[Pokey]

Sorry for my extended hiatus from the forums, everybody -- real life came nipping at my heels a few months ago and I wound up giving up poker playing and poker writing for a bit. The past few weeks I've been dipping my feet back in the soup, and I think I'm finally ready to go back to it. Rather than just dribbling back into the forums, I figured I'd come back with another PSA. You can thank z28dreams for this one: it was his suggestion. He PM'ed me the following:

[ QUOTE ]

I believe that my biggest "leak" is in losing value on the turn or river though, but I'm never sure where I should be betting.

Could you a handful of examples where a river bet might be appropriate, but a typical USNL player might miss it? I'm not talking razor-thin value bets, but common spots that we should always be betting. Examples might be mid-pocket pairs on low boards (99-TT), top pair weak kicker, or strong hands that end up facing a 3-flush/paired/straight-completing board.


[/ QUOTE ]

I think this is an outstanding question, and it's the kind of question that all uNLers should be asking themselves regularly.

So, without further ado, let's try and find a roadmap to Valuetown.

<font color="blue">When should I bet the turn?</font>

As a general rule, there are a few good times to bet the turn:

1. When your hand is strong enough.
2. When your opponent is likely enough to fold.
3. When neither #1 nor #2 applies, but TOGETHER they make betting +EV.
4. When your hand needs protection.
5. When you are in position and want the option of a free showdown.
6. To better disguise your hand.

Let's look at these one at a time.

1. When your hand is strong enough.

Could this be any more vague? Probably not, but it's as good as we can get. "Strong enough" is going to be opponent-specific. A few things to consider:

- betting for value should happen less often against multiple opponents than heads-up. The strength required to consider a hand "strong enough" is going to go up as the number of opponents rises, simply because the odds of some sneaky lil' monster increases with the number of opponents out there.
- betting for value should happen less often when your opponent is tight. This one is obvious -- if your opponent is tight he's not going to call without something, and when you get called you're less likely to be ahead.
- betting for value should happen less often when your opponent is aggressive. While we think of aggro-monkeys as always wanting to get involved in huge pots, the reality is that most of them look so aggressive because they're always either betting ... or FOLDING. If you bet into a maniac, he's likely to release and move on to the next hand. If you check into him, he's far more likely to bet it for you, at which point you can snap off a bluff or two (or three, or four).
- betting for value should happen less often when the board is dry. You've got A7 on a board of J732 rainbow; what the hell is your opponent going to call with? Nothing that makes you happy, that's for sure. Now if the board were 9872 with two hearts, you've got more chances of getting a call from someone fishing for a hand.
- betting for value should happen less often when you want a small pot. Generally speaking, you want a small pot if you have a weaker hand, because if two stacks go into the middle and you've got TPTK you've usually made a mistake. Before you build a pot, make sure you WANT a big pot.
- betting for value should happen more often against calling stations. With JJ on a Q942 rainbow board, you don't want to build a pot against a tough and clever opponent, but against a calling station you're WAY ahead of his range. Make him suffer for being too loose.
- betting for value should happen more often when your opponents are likely to be drawing. If you know your opponent is the kind of person who never folds suited cards, go ahead and value bet any pair on a two-suited board. Just be sure you can fold if the third card falls....
- betting for value should happen more often when your hand is subtle. If you raised 54s UTG and the flop comes 876 rainbow, start shooting -- nobody is going to see you coming, so you're very likely to get called by weaker hands.
- betting for value should happen more often against bad players. Bad players don't understand hand values, don't understand when to fold, and don't understand when they are beaten. Charge them for the lesson.

In short, value betting should happen not only when you are sure your hand is best, but when your hand beats villain's range.

One more point: "raising for value" is harder than "betting for value," because raises often shut down your opponents, either causing them to fold and stop spitting money into your pot or causing them to check-call, adding far less to the pot than if you'd let them keep firing.

2. When your opponent is likely enough to fold.

Again, this is going to be opponent-specific and board-specific. Some important points:

- folding equity is stronger when a scare card falls. If you check-called your 88 on a board of Q74 with two hearts, consider betting when that K[img]/images/graemlins/heart.gif[/img] hits on the turn, especially against the kind of opponent who doesn't bet draws (you know which ones I mean).
- folding equity is stronger with better players. While bad players call too often, good players often fold too often, especially at uNL levels. Typically bad players are calling stations and "good" players are weak-tight. Make use of that, and force them to fold on the turn.
- folding equity is stronger if you've shown strength earlier in the hand. A preflop raise, followed by a flop bet, followed by a turn bet is EXTREMELY SCARY for most players, especially ones who "understand" continuation betting. You'll often find players who "test" the preflop raiser to see if he'll keep firing on the turn; against this sort of opponent, a "second barrel" often has ridiculously high folding equity. (SIDE NOTE: against this type of opponent, checking behind on the turn with a good hand is often extremely profitable, since they will either try to bluff the river or check/call your "obvious bluff" on the river. The preflop raise / flop bet / turn check / river value bet can be a very potent combination if used sparingly and if used against the right type of semi-thinking opponent.)
- folding equity is stronger when the turn could have logically improved your hand. Say you were in the big blind, the flop comes A72, you checked through, and the turn is another 7. A bet here is going to have some solid folding equity -- it's unlikely that anybody has an ace (the flop was checked through), and that second seven is going to put the fear of trips in your opponents. Not only was it an unlikely card to have improved anybody, but it's also going to get hands like 88-KK to fold quite often. Aces will often still call you down, but that's a risk you've got to take sometimes. Be realistic, though: if you were the preflop raiser, the flop was 765, and the turn was a 3, people aren't going to believe that you've got a 3....(this might be a good time to value bet your pocket pair 99+, though -- just strongly consider checking behind on the river if your opponent calls this turn bet.)
- folding equity is stronger when the board matches your "likely holding." People typically assume the blinds have crappy random low cards and people typically assume that the preflop raiser has AK or a strong pocket pair. Don't pretend that the board pairing deuces helped your hand when you've got naked overs, and don't pretend that the KQJT board matches your big blind -- you're not going to get folds.
- folding equity is weaker in a multi-way pot. Don't rely exclusively on folding equity here unless you've got a good reason to do so.
- folding equity is weaker when your opponents are calling statiosn -- enough said.
- folding equity is non-existent if your opponent's hand is strong. He's never folding the flush. He's never folding the straight. He's never folding his set. Don't even try.

3. When neither #1 nor #2 applies, but TOGETHER they make betting +EV.

This is commonly called "semi-bluffing." You don't have a strong enough hand to warrant a bet (usually you're on the draw). You don't have enough folding equity to be +EV (you think your opponent probably has a good hand). But together, the two sources of value make the play +EV.

Example:

You've got 6[img]/images/graemlins/club.gif[/img]5[img]/images/graemlins/club.gif[/img] in the big blind. CO (TAG) raises to 3xBB, button calls, SB calls, you call.

Flop: A[img]/images/graemlins/diamond.gif[/img] 7[img]/images/graemlins/club.gif[/img] 9[img]/images/graemlins/club.gif[/img]

SB checks. You check. CO bets 8xBB. Button folds. SB folds. You call. (Not necessarily your best play, but let's say this is what you do for the example.)

Turn: 9[img]/images/graemlins/heart.gif[/img]

You check. CO bets 20xBB. You should consider pushing.

Let's say that your opponent either has an ace (70% of the time), a big pocket pair (20% of the time), or junk (10% of the time). Smooth-calling for flush value is pretty weak, here, because you're out of position and unlikely to get called by a hand that you beat if you manage to hit your flush. The pot equity isn't here and the implied odds are pretty poor, so you can't just smooth-call and hope to get lucky on the river. Folding is an option, but you've got lots of outs to a very likely winner. Also, your opponent doesn't NECESSARILY like his hand right now! If you push, things change around quite a bit.

First off, your push will be a big enough bet to put the fear of jeebus into your opponent. If he folds, you walk away with a 48 BB pot with nothing but potential and guts, and that's always a nice situation. If he calls, you've still got twelve outs to a very likely winner, and that means you'll win money some of the time.

Now, if you KNEW your opponent would call 100% of the time and be ahead, pushing is -EV (your hand strength alone doesn't make this a +EV move). Likewise if you KNEW your opponent would win 100% of the time at showdown, pushing is -EV (your folding equity alone doesn't make this a +EV move). However, sometimes villain will fold the best hand (value for you) and sometimes villain will lose the hand when he calls (value for you). Often, the COMBINATION of these two bumps the play into +EV territory.

Here you're risking about 80 BBs on the push -- if your opponent folds, you win ~50 BBs. If you get called, you'll win ~120 BBs about 1/3rd of the time and lose 80 BBs the other 2/3rds of the time. Let's say your opponent folds 25% of the time, here.

EV from push equity: 1/3*(+120) + 2/3*(-80) = -13.3 BBs. This happens 75% of the time, so the total EV here is 0.75*(-13.3) = -10 BBs.
EV from fold equity: 50 BBs. This happens 25% of the time, so the total EV here is .25*(50) = 12.5 BBs.

EV from pushing is +2.5 BBs.

Note: if you always lost when you got to showdown, the EV would be 0.75*(-80) + 0.25*(50) = -47.5 BBs.
Likewise, if you were always called when you pushed, the EV would be -13.3 BBs.
It is only because of the combination of some folding equity AND some showdown equity that this move is +EV.

For semi-bluffs to work, the following must all be true:

- You must have outs. With outs, you can win at showdown if you are called.
- You must have some folding equity. If your opponent never folds, you can't be semi-bluffing.

4. When your hand needs protection.

This reason to bet is misused more than any other. You raised preflop 4xBB with KK. The flop comes A[img]/images/graemlins/heart.gif[/img]K[img]/images/graemlins/spade.gif[/img]7[img]/images/graemlins/diamond.gif[/img] rainbow, you bet 6xBB (ooooh, sucker him in!), your opponent calls.

The turn is Q[img]/images/graemlins/heart.gif[/img] putting two hearts on the board. Your opponent checks, you push for 90 BBs, your opponent folds.

"Whew!" you say. "I protected my hand from the draw!" The reality is that "protection" is an excuse for timid play, here.

And yes, pushing can be TIMID: you know your opponent will fold, and rather than trying to extract value while you're WAY ahead, you go for the safe win. What hand are you afraid of, here? JT? Do you really think JT is folding here?? Or maybe you think your opponent was calling you with J[img]/images/graemlins/heart.gif[/img] 7[img]/images/graemlins/heart.gif[/img] and just improved his hand to a superdraw. Unlikely as this may be, you're still ahead -- make him pay for his draw! Don't chicken out just because your hand isn't the absolute nuts.

Here's when your hand needs protection:

- When you are ahead, but many, many hands could destroy you on the river. If you've got 99 and the board is now 8[img]/images/graemlins/heart.gif[/img]5[img]/images/graemlins/heart.gif[/img]4[img]/images/graemlins/spade.gif[/img]3[img]/images/graemlins/spade.gif[/img], you've got a reason to protect. If you've got AA and the board is 8754 rainbow, you're not protecting anything against anything -- bet for value or check behind with an unusual read.
- When you are out of position, be more likely to protect your vulnerable hands. You don't want to fold the best hand, so getting an opponent to fold when you're OOP is often a good thing.
- When the board is extremely draw-heavy.
- When your opponent likes to bluff.

These last two combine in some pretty spectacularly ugly ways at times. For instance: say you've got black aces and the board is J[img]/images/graemlins/heart.gif[/img]T[img]/images/graemlins/heart.gif[/img]6[img]/images/graemlins/diamond.gif[/img]5[img]/images/graemlins/diamond.gif[/img] and your opponent is loose and aggressive, tricky and bluff-happy. Damn near every card in the deck is a scare card for you: villain could complete a straight with any 2, 3, 4, 7, 8, 9, Q, K, or A.

Also, any red card could complete a flush. Also, any J, T, 6, or 5 could give villain trips. That means that -- literally! -- any card in the deck is a potential scare card on the river. You know that your opponent is likely to have somewhere between zero and 15 outs, but you have no idea which ones are the magic ones. Given your opponent's style, you're going to have to call his river bet out of fear that it's a bluff. So what do you do? You bet the turn -- HARD -- to protect your hand. Doing so reduces his implied odds if he hits and charges him a big price for hunting. It also makes you look committed, decreasing the likelihood that villain chooses this hand to get all tricky with a bluff. Feel free to bet the pot, here.

Don't go nuts with your protection bets: after all, you DO have the best hand -- that's what you're protecting, remember?

Make a bet that your opponent can talk himself into calling, but that makes calling -EV for him. After all, we're trying to win MONEY, not POTS.

5. When you are in position and want the option of a free showdown.

Here's another good one that uNL'ers often forget. You've got 88, so you raise 4xBB preflop. SB and BB call. The flop comes A74 with two hearts. The blinds check, you bet 9xBB, SB calls, BB folds. The turn comes with a black 2 and SB checks. What to do? Many uNL'ers will check here, hoping to get a cheaper showdown. Of course, this could create some very tough decisions for you on the river. What do you do when the river card is a heart, or another A, or a 7, or a 4, or a 2, or an overcard, or something that looks totally blank, and SB bets into you? Well, you cry, because you've got no idea what's going on.

However! If you bet the turn, look at all the good things that happen:

- You charge the heart draw to see another card.
- You get more value from a stray 7 (or 4!) that happens to be overly optimistic.
- You might fold out 88-KK, getting them to finally believe that you have an ace.

Even if you are called on the turn, when the river card comes, you'll very often get a chance to check behind and see how the cards look. You get to set your own price for the showdown, losing very little to Ax or overpairs, and winning unusually much from weaker hands and draws. In a sense, this reason combines protection, bluffing, and pot control into one reason.

Make sure that your betting strategy is internally consistent and well-disguised -- don't automatically slowplay the turn with stronger hands and automatically bet weaker hands on the turn, or you'll be obvious enough that you'll induce a ton of turn bluff check-raises that you won't like. In other words, bet your strong hands on the turn but ALSO bet the weaker stuff on the turn sometimes.

Don't try this move on tricky-trappy opponents -- they love to smooth-call flop bets with monsters and then check-raise the turn. You'll hate it. Against these opponents, checking behind on the turn can be the better move, particularly if you've got outs to improve (even if it's only a few outs). Luckily, these folks are relatively rare. Against the typical opponent, calling the preflop and flop bets and check-raising the turn means you can safely let your hand go, especially if you're making this kind of play with very few outs to improve (like a pocket pair).

6. To better disguise your hand.

Here, the idea is that checking the turn would give away too much information about your holdings. The most obvious example is when you're OOP with a flush draw -- check/calling the turn and leading a river flush card is equivalent to renting billboard space to announce your hand to your opponents. However, leading out on the turn makes it much sneakier if and when you hit on the river, and makes it more likely that you get paid off handsomely when your ship does come in. Also, if your turn bet is a continuation of previous aggression you will have the added benefit of folding equity to make your move more +EV. Many people would consider this a variety of semi-bluffing, but the intent is very different here: should your opponent choose to fold on the turn you wouldn't be too upset, but this bet will not be all-in, and the purpose of this bet is not to fold your opponent. Rather, the goal is to make your river payoff much larger if you are lucky enough to hit your gin card.

Semibluffs want folds; disguise bets want calls.

----------

<font color="blue">When should I bet the river?</font>

For river bets, things are a bit simpler: now, your value comes from only three sources:

1. Getting your opponent to fold when he has the best hand.
2. Getting your opponent to call when he has the worst hand.
3. "Value bluffing."

Once again, let's go through them one-by-one.

1. Getting your opponent to fold when he has the best hand.

At the river, you either have the best hand or you don't. If you don't have the best hand then your only hope of leaving with the cash is to convince your opponent to fold. River bluffs take lots of courage, but if they are done properly they can be extremely profitable. Unfortunately, most people either bluff the river WAY too much (spewing cash left and right since their opponents know not to fold) or WAY too little (never betting unless they are fairly sure they are ahead, and therefore rarely getting called by non-monster hands). How do you decide if a river bluff is a good investment?

First off, we should bluff rivers more against good players than against bad players. Good players can see that they are beaten; bad players cannot see beyond their own cards. If a bad player likes his hand, he's not folding -- it doesn't much matter what you say about your holding. Don't bluff the calling stations -- it's an expensive habit.

Secondly, bluff rivers more when your table image is tight and strong. If you've been showing down monsters for the past hour, getting involved in very few hands, betting strongly, and flashing nothing but the nuts, feel free to throw your weight around a bit on the river. You're air-tight play has earned you mountains of folding equity; cash in by bluffing more freely at the river. However, if you've been mixing it up left and right, and if you were caught betting with air a few times (especially against THIS opponent!), feel free to check behind and lose a smaller pot rather than a bigger one. If your table image is crap, your folding equity is floating in the toilet right next to it -- understand this, and save your money for when you get a hand.

Third, bluff rivers more when your betting tells a logical story that means you've got a good hand. Your opponent limp-calls preflop. The flop comes KT8 with two hearts -- your opponent checks to you and calls your bet. The turn is 4[img]/images/graemlins/spade.gif[/img] and your opponent check-calls your bet. The river is 3[img]/images/graemlins/heart.gif[/img] and your opponent bets. He's telling a consistent story: "I've got a speculative hand ... I've got a flush draw ... I've still got a flush draw ... I've got a flush!" This is the kind of bet that has TONS of folding equity, even if your opponent's actual holding is 9[img]/images/graemlins/spade.gif[/img]7[img]/images/graemlins/spade.gif[/img]. Too often, though, we tell an INCONSISTENT story: we raise preflop. The flop comes 876 rainbow, our opponent checks and we bet. He calls. The turn is another 7. Our opponent checks and we check behind. The river is a 3, our opponent checks and we...bet? What strong holding is actually consistent with this betting strategy? Our opponent knows we're not doing this with an overpair, he knows we don't have a straight, and he knows we don't have a 7. Our hand screams of missed overcards, and a savvy opponent is going to call with his T6o and piss us off as he rakes the pot. "How can you call with that crap??" we scream, but the answer is obvious: "because you only make that play with air." Make sure that your bets make sense. So if you raised preflop, bet the flop, and checked the turn, bet a river scare card: an ace, a third suited card, possibly a king (on an aceless board) -- that sort of thing. Don't pretend that the river 3 made your straight when the board is now JT653 -- nobody's going to believe it. Conversely, don't pretend you were limped preflop and were check-calling all the way with AK; it's just not credible.

Finally, bluff rivers more when you've got a read that your opponent knows how to find the "fold" button. This is related to the first point, but still separate enough to warrant mentioning. Some ultra-tight nits cannot ever fold after raising preflop -- they play so few hands that they are simply incapable of letting go of a hand once they've got one. Know this, and avoid bluffing them. Alternatively, some maniacally aggressive LAGs will release a hand the moment they face pressure; fight against them. Against these folks, river raises or check-raises are often the most profitable play you can make. Use it.

2. Getting your opponent to call when he has the worst hand.

Ah, the art of poker. You've got the nuts -- the absolute, unburnished nuts! Now what? Well, now you've got to figure out how to get paid. One of the greatest difficulties in poker is deciding on bet sizing. We've got a dilemma: the bigger our bet, the bigger our payoff when we get called but the less likely our opponent is to call us. How do we maximize value when we think our hand is best? We pick our bet size carefully. Some general hints:

- Bet bigger when our table image sucks. If we've been playing crappy poker, or if we've been unlucky enough to get caught stealing a few times, or if we've not shown down any winners lately, our table image will look bad. This is especially true if this particular opponent has caught us stealing recently. Feel free to make large bets -- it'll look desperate and scared, and our opponent will call more often than usual.
- Bet bigger if you're "on tilt." I put "on tilt" in quotes because I'm assuming you're NOT on tilt (we're beyond emotional play, right? [img]/images/graemlins/grin.gif[/img] ) No, what I mean here is that you can bet bigger if your play will *look* like a tilt play to your opponent. The last hand your nut flush lost to a rivered full house? You're on tilt, baby! If you're lucky enough to catch pocket rockets this hand, play it to the bone -- nobody is going to believe you, and you'll be paid off because "he's tilting off another stack; I'll bet my 3PNK is good!"
- Bet bigger if your opponent is bad and/or unobservant. Against a good opponent or a watchful one, you can't get away with sizing your bets based on your hands. But against an idiot you can bet big with a big hand and small with a small hand and suffer no negative consequences. Feel free to make these kinds of plays against the hopeless goobers sitting at your table; just be sure you behave yourself when a smart player is in the pot.
- Bet bigger if your opponent sees too many showdowns. This is a useful number to have in your HUD for this reason alone.
- Bet smaller if your opponent is timid. A weak-tight nit is not going to pay off a pot-sized river bet without a strong hand; make sure you give him enough odds to talk himself into a crying call.
- Bet smaller if your opponent couldn't possibly have a big hand. If you've got AA on a board of AAQ72, you simply can't get away with betting much under normal circumstances. Make a small bet and be happy that you got paid off at all.
- Bet smaller if your opponent likes to raise or check-raise weakness. Some opponents LOVE to push people off their hands on the river; if that's the case, do whatever it takes to induce a bluff. A common betting pattern that might do this at uNL is "raise preflop, 3/4ths-pot flop, 1/2-pot turn, 1/2-pot river." That small river bet can induce a maniac to come over the top with a bluff raise, hoping to drag down a big'un. This is especially true if the "obvious draw" misses on the river.

(Note that many of these hints work the other way around for sizing bluffs.)

3. "Value bluffing."

Of all the ideas in this post, value bluffing is probably the least understood. It's the river equivalent of a semi-bluff: a bet that is not +EV as a bluff, is not +EV as a value bet, but is +EV as a combination of the two. How can you possibly have a situation at the river where your bet is "part bluff, part value bet"? Well, because your opponent isn't holding a hand; he's holding a range.

I'll repeat that, because it's a really, Really, REALLY important concept: your opponent isn't holding a hand; he's holding a range. Anybody who says he can size up an opponent -- look him in the eye, stare him in the soul -- and identify his exact hand holding ... well, that person is a liar. It's what good players say to freak out their opponents.

The truth is that unless an opponent is woefully straightforward in playing style, you won't know their exact holding. Best case, you'll know their range. Of course, this works the other way around, too: your opponent can't identify your exact holding, either. Now, the combination of these two scenarios leads to the interesting (though unusual) situation of the value bluff.

Let's say that your opponent is an ultra-loose but not completely stupid player. He's not a "good LAG," because they don't exist at uNL (or SSNL, for that matter). Rather, he's a "semi-thinking LAG." A while back, someone coined the phrase "1.5th-level thinker" to represent a player who thinks about your hand but always puts you on AK. It was a joke, but beginning hand readers will often do something very similar to this, putting you on an overly narrow hand range and sticking to it despite any evidence to the contrary. So, let's say this is the kind of opponent you have: loose enough that you can't predict his holdings all that well, smart enough to try to put you on a hand, but amateurish enough to do a really bad job of it.

The current board is AJ974 with no flushes. You have QJo and you attempted a steal. At the river, you've got second pair.

Your opponent's range is quite wide, but you think there's an 75% chance you're beaten. If you bet the pot, there's a 1/3rd chance he'll fold a better hand than yours and also a 1/3rd chance he'll call with a worse hand than yours. How does this work out?

He folds 1/3*(75%) + 2/3*(25%) = 41.7% of the time. Here you win 1xPot.
He calls with a worse hand 1/3*(25%) = 8.3% of the time. Here you win 2xPot.
He calls with a better hand 2/3*(75%) = 50% of the time. Here you lose 1xPot.

Your net is 0.417*(+1) + 0.083*(+2) + 0.5*(-1) = +0.083xPot, making this a winning strategy.

Note that this strategy doesn't work as a bluff -- villain only folds 1/3rd of the time, and you're risking the pot on the bluff (you lay 1-to-1 odds on a bet that's 2-to-1 against you winning). Note also that this strategy doesn't work as a value bet -- villain's hand beats yours 75% of the time, but you're only getting twice your money when you win (you're getting 2-to-1 odds on a 3-to-1 longshot). In other words, the bet is -EV as a bluff, -EV as a value bet, but +EV as a combination of the two. Villain's sloppy hand-reading combined with villain's unidentifiable hand makes this a fuzzy enough betting area to be +EV.

----------

If you've slogged through this all, I hope you picked up a trick or two. Good luck, and I'll meet you at Valuetown.