Although it often seems like NLHE and PLO are the only kinds of poker, the draw and stud games that came before them still see a decent amount of action online. While chances are nobody you know plays any of them particularly well (or at all), there are still a few professionals out there who specialize in Stud, Stud/8, and Razz. Many more play high stakes HORSE, where someone –EV in the stud games will find it very hard to be a winning player.

As the simplest of the three Stud HORSE games, Razz should be the one that has the highest average level of play. Many NL pros, who know big-bet math and theory cold, really struggle with Razz, due to a lack of understanding of the fundamentals (to be fair, the ratio of NL learning tools to Razz is about several hundred to one). In addition, being good at NL is not the same as being good at poker; while good poker players absorb the fine points of the game intuitively, those who are simply good at no limit miss the 2-3 plays a session which turn a losing Razz player into a winning one.
The problem when writing this article, though, is pure Razz strategy may only appeal to about eight people. So today, I want to write about Razz as a way of demonstrating lessons to carry back to NL. The easiest way to do that is to discuss a play that is ubiquitous in every poker game – the bluff.

In well-played NLHE, bluffing is not just an integral part of the game – it’s mandatory. It’s easier to c-bet a ton of flops than to actually flop pairs, and flop bets probably account for most of the pots you win. Even at micro-stakes these days, it’s probably impossible to be an NL player who rarely bets without a pair and win money. At the same time, it’s extremely difficult to bluff well; there are a few $2000 E-books out there by world class poker pros that are worth the price, because they explain how to evaluate board textures, rank them against your opponent’s range, and use that information to fire a second and third barrel only when the other guy is unlikely to call. Even these E-books and top-flight coaches don’t automatically transform students into experts overnight, because bluffing in NL is a remarkably hard task with extremely randomized feedback. PLO is another level of magnitude above NLHE; you could probably write an entire chapter in a long book simply on when and how to bluff with an ace in your hand on a board with a flush.

Razz is much easier to play. Before the river, you know two thirds of the other player’s exact hand, and if he’s playing a TAG style, you can predict your equity vs. his range down to a few percentage points. This makes it incredibly easy to bluff in small and medium pots (since it’s a limit game), but harder to bluff in large ones (since it’s a limit game).

There are times when it is still possible to bluff people off the best hand in big pots, and this is when the Razz pro who wins a 7 BB pot 50% of the time by bluffing correctly can crush an otherwise decent NL pro. The second player, who has the same winrate in every other situation but misses 3-4 hours’ worth of effort by always losing that one huge pot, may not even be a winner in a tough game. The reason this happens – and it happens a lot – is because aggressive players who are good at NL but not poker often correctly follow the math, but forget the most important thing in poker: how to properly sell one’s hand to an attentive audience.

As an example, I’d like to take you through a pretty standard Razz hand. I’m assuming you guys know most of the basics simply because you’ve made it this far into an article with “Razz” in the title, so I’ll skip over basic explanations like “he probably has a three card eight,” etc. This article does assume a particular kind of player, though, so don’t try this against just anyone.

On third street in an 8 handed game, a king brings it in, two people fold, a 3 raises in middle position, three more people fold (the dead cards are irrelevant), and you decide to call with 857 (the 7 is showing) on the button. On fourth street, you catch a five while the 3 catches a 9, so you bet your 8575 into his xx39 hoping he’ll fold, but he decides to call. On fifth, we catch a six (85756) while he catches an eight. We bet, but after taking a while to act, he finally calls with xx398. This means he is likely a bad or at least ignorant Razz player…but possibly a thinking player calling because we called his raise on the button. In his mind, we are playing a lot of hands and have a wide range. In this case, his stubbornness is helping him, because he most likely has a made 98 by now and we are about 70/30 underdogs vs. his hand.

On sixth, we catch a 9 (857569) while he catches a king. At this point, although our 98765 is guaranteed to be worse than his 98, we’re showing xx7569 vs. xx398K. We’re never getting raised, and even if he doesn’t fold now, he often will on the river, so it’s time to make the automatic two barrel bluff. Right?

Every NL pro I’ve ever seen playing Razz will always bluff here, going purely off the simple equity calculation (“I’m not drawing dead”) and instincts saying “he’s bound to fold way more than I need to make this work.” But automatically bluffing here, while profitable, doesn’t actually sell your bluff.

Think about this hand from the other player’s point of view. If you are this guy and you called fifth street with A2398, you should now be sure you are behind, but you’re not going to fold yet. The pot is now big enough that you have the odds to call sixth and fold the river unimproved, and when you do catch a 4, 5, 6 or 7, you’re going to call that one last bet. Depending on the dead cards, that can be over a third of the deck, so, when we do bluff both sixth and seventh (rest assured we’re betting the river whether we improve or not), we automatically accept that we’re going to lose the pot over a third of the time.

But there’s a way to win this pot nearly 100% of the time – by not bluffing sixth, especially if we have a reputation as tricky or even prone to Fancy Play Syndrome. Imagine we check here, instead. What does someone checking xx7569 into xx398K tell our opponent? It’s either that he’s already drawing dead, or we’ve paired somewhere and he’s ahead. Most of the time, though, a thinking player will assume the first one; after all, if we’ve paired, why wouldn’t we just bet again?

So in this case, simply by delaying our bluff one street and then automatically betting every river, we’re acting like we are (badly) slow-playing a monster, and there’s a good chance that on the river, our opponent will laugh at our “mistake” and fold hands that have actually improved to an eight.

What if he decides we’ve paired and manages to bet into us on sixth? That shouldn’t really ever happen, but if it does, we’re instantly check-raising and following through on the last card. This will make him fold just as often as simply betting would (60%+) even if he doesn’t insta-fold to the CR, but with one more BB in the pot, it’s even more profitable.

Pulling off this kind of play just once, especially if there were a couple of other people in the pot who folded somewhere, represents up to a hundred hands’ worth of EV for a 2 BB/100 winner. It’s a huge deal to win a pot this way in any kind of limit game. But how does it translate back to NL?

I’ve frequently heard the line that most NL players don’t bluff enough. That’s not entirely accurate, because as I’ve said, everyone in NL bluffs all the time. But it’s true that very few people bluff well in big pots. A lot of your more aggressive opponents (especially in PLO) use naked aggression on every street to try to run you over, and these days, everyone is becoming increasingly aware of bet sizing as a function of stack sizing, but it’s pretty rare to find someone who is good at telling opponents believable stories about their hands.

The latter is what makes a decent player into a very good one.

-adanthar

Recent Scores for adanthar

<table width="90%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td><table width="100%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td valign="bottom"></td><td>$43,290.00 </td><td valign="bottom"></td></tr></tbody></table> </td></tr><tr><td>$2100 buy-in, SCOOP-13-H: $2,100 Razz, $150K guaranteed on PokerStars. 04/08/2009, 2 place for 43,290.00 </td></tr></tbody></table> <table width="90%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td><table width="100%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td valign="bottom"></td><td>$16,500.00 </td><td valign="bottom"></td></tr></tbody></table> </td></tr><tr><td>$300 buy-in, FTOPS Event #19 on FullTiltPoker. 05/14/2009, 3 place for 16,500.00 </td></tr></tbody></table> <table width="90%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td><table width="100%" border="0" cellpadding="0" cellspacing="0"><tbody><tr><td valign="bottom"></td><td>$2,636.70 </td><td valign="bottom"></td></tr></tbody></table> </td></tr><tr><td>$215 buy-in, Sunday Million [$215 NLHE - $1.5M guaranteed] on PokerStars. 11/23/2008, 41 place for 2,636.70 </td></tr></tbody></table>
---