically. Given this, it is going to be hard to fold the river. Thus, we've managed to get of 40bb in, the majority of it post-flop, with one pair. That might be bad. What I'd really like to have here is 60 or 70 bb, such that I can make better decisions than my opponent on the late streets, and not be locked-in to decisions by the felt.
-----
A few (slightly) more refined thoughts:

It seems like 40bb is a depth where this strategy wishes it had one more bet. I feel like the situation on the river is such that our opponent can potentially play perfectly, but we can't. If the effective depth was another pot-size bet deeper, then we'd almost certainly have the room to outplay our opponent on the river. At 40, it doesn't seem to be the case. The money is going in, and we're reduced to our straight-math odds. That seems fine for a cash game, but i feel like this may be a situation to avoid in a tourney. I don't like saying that, because we seem to be making +EV plays the whole way, but the sum total of them seems to lead us to potentially difficult situations. Assuming we have read his hand correctly, 1/3 of the deck is going to annoy us on the river, and if he is a good player, he is going to get away from our river shoves plenty of the time with the hands he misses with.

Winding back out, my sense it that we might want to adjust our turn (and flop) play here. If we really feel like he hit the ten, one option is to try to check-raise all in on the turn. We could make that more easily accomplished by using a slightly larger flop check-raise. That would nullify his ability to outplay us on the river, at least.

And sure, he might take the free card, but then he will almost have to bet the river with any hand he would call the river with, so we can very profitably check-call (or even check-raise), without exposing all of our stack when scary cards come on the river. It just seems more flexible.

I guess my bottom line is that I feel like 40bb is somewhere between the short depth where you can't make a huge mistake getting it in with Aces (say 15-20bb or less) and the deep depth where it's likely awful if you get it all in with Aces post-flop (say 200bb). But we seem to be taking a line that is pretty much guaranteeing we get it in, and I'm a bit uncomfortable with that.

Now, part of the strategy is to take advantage of reads. You make it clear that your flop and turn play is influenced by your real-time table reads of opponents. But I wonder how much would change if we had no read? We are virtually obligated to check-raise either the flop or turn, i think, and preferably the flop. Being out of position and without many scare cards that can fall on the turn, we're likely betting there. And then the river plays automatic given our stack size.

Like you said, you don't trust the opponent enough to fold on the river if he shoves. What about if he check-raises the turn? Because if we can't fold there, i don't think we're giving ourselves much room to outplay him.

And I'm also not sure we can't get more money in earlier on. This opponent has fulfilled a request to put half his stack in by the turn with mediocre top-pair and junk kicker, with a gutshot. Is he really going to throw it away pre-flop or on the flop if we put a little more heat on?

Ok, I'm rambling. I'll stop.

[brizzology]

Great read.

I think the sentence "The turn was a ten, for a board of 853/T" should read "board of 863/T".

Looking forward to more of your articles!