Sizing Bets in No-Limit Holdem

The obvious difference between Limit and No-Limit Hold'em is the variability in the size of bets you can make based on the situation. While this ability changes the game, it is extremely important to choose your bet and/or raise size appropriately to accomplish the goals of No-Limit Holdem:

1) Force your drawing opponents into making a mistake by calling.

2) Maximize your profits from opponents when you have a made hand.

3) Minimize your losses in situations where you don't have the best of it.

These two goals are profound, yet contradictory. You want to bet enough so that your opponent will make a mistake (mathematically) by calling, but you want to keep him putting chips in the middle as an underdog.

First of all, you need to evaluate your opponents to accomplish this. The initial evaluation is that 3/4 of your opponents (at most small stakes games) have no concept of pot odds or anything else--all they know is that they have 4 hearts and another heart usually means they win the hand or that Norman Chad said on ESPN that 98o is a coin flip against 77.

Second of all, you need to use what you know from the above, and what you glean from your opponents play to make appropriate decisions on how to best approach your post-flop play.

Example: You're on the button in a $200 NL game with 8s8d. Two players limp, you raise to $8 and the BB and both limpers call your raise. There is $33 in the pot when the flop comes Jh-8s-3h. The limpers both check to you. What is your action??

Let's examine each potential action and analyze it's merit:

1) Check. This makes me scream almost as much as cold calling raises in LHE. Slowplaying can only get you in big trouble. You have a set. Right now, you're only behind JJ, but there are two potential draws out there to cause you trouble. There is a heart flush draw (Xh-Xh-Jh-3h) and a few potential straight draws--the open ended draw (Jh-Tx-9x-8s , for the player with T9), or some gutshots (Qx-Jh-Tx-8s, for the player with QT), and some player may actually even have both draws (if they have Th-9h, Qh-Th, Qh-9h). Checking here gives these drawing players free cards, and if I'm drawing here and you check, I'm just checking behind, ESPECIALLY considering your raise. Giving free cards to drawing players is the single worst error you can make in holdem. Don't get fancy here and go for a check-raise, because even if it works, and a player bets large enough and you raise enough to create a pressure point, now you may be playing a vulnerable hand for your entire stack against an opponent with potentially as many as 15 outs (which actually would make him a favorite to get there by the river)...Checking is very bad here.

2) Bet. Aggression is by far best here. But now, how much do you bet? It depends mostly on who or what draws you want to price out of the action (or what players you want to eliminate from the hand). First you need reads on players, so I'll give them here:

MP1 is a solid player by all accounts, VP$IP of 19%, PFR of 9%, AF-T of 2.1. He has $175 in his stack.
MP2 seems to call a lot, VP$IP 32%, PFR 4%, AF-T of 0.85. He has $140 in his stack.
BB seems to be quite loose, but aggressive, VP$IP 44%, PFR 19%, AF-T of 3.2. He has $260 in his stack.

You have $225 in front of you.

What does this tell you? MP1 may actually know what the hell he's doing. If you make a smart bet, and don't give him odds, he's probably smart enough to dump his hand. MP2 will be tougher to get to leave. The BB might re-raise you if you fire out, which is exactly the result you'd like. He'd fire with a draw, top pair, or just because he liked the way the sunlight hit his monitor.

Back to the topic at hand. How much should you bet? Let's put each player on a range of hands. MP1, being very solid and tight, likely has a hand like a small pair, AQ, AJ, KQ, KJ, QJ, or a suited connector (because he limped, which means he likes his hand a little bit, and called a raise, which means that he likes the odds he's getting to hit something). Does he have a hand like T9? Unlikely--he wouldn't think that hand is worth a limp-call. QT? Possibly. Could they be suited? Very possible. So, I have him limited to big Aces (AQ or AJ, likely offsuit), suited Broadway (two face cards), or MAYBE lower suited connectors. What could he have? He could have TPTK (you hope--AJ makes him top pair top kicker, the dream hand for an opponent to have vs. a set), a gutshot (with QT), or a flush draw (with two hearts). AJ has him drawing just about dead (since he needs either running Js, or running As. The gutshot or the flush draw are your worries here. A gutshot draw gives him 4 outs twice, a 16% chance of hitting, and 10.75-to-1 odds against hitting his hand. The flush draw gives him 8 outs (because the 8h gives you the stone cold nuts) twice, a 32% chance of hitting, and 4.875-to-1 odds against hitting his hand.

If you were playing him heads-up, this would be easy. You would pick the worst case scenario and bet so that he would be incorrect in drawing to his hand. Here the worst case is a flush draw (short of JJ, of course), and you need to give him worse that 4.875-to-1 odds on his money to force him into a mistake by calling.

He needs to get worse than 4.75-to-1 odds on his money to call. We can accomplish this by betting $9. This makes the pot $42, and makes him call $9 to win $42, which is only 4.67-to-1 on his 4.875-to-1 shot. This is an incorrect call for him, so if he folds, (we're talking heads-up here) you win the pot, and if he calls, you win because he's made a serious mistake in calling without odds to make the call.

Of course, we're NOT heads up, so it's a different situation. You have two other players to get through to get to this solid player, who is likely smart enough to fold to that bet anyway. Let's look at the BB, the aggressive player. He could have any two cards, but for him to call this raise, considering his general level of aggression, he probably has suited cards. Could he have Th-9h (the biggest draw here)? Yes. Plan for that. He has 14 outs here (8 hearts--except the 8h) to make his straight or flush, and he's actually a 56% chance to get there. He has 2.36-to-1 odds to make his hand, so you have to force him into a mistake with your bet size. The appropriate bet here (apologies for the algebra) is one that makes 2.36 x > 33 + x. To solve the equation for x, you come up with x > 24.26. If you bet $24.50 (3/4 pot), he does not have odds to draw to his monster draw and would be making a mistake by calling.

Since this player is the first you have to deal with, we'll pick up the action here. You bet $25, a nice round number, a little more than 3/4 pot. He can either fold (you win the encounter), call (a CRITICAL mistake, and out of character for such an aggressive player), or raise, because he either made a hand (two pair, other set), has over-valued his draw (flush, straight, or both), or thinks you're bluffing. A raise here by the BB will make this hand heads up, undoubtedly, unless MP2 is an idiot, or MP1 has a set.

We'll deal with the raise first. Since he's too aggressive to limp-call JJ, you have to assume you're miles ahead of a drawing opponent. Any raise he makes here to your $25 bet has to be at least $50, so your next move would be clear--all-in. It would again force him into another huge mistake. He would be calling $192 to win the $108 in the pot, $50 of which is his, and he would be calling it as a massive underdog. Again, unless he's not only an idiot, but a lucky idiot, you win the hand right here.

He won't just call that bet, so that's eliminated. He folds. MP1 sees your $25 bet and realizes he's behind. He folds as well, leaving you heads-up with the MP2 calling station. When a player is non-aggressive (passive, with an AF-T under 1.0), he will call more frequently, and usually call larger bets than their more aggressive brethren. If he does call this bet, you've forced him into another error. Let's say he calls, because that's what calling stations do.

Now there is $83 in the pot, he has $107 in front of him, and you have $192 in front of you. The turn card is another heart, and not the 8h. What do you do? If he has the flush, you still have 9 outs. How much of a bet can you profitably call after a check? You have 4.11-to-1 odds to make a full house on the river, which will give you the win, and most likely your opponent's stack. This is where implied odds come into play as well, but on face value, your opponent must bet at least $27.67 to make your call unprofitable (1/3 pot) on it's face value, but you also must consider the fact that if you call this bet, he will be pot committed and get the balance of his stack in on the river. So the pot's effective size here isn't $83, it's actually $190 due to the implied odds you're getting for him to get the rest of his stack in the middle. His bet to make you unprofitably call still has to be 1/3 the size of the pot based on the pot odds, so his minimum bet to push you out has to be at least $63, which would completely pot-commit him.

If he bets $63 or less, you can either call the bet or raise him all-in. If he bets more than $63, you can assume he made the flush and properly priced you out--fold the hand and walk away.

Now let's say that the turn card doesn't complete any draw--it's a blank. It's you and MP2 heads up, and the pot here is $83. What do you do? He already called a ridiculous bet on the flop, so make him call another one on the turn. Assume he has the 14-out draw and bet accordingly: 3/4 pot (as detailed above). Bet out $60. See if he calls. If he calls, he makes another mistake by calling, and yet another by pot-committing himself to a draw on the river. Another victory for you.

Finally, let's say you hit gravy, and the turn card gives you either a full house by pairing the board or quads. What now? You have the nuts (assuming again, he doesn't have quads to your full house or JJ for a bigger boat). You need now to bet enough to KEEP HIM IN THE HAND. Maximize profit. He has $107 in front of him, so your absolute betting cap is about half his stack. Bet out more than the $25 you bet on the flop, but less than $50 to keep him from panicking over the size of his remaining stack (which won't be remaining much longer). Let him hang himself by calling. I'd toss out just enough for a flush draw (9 outs) to properly call, about $30. If the turn card was the 8h, do the same thing. LET him call you and feel good about it, then pounce on the river for the rest of his stack.

No Limit Holdem is a complex game, and when you consider than truly good players do what was detailed above in their heads within about 15 seconds, you realize exactly how complex and challenging a game it is.

Off the Schneid...and A Little Bit About Winrates

Well, my breakeven streak is apparently over. As some of you reading my forum saw, I had a monster session at the $50 NL at Full Tilt. I had run up my initial buy-in up to almost $180 in a matter of 40 minutes before I lost a little back with a set-over-set beat and another weird beat. More than that, I have confidence in my game again (not that I was lacking in that regard, but positive results have a reinforcing effect on the mental side), and it's just in time for me to play in tomorrow's Green Chipper--even though PLO isn't my game.

A lot of people have asked me in various forums about winrates, both in Limit Holdem and in No-Limit Holdem. It's in our nature as poker players to be competitive, and many of us actually wonder how we're really doing. Those of us who are winning players--exactly HOW winning are we? We can see the results in our bankroll, but if a player plays 10,000 hands a month and wins $1000, it's vastly different than a player who wins the same $1000 over 4,000 hands.

This is another use of PokerTracker. The software tracks your play and winnings as well, and expresses your winrate (how much you win--standardized) in terms of Big Bets per 100 hands.

Limit Holdem: In Limit, a Big Bet is the amount you would bet on the turn or river. In a 3/6 game, the Big Bet amount is $6. Because of the nature of Limit Holdem, winrates in terms of Big Bets per 100 hands (or per hour) are often lower than their No-Limit counterparts, but since the Big Bet is far larger in most cases, it can still be used to approximate how well a player is doing:

Here are some examples of winrates and what they really mean:

If Joe Holdem has a winrate of 2.2 BB/100 hands at 3/6 LHE, that means that he wins an average of $13.20 for every 100 hands he plays. This winrate is representative of a successful Limit Holdem player. What this doesn't take into account, however, is multi-tabling and table hours--hourly rate. Say that Joe plays 2 tables at a time all the time he plays. If the average online table plays 60 hands in an hour, and Joe plays 2 tables, he plays 120 hands an hour, so at 2.2 BB/100, his hourly win rate (that PokerTracker also calculates) is 2.64 BB/hr or$15.84/hour. For an online small-stakes player, that's extremely respectable. If Joe plays 15 hours per week, he'll make $237.60/week--not bad at all for a hobby, huh? As the number of hands you play go up (over the long run), your winrate becomes more significant. Winrates in the short term (on good runs) aren't significant in terms of classifying you as a player, but over time, it gives a good idea of where you stand among players.

Winrates start to become significant over about 10-20,000 hands. At most online games, if you're playing a single table:

0-1 BB/100: You're a marginally winning player.
1-2 BB/100: You're fairly successful, but can probably use a little work on your game to get over the hump.
2-3 BB/100: You're a successful online player.
3-4 BB/100: You're an outstanding Limit Holdem player, and if you can sustain this rate, you're doing extremely well. You're in the top 5% of poker players.
>4 BB/100: You are CRUSHING the game. You're one of the best players at that limit in the world. Very few players ever touch this level even over as few as 10,000 hands. This, if it were sustainable long term (which, IMHO, it just plainly isn't) would put you among the top 0.1% of all poker players at that limit in the world. You are elite.

If you're playing multiple tables, the focus changes, as your winrate in terms of BB/100 comes down a bit, but your hourly rate (since you're playing vastly more hands) goes up:

0-1 BB/100: You're a marginally winning player.
1-2 BB/100: You're reasonably successful.
>2 BB/100: You're a VERY successful online player. This is where most online multitabling LHE pros live. At higher limits, you can make a VERY nice living being here. An example:

A small-stakes online pro is 3-tabling 10/20, and is winning at 2.1 BB/100. His tables deal about 45 hands per hour--he's playing about 135 hands/hour, and is winning therefore 2.835 BB/hour. 2.835 x $20 = $56.70/hour. Over a 30 hour week (which is a TON, BTW, and not easy--it's a real grind), he's making a hair over $1700 a week, or $88,000 yearly. See what I mean?

>3 BB/100: You're a liar. Nobody beats the online games multi-tabling for this type of winrate long-term. Nobody. :-)

No-Limit Holdem: Since PokerTracker was originally designed with Limit Holdem in mind, it keeps track of big bets in a similar way. In No-Limit Holdem, just as in Limit, a Big Bet is the same as twice the Big Blind. (Note: In a 3/6 LHE game, the SB is $1, and the BB is $3, a Big Bet is $6--in a 1/2 NLHE game, the SB is $1, the BB is $2, and a Big Bet is $4) So, in reporting NLHE winrates, unlike in LHE, you can't just multiply the stakes by the winrate--we record NLHE winrates in terms of PTBB/100 hands, which is slightly different than a corresponding Limit winrate.

Another example: Joe Holdem decides that Limit is too much of a grind and switches to NL. He starts playing 0.50/1 NLHE (known as $100 NL for it's $100 max buy-in). After playing a while, he sees his winrate is 8PTBB/100. What does this mean? PokerTracker sees a Big Bet in this game ($0.50 SB and $1 BB) as $2. So, he's winning $16 for every 100 hands. He's doing quite well. In fact, if he were playing 3/6 LHE, he'd need a winrate of 2.67 BB/100 to match his earnings in this game. He'd need to be a VERY good LHE player to sustain that winrate. In NL, as you will see, 8 PTBB/100 is very sustainable for a good player.

What do these winrates mean? For a single tabler:

0-3 PTBB/100: You are a marginally winning player.
3-6 PTBB/100: You're doing well, but you can stand to plug some leaks to play better.
6-10 PTBB/100: You're playing solid poker, and doing very well at this limit. You're among the top 10% of players in that you're a fairly big consistent winner.
10-15 PTBB/100: You are CRUSHING this game if you're here long term. You're in the top 2% of players if you're here long term.
>15 PTBB/100: Unsustainable long term, you are elite if you're here over the long term. I'm talking top <1% of players at the limit worldwide.

Again, just as in LHE, your winrate means more the more hands you have played to establish and back it up. Winrates at a given limit really aren't significant (and I can't stress this enough--for your own good, DON'T overestimate yourself over a big winrate over 5,000 hands) until you've played over 10,000 hands at a given limit. Some players discount winrates until you're over 25,000 hands at a given limit.

If you're multitabling, you can still pretty much maintain the same numbers as above, but more toward lower ends. For example, if you're winning at 10 PTBB/100 while multitabling, you're crushing a certain limit. Winrates at >15PTBB/100 for consistent multitablers just doesn't happen long term. You're very successful multitabling if you win 6-8 PTBB/100, which can be done. Think about this example:

Back to our friend Joe Holdem. He's playing 0.50/1 NL ($100 NL), he's 4-tabling and winning 6.5 PTBB/100. His tables average about 50 hands per hour. What is his hourly rate?

6.5 PTBB/100 sounds like a very modest winrate. Let's look closer. He's 4-tabling, and this means he's doing very well for himself. How well? 6.5 PTBB/100 = $13/100 hands. He's playing 200 hands/hour, so he's making $26/hour. Again, not bad for a hobby. Compare this to a player who is crushing the same game but only single tabling. He wins 13 PTBB/100--double Joe's winrate. At 50 hands/hr, he's making 6.5 PTBB/hr, or $13/hour. The multitabler, despite half the winrate is making twice the money.

Multitabling, if you can do it, can be VERY profitable. Remember, the best way to read your opponents, especially when multitabling, is using PokerTracker and PokerAceHUD.

Best of luck at the tables, all!

Frustration or Introspection??

It's very frustrating for a player like myself to not see the results I'm used to seeing. I'm right now in the midst (hopefully at the tail end) of about a 5k hand break-even streak. Most of my losses have come at my "home site" of PokerStars. Sometimes it's just my being impatient, sometimes it's just a matter of me being wrong on a read or running into a cooler (99 vs. TT, etc.).

All I know is that it needs to stop. I had a fairly lame month last month, only netting about $800 or so, and even though much of it is due to my playing less, some of it is due to bad play.

That's why I'm writing this post.

When you get into that bad groove, you need to get yourself out of it. Pronto. No way is better at getting you out of the doldrums than getting back to basics. Think about hands you're playing--review your PokerTracker info--look at the hands that are costing you money. Look at the hands that you lost the most money on. Were they due to just poor play? Was it an unfortunately unlucky break? For me, most can be attributed to poor play or a blown read.

What I do is go session-by-session, and I find those sessions (NL) where I lost a stack or a good chunk of one, then go hand-by-hand to see where the money's disappearing to. I look at the opponent, their PT numbers, then break down the hand in the hand replayer. I look to see if someone sucked out on me, was it because I gave them adequate odds to call? Or was it because I tried to slowplay and got it stuffed down my throat? I look at hands I played out of position, in position, and just look at each mistake I made--and how to correct it. One of those introspective sessions is overdue, IMHO.

It's amazing how much either success and failure can change your game. What separates the truly great players from the merely good is how you react to either success or failure and how you adjust to make your game the best it can be.

Utilizing PokerAceHUD to the Fullest

I would hope that many of you have taken advantage of VPP's many promotions to get a license for PokerAceHUD, what I believe is the top add-on for PokerTracker.

This program allows you to display statistics on your opponents, as an overlay on the poker table, in REAL TIME. I'm sure you can see how this could be an advantage. As you play at a table, PokerTracker analyzes the hand histories you generate and not only tracks YOUR play, it also provides the same statistics on your opponents. With this information, you can...

--Help put an opponent on a range (VP$IP)
--See how solid a hand he/she will raise with (PFR)
--Determine how aggressive an opponent is (AF)
--See how reliable these stats actually are (number of hands played)

I have my own system with PAHud:

***WARNING: This presupposes that you have read the tutorials on http://www.pokeracesoftware.com, so if you haven't read them, make sure you head over there, as Josh has really put together some nice info.

I use the "Combined Stats" option, and put in to this line the primary stats you look at to get an "at-a-glance" read on a player:

VP$IP/PFR/AF-T/# Hands

I also use the color range option to sort these players. In recent updates, Josh has added the ability to change the color of each stat within a combined stat line to match pre-determined ranges. I use pretty basic color schemes--first selecting VP$IP (Voluntarily Put Money into the Pot) and setting say <20%>55% as green. Then, I'll do the same for PFR--<5%>15% to be red. These are oversimplified and not my actual ranges, but I'm sure you get the point. You want to be able to look at the line above the player's name and get a quick and dirty read on how they play, and this is a fairly easy method. The same goes for aggression factor. Your goal here is to essentially render the auto-rate rules I wrote to be obsolete. You want to look at a player's line, and see a red VP$IP, a red PFR, and a red AF-T and say, "He's tight-aggressive/aggressive," and know that if he's raising he's raising with a solid hand.

PokerAceHUD also allows you to use "Pop-Up" stats, which are stats that you can access by clicking on the opponent's name at the table. The stats you select will then pop up, stay for a few seconds, then disappear until you click on them again. When playing NL, I'll put these type stats into the "Pop-Up:"

--Call PFR--if I raise, and he flat calls, how likely is it that he's trapping me?
--Aggression Frequency--a PAHud-exclusive stat, it gives more insight to Aggression than the old aggression frequency does. I select it by street (flop, turn, and river) and use to to see how likely a player is to bet when checked to, and if that bet means you're beat or if he's just firing to fire a bullet.
--Went to Showdown--how far will my opponent take a marginal hand? Will he play his TPGK hand all the way even though I'm betting/raising him to death? The higher this number, the more likely it is...
--Won $ at Showdown--is he taking only monster hands to showdown, or is he value betting with marginal hands to chase you away?
--Won $ When Raised Turn/River--Don't you just love when you're leading out, leading out, and some donkey decides to raise on the turn, usually a min-raise? This stat (over 100s of hands) lets you know how likely he is to have a solid hand. If this number is low, he's a bluffer and is trying to buy a pot (more often).

My next article will have some screenshots on how I use PAHud, so stew on this for now, and continued luck at the tables!