Thanks to everyone who participated in the first installment of this little experiment. The question I left you with, essentially, was about whether a suited connector like 87s should be raised in early position, and if so, why?
As I expected, a lot of the comments hinted at the right answer without quite nailing it on the nose. That’s good, as it will give us a chance to address some common misconceptions. Here’s a representative example: “when you raise pre flop, you don’t raise a hand, you raise a range of hands. You want your range to be balanced, so you have to put big hands and hands with potential. Even when 87s is behind pre and post flop, it’s your whole range that matters.”
This seems to imply that opening 87s is -EV, but we should do it anyway. This should never be the case. The route to optimal play is to make the most +EV decision at every opportunity. Folding has 0EV, so if raising is -EV, then folding is better. Contrary to popular belief (a belief that I myself espoused until not too long ago), you shouldn’t make plays “for the sake of balance”. At equilibrium, the EV you gain from, for instance, giving your opponent more incentive to pay off your strong hands because he knows you also raise weak ones, can’t be so great that it overwhelms whatever you lose by playing weak hands. If it were, your opponent could just revert to not paying off your strong hands, let you get the occasional bluff through, and unilaterally increase his own EV.
What this hints at, however, is that the general composition of your range may actually make it profitable to play other sorts of hands. This is what people have in mind when they talk about “implied odds”, although I find that term misleading because it conjures up images of “slot machine poker”: making big hands and then getting paid off. In fact, as we saw in the hypothetical above, hands can outperform their hot-and-cold equity not only by getting paid big on certain boards but also by bluffing on boards that are good for other parts of your range.
There’s a small universe of hands so strong that they can be opened and, even with six or seven people left to act, be a favorite either to win the blinds or be called by worse. Your opponents simply won’t be dealt better hands than, say, QQ or AK often enough to both defend the blinds vigorously and consistently be ahead. So, you certainly will open hands like that from early position. Just for the sake of argument, let’s say that range is {AQo+, AJs+, TT+}
Now let’s think about how that range will play on two different boards: 654r and AT5r. On the 654 flop, the danger is apparent: this range never makes a stronger hand than one pair, which gives opponents the opportunity to make both bluffs and thin value bets with impunity. Having 87s will not necessarily “protect” you from this strategy. It will, however, give you the opportunity to win a big pot on this flop. This is why I say that it’s not about making a -EV raise with 87s in order to protect some other part of your range, but rather about the composition of your range making it profitable for you to include some different types of hands.
The danger on AT5r is less apparent. Many people are perfectly happy to see a flop like this with AQ and then with the pot and perhaps even another bet before eventually taking it down with a turn or river bet. I even see people who want to bet hands like KK on this flop to “represent the Ace” because the board so strongly favors the pre-flop raiser.
The reason you may have trouble winning big pots with objectively strong hands like AQ or AK is that your opponents correctly perceive you are rarely bluffing on a board like this, because it is so good for your range. If the weakest hand in your range is JJ, well, I guess you might as well turn it into a bluff, but that isn’t going to be terribly profitable. You may not realize it, but a very tight range is missing out on profitable bluffing opportunities for lack of hands that strongly prefer bluffing to checking on a board like this. Playing 87s may not cause you get more action with your strong hands, but it will enable you to profit from your opponent’s folds. This, too, deserves to be called “implied odds”, though it usually isn’t.
More useful terminology has recently entered the poker lexicon. Videos from Ben Sulsky, Sean Lefort, and others use a variable, r, to represent the percentage of its equity that a particular hand can realize after postflop play. This is often used in analysis of blind defense, to determine exactly how little of its equity a hand can realize while still being a profitable call.
In this case, however, we are looking at a hand that will have r > 100%. This is because 87s can win several times the pot on boards where it has near-100% equity and also have a reasonable chance of winning the pot on boards where it has near-0% equity. Thus, even though its hot-and-cold equity is poor relative to other hands in your opening range, the overall composition of that range gives your opponents incentive to play in certain ways against you, which makes it possible for a few weaker hands in your range to outperform their equity.
You have to be careful not to take this too far. If your opponents are so loose and poor at hand reading that you can play a very tight range pre-flop and still happily put your stack in with AQ on AT5r, then 87s will probably not have the implied bluffing outs discussed here, and you may be better off not playing it in early position, especially if your opponents are also very loose pre-flop (and such opponents almost always are). Likewise if you are going to pass on the profitable opportunities that playing this hand presents because you are averse to bluffing. I don’t think you’ll show a profit playing this hand if your only goal is to make two-pair or better and then hope to win a big pot.
There’s room to include only a few such hands in your range – if you go overboard, then you risk exploitation, quite possibly unintentional, by loose opponents, or by passive opponents who don’t attack you on 654r. The stronger suited connectors are the best candidates for this. Their raw equity is higher than that of lower suited connectors or suited gappers, and their ability to realize equity is better than that of off-suit broadway hands like ATo, which tend to be strong or weak on the same sorts of boards that are already good or bad for the majority of your range.
As one commenter pointed out, PokerSnowie does in fact advise folding this UTG at a six-handed table with 100BB stacks. That seems excessive to me, but with sufficiently skilled and aggressive opponents (getting 3-bet when you’re holding 87s sucks), I could see this being correct. In practice, it’s a hand I play 100% UTG in, say, a $1/$2 ZOOM game with a $200 stack. At 250BBs, I think the case for playing it only gets stronger, and it’s stronger still with typical live opponents whose three-bet ranges are, shall we say, sub-optimal.
How will your opponents know you dont open 78s utg?
If they bluff you more on low boards you can bluff catch more.
If they thin value bet more on low boards you can bluff raise more.
You can still bluff with kq kj qj on at5r board.
These hands have overcards to the second high card on board and have
A draw to the nuts. Why do you need 78s to bluff there?
Good questions:
“How will your opponents know you dont open 78s utg?”
They might not. This is why I emphasized that playing the 87s is not an advertising or loss-leader play. In many cases their exploitations of you are not deliberate. It’s just that they correctly fold JT to a bet on ATx, and you could profit from that if you saw that flop with more air.
“If they bluff you more on low boards you can bluff catch more. If they thin value bet more on low boards you can bluff raise more.”
Not necessarily, because they can also value bet more. IF your range is big Aces and big pairs on 456, your opponent can both bluff you and confidently play hands as weak as 65 for value. If you try to open a bluff-heavy 3-bet range on that flop, you get owned by people who jam like A8 or 54 or 76, not to mention the frequency with which your opponent will have a legitimately strong hand.
“You can still bluff with kq kj qj on at5r board. These hands have overcards to the second high card on board and have A draw to the nuts. Why do you need 78s to bluff there?”
Hands like JT and 99 still have a pretty trivial fold on AT5 even if these hands are in your range. KQ is actually still a pretty strong hand to turn into a bluff on ATx (a good chunk of folding ranges will consist of weaker hands), and if your range consists only of Ax and hands with 10+ outs, then JT and 99 still aren’t indifferent to calling unless your bet is very small. Basically I think your range is strong enough to support a wider bluffing range than just very strong semi-bluffs.
Back in the day we used to call this Shania.
I like the analysis, but there is one thing that does puzzle me. The idea of R is interesting, and something I’ve seen some of the sit and go wizards talking about in the last year or so, and indeed, playing with CREV led me to think about what ‘my share’ of the pot was in any given spot, which is related I think.
But I cannot for the life of me see why the preflop equity of a hand has any substantial informational content, when we’re talking about poker with any decent amount of cash behind, so it seems to me to be a weird framing effect to use it in building a metric for analysing our decisions.
Ultimately we care whether our share of the preflop pot, given all the flops and lines and so on, is greater or less than what we have to contribute to it. So it’s an ‘is X > Y’ question, and we’re transforming into ‘is X=r*Z>Y?’; these transformations are helpful if the new variables provide new information or insights (a great example of a good transformation is in option pricing, where the fiction of volatility allows comparison of otherwise incomparable instruments, a good example of a completely useless and even counterproductive one was the whole non-showdown winnings debacle).
My question is whether the Z (and thus the r) are really all that helpful in making preflop decisions. Things like how often we have to fold, and implied odds and flopability, and so on, are what make the decision hard, and I’m not sure that the fact preflop equity is anything other than weakly related to the things that matter.
(Sorry if this is pedantry &/or idiocy.)
I remember Shania. If I recall correctly, though, the explanation there was more of the advertising/loss leader variety. The example I remember involved comparing a range of {AA} to {AA, 72o}, with the argument being that you would lose money with the 72o but not enough to compensate for how much you’d gain with the AA. That shouldn’t be true at equilibrium, as Villain could do better by just reverting to whatever his old strategy was when we didn’t have 72o in our range at all. That strategy would be less profitable than it as, but still more profitable than whatever weird adaptation he’s making to the {AA, 72o} range.
As for the value of ‘r’, I think it’s mostly answering an objection. When people see weird peels from the BB, they say, “Ah, yes, but won’t it be hard to play post-flop?”. R provides a way to quantify that. Often, the answer is “yes, but not that hard.” I do think there are some cases where it can be estimated based on empirical evidence and where we can talk about factors that increase or decrease r. I find it a useful vocabulary to have (and more sensible than “Shania”!).
“This seems to imply that opening 87s is -EV, but we should do it anyway. This should never be the case. The route to optimal play is to make the most +EV decision at every opportunity.”
I assume by “optimal” you mean GTO/balanced. Of course -EV decisions are never part of a GTO strategy, but we know that GTO isn’t always the most profitable way to play. Do you think there are valid exploitative reasons for making a decision that isn’t the most +EV immediately?
An example might be if you have 72o on BTN and both blinds are folding 85% of the time to steals. The most +EV line for this hand would be to steal, but if you’re stealing 100% of your range, your opponents are more likely to find and fix their leak, taking away a big source of profit for you in the future. How do you weigh the long-term EV of them continuing to fold too much versus the immediate EV of exploiting them at every opportunity?