Wizard strategy for Ultimate Texas Hold 'Em
In preparing for Ultimate Texas Hold 'Em I wanted to present a simple strategy for the game. Here is what I came up with.
4X Raise
Follow the optimal strategy below.
2X Raise
Raise with any hidden pair or better
or
Raise with 4-flush including suited 10 or better or pocked twos
1X Raise
Raise with a hidden pair or better.
or
Raise if less than 21 outs beat you.
Here is an example of the "21 outs" rule.
The this example the following number of each rank will be beat you in the dealer's hole cards:
Aces: 4
Kings: 3
Queens: 4
Jacks: 3
8's: 3
6's: 3
5's: 3
Total = 23
Because 21 or more outs beat you -- fold.
I must admit that the "outs" idea comes from the James Grosjean strategy card. So does the second rule under 2x raise. However, I feel there is no way to address the game without using these common ways to express poker strategy.
I asked JB to analyze this new simplified strategy. Here are his results:
TRIALS ................. 12,001,119,847
RETURN ................. -318,576,326
RETURN % ............... -0.026546
TOTAL BET .............. 50,163,071,530
RETURN OF AVG BET ...... -0.006351
As a reminder, the house edge under optimal strategy is 2.185%. So, this is only 0.47% higher. Yeah, 0.47% is a lot, but keep in mind that is relative to just the initial bet, and the game also includes the blind and raise bets. The element of risk is 0.526% under optimal strategy and 0.635% under this Wizard strategy, which is only 0.11% higher. That is about one full bet every 900 hands.
So, you guys are the first to see the "Wizard simple strategy" for Ultimate Texas Hold 'Em. What do you think?
I really enjoy this game and have noticed that most people play it very poorly and even criticize my play (Grosjean strategy.) Your strategy is simple enough that more people might bother to learn it. Then again the average player just can't come to grips with raising 4x on any ace, so maybe not.
Quote: jml24We should probably be happy people check pre-flop with good hands. If everyone played correctly the casinos would probably start reducing the pay table on the blind bet.
I agree with that principle regarding every game. That is why I say that I'm not trying to save everybody, only those who will listen. It takes some bad gamblers to subsidize the good ones.
In other news, JB Emailed me some improvements over the strategy above. I'm still going over that with him. So, stay tuned for an even better one.
Before it said, "Raise with 4-flush including suited 10 or better or pocked twos."
I'm going to reword that to "Raise with 4-flush including a hidden 10 or better to the flush in the hole." This makes it clear the player must have a good kicker, in case the dealer gets a flush too.
The pocket twos was evidently not very helpful to add and doesn't happen very often, so I dropped that.
JB suggested I add a rule under 1x saying to always raise with a straight or higher on the board. However, I feel that case is covered under the 21 strikes rule. If there is a pat straight or better on the board there will be fewer than 21 outs that improve it.
Here are the results of this modified strategy. Previous numbers are in parenthesis.
TRIALS ................. 4,301,049,313
RETURN ................. -104,650,743
RETURN % ............... -0.024331 (-0.026546)
TOTAL BET .............. 17,914,312,086
RETURN OF AVG BET ...... -0.005842 (-0.006351)
What do you think?
Quote: jc2286Isn't pocket twos already included by the first 2x rule anyway?
The first rule says to check pocket 2's.
Quote: TranscendFor the 21 outs rule, does that mean each card has a max possible value of 4 based on how many are left to show up then you add up the total unseen of cards that can beat you?
Yes. So if there are five different ranks on the board, and I assume none match yours, lest you would have a hidden pair and raise, then there are 15 strikes already against you (5 ranks * 3 suits left in each). Then just add 4 for each other rank that beats your highest hole card. If there is a 4-straight or 4-flush on the board, add strikes for that too.
Quote: WizardThe first rule says to check pocket 2's.
Isn't pocket twos a hidden pair?
2X Raise
Raise with any hidden pair or better
Quote: WizardThe first rule says to check pocket 2's.
Yes. So if there are five different ranks on the board, and I assume none match yours, lest you would have a hidden pair and raise, then there are 15 strikes already against you (5 ranks * 3 suits left in each). Then just add 4 for each other rank that beats your highest hole card. If there is a 4-straight or 4-flush on the board, add strikes for that too.
Thank you for the clarification as well as JB and yourself doing the legwork
Quote: WizardI may as well say that tomorrow I'm going to be making new gambling videos for pai gow (tiles),
Can you have Juicie Jennie star in the tiles video?
Quote: jc2286Isn't pocket twos a hidden pair?
Good point. Here is the latest wording of the strategy to make clear to not raise 2x with pocket deuces unless they contribute to a two pair or better:
4X Raise
Follow the optimal strategy below in the original post.
2X Raise
Raise with:
A two pair or better
or
A hidden pair except pocket twos.
or
Raise with 4-flush including a hidden suited 10 or better.
1X Raise
Raise with:
a hidden pair or better.
or
if less than 21 outs beat you.
Any questions or suggestions?
Quote: sodawaterCan you have Juicie Jennie star in the tiles video?
I would be happy to if she were here. If anyone in Vegas tomorrow would like a cameo, let me know.
The nitpicker in me feels compelled to point out that the proper wording is "fewer than 21 outs beat you." The poker wording might actually be "the dealer has fewer than 21 outs" while "fewer than 21 cards beat you" would avoid poker jargon.
1) You recommended initially (then retracted) betting pocket deuces if the flop gives you a flush draw? (Ex: 2c2s on a board of As10s5s) Is that correct?
2) You recommend checking when you've paired a two on the flop? (Ex: 10s2d on a board of 9s5h2c)
3) You recommend betting pocket deuces if there is another pair on the flop, but not if there isn't?
If I'm understanding all that correctly, I'd like to know the reasoning behind #3 in particular. If there is a pair on the board, and I have pocket deuces, I'm scared to death of a second pair coming on the turn or river, counterfeiting my deuces. Are you really in better shape with a paired board than a non-paired board?
On a side note, UTH is one of my favorite table games, and I pride myself on playing as close to Wizard Strategy as possible. I've mastered the preflop action, and now I want to polish the rest of my game, and your analyses of table games is the gold standard, IMO. Thank you for all the hard work you do.
Quote: DeucekiesTo clarify regarding pocket deuces on the flop. Am I correct on the following:
1) You recommended initially (then retracted) betting pocket deuces if the flop gives you a flush draw? (Ex: 2c2s on a board of As10s5s) Is that correct?
Yes.
Quote:2) You recommend checking when you've paired a two on the flop? (Ex: 10s2d on a board of 9s5h2c)
No. You always raise if you pair the flop. If I wrote that before there was a misunderstanding.
Quote:3) You recommend betting pocket deuces if there is another pair on the flop, but not if there isn't?
Always raise with two pair or better.
Quote:If I'm understanding all that correctly, I'd like to know the reasoning behind #3 in particular. If there is a pair on the board, and I have pocket deuces, I'm scared to death of a second pair coming on the turn or river, counterfeiting my deuces. Are you really in better shape with a paired board than a non-paired board?
Yes. That is what the math says. I hate to give sound bite answers, but you've already got a two pair and have hope for a full house.
Quote:On a side note, UTH is one of my favorite table games, and I pride myself on playing as close to Wizard Strategy as possible. I've mastered the preflop action, and now I want to polish the rest of my game, and your analyses of table games is the gold standard, IMO. Thank you for all the hard work you do.
Since you take the game seriously, I would suggest getting the Grosjean card or follow the strategy at Discount Gambling if you can understand it. The strategy on my site is too complicated. I probably will remove it when I post this simple strategy.
Quote: DeucekiesTo clarify regarding pocket deuces on the flop. Am I correct on the following:
1) You recommended initially (then retracted) betting pocket deuces if the flop gives you a flush draw? (Ex: 2c2s on a board of As10s5s) Is that correct?
2) You recommend checking when you've paired a two on the flop? (Ex: 10s2d on a board of 9s5h2c)
3) You recommend betting pocket deuces if there is another pair on the flop, but not if there isn't?
If I'm understanding all that correctly, I'd like to know the reasoning behind #3 in particular. If there is a pair on the board, and I have pocket deuces, I'm scared to death of a second pair coming on the turn or river, counterfeiting my deuces. Are you really in better shape with a paired board than a non-paired board?
On a side note, UTH is one of my favorite table games, and I pride myself on playing as close to Wizard Strategy as possible. I've mastered the preflop action, and now I want to polish the rest of my game, and your analyses of table games is the gold standard, IMO. Thank you for all the hard work you do.
Wiz,
Thank you (and JB) for this! I think it will help a lot of people. I've been playing JG's strategy, and while I'm not shy about putting up 4x when called for in general, can we get down in the weeds a bit on queens? They have been biting me a lot ever since I started playing the game, and I modified my BS to the following, when the queen is only playing as a kicker at the end and the other card does not improve the hand (changes to your strategy otherwise);
4x check Q-10o, Q-8s or less
2x " " (no 4 to a flush or 4 to an outside straight) (see below)
1x bet if board holds at least a lower pair plus an overcard, or at least 2 overcards (which can include AA or KK pairs); if this happens on the flop, bet the 2x then. Otherwise fold.
I have been winning more consistently (growing my session pile from the play) since making this change, but I do miss some wins where the queen would have survived as the kicker playing your more aggressive strategy. I'm curious how much if any edge I'm giving away by doing this, or whether it actually is a good strategy, possibly with some threshold tweaks. I may just be on a run of good variance, and was on a long run of bad queen variance before that led me to this; if it's a minor difference, I may continue doing it.
EDIT: I think this question mostly falls under your "21 or under" rule, but it's likely I'm folding more often than applying that rule strictly would dictate, just for clarification.
Quote: Wizard
No. You always raise if you pair the flop. If I wrote that before there was a misunderstanding.
Looks like I just misread your post. Sounds good.
Quote: Wizard
Yes. That is what the math says. I hate to give sound bite answers, but you've already got a two pair and have hope for a full house.
This must be the rub. The reward of the full house outweighs the risk of the counterfeit.
Quote: WizardSince you take the game seriously, I would suggest getting the Grosjean card or follow the strategy at Discount Gambling if you can understand it.
I will certainly look in to that. Thanks again, and thanks for the quick responses.
Quote: Deucekies
This must be the rub. The reward of the full house outweighs the risk of the counterfeit.
Actually i think the "rub" is more because when the board pairs, your ante gets paid when you win with pocket 2s instead of pushing.
And great work wiz!! Thank god poppies will hate this strategy though...lol
Quote: WizardSo, you guys are the first to see the "Wizard simple strategy" for Ultimate Texas Hold 'Em. What do you think?
Doesn't this get outperformed by Grosjean's simple strategy, which you can get on the little card that fits in your wallet (I don't play UTH any more, so I can't find the card. Back before the casinos wised up and stopped over-comping that game, it was great, though. That little strategy card was the best $4 I ever spent)
IIRC I bought the card from a link on your page....
Quote: AxiomOfChoiceDoesn't this get outperformed by Grosjean's simple strategy, which you can get on the little card that fits in your wallet (I don't play UTH any more, so I can't find the card. Back before the casinos wised up and stopped over-comping that game, it was great, though. That little strategy card was the best $4 I ever spent)
IIRC I bought the card from a link on your page....
There is no doubt the Grosjean strategy is stronger. However, mine is easier to memorize. How much work one is willing to put into the game is up to them. I'm just looking to present something to the lazier gamblers.
Quote: Wizard
2X Raise
Raise with:
A two pair or better
or
A hidden pair except pocket twos.
or
Raise with 4-flush including a hidden suited 10 or better.
Nice work as always!
To get rid of a line I would recommend combining the fist two rules to "any made pair or better", unless "made pair" is not understood outside the poker community, then you might go with "any hidden pair or better (except pocket deuces)".
(also, please refrain from writing "twos" in the future, you are the Wizard forgodsakes ;-))
You didn't include the open-ended straight holding 9+ or better from the 2x. It might not be a big difference, but it is very easy to remember, especially for poker players, which is why I think it is worth including. Can you provide any numbers with the straight draw included for comparison? Thanks!
Quote: CanyoneroYou didn't include the open-ended straight holding 9+ or better from the 2x. It might not be a big difference, but it is very easy to remember, especially for poker players, which is why I think it is worth including. Can you provide any numbers with the straight draw included for comparison? Thanks!
The problem is, as I understand the optimal strategy, the open ended straight is very sensitive to whether you're using 1 hole card or 2, whether there's 3 to a suit or 2 to a suit or a rainbow board, and how high the straight is. If there was a simple rule, sure. I think it's no big loss because most of the starting hands that give you a straight draw you can correctly 2x on the flop, you're 4Xing with pre flop.
Quote: CRMousseauThe problem is, as I understand the optimal strategy, the open ended straight is very sensitive to whether you're using 1 hole card or 2, whether there's 3 to a suit or 2 to a suit or a rainbow board, and how high the straight is. If there was a simple rule, sure. I think it's no big loss because most of the starting hands that give you a straight draw you can correctly 2x on the flop, you're 4Xing with pre flop.
You are right, the wording should be:
open-ended straight draw holding two cards 9 or better
board texture doesn't matter here, expectation of raise is (almost) always better than check
Quote: CanyoneroYou are right, the wording should be:
open-ended straight draw holding two cards 9 or better
board texture doesn't matter here, expectation of raise is (almost) always better than check
Pretty sure any straight draw on a one-suit board is a check, and that you still check 98 on a JTx board. But those are really starting to get into fine points almost not worth bothering for a 'simple' strategy. I, too, am curious of the cost of not betting straight draws.
I think the main thing at the end is how scary the board is. but a simple method might be.
Over-riding rule - if you beat the board by more than kickers or the board has a 5-card hand, then play.
(i) VWXYZ No Pair - Really scary board, need to equal or beat bottom pair, not scary at all need 2nd Nuts.
(ii) WW XYZ Pair - 1st thru 3rd nuts depending on scariness.
(iii) 2 Pair or Trips - 4th/5th nuts acceptable, also play the board sometimes.
(Note this means that you sometimes would fold pocket 2s - either with a scary non-pair board or 2 Pairs as you are now playing the board).
Similarly for x2.
Over-riding rule - unless the board is suited, if you beat the board by more than kickers, then raise x2.
(i) XYZ No Pair - any flush draw, open ended straight, Nut kicker any possible straight, any pair (i.e. anything that's promising and better than "just nut kicker").
(ii) XYZ Suited - Any good pair, any pair & flush draw, good flush draw (i.e. has to be better quality than above) .
(iii) XXY or XXX - 1st (or 2nd) Nut kicker, good flush draw.
Quote: CRMousseauPretty sure any straight draw on a one-suit board is a check, and that you still check 98 on a JTx board. But those are really starting to get into fine points almost not worth bothering for a 'simple' strategy. I, too, am curious of the cost of not betting straight draws.
Yeah, I was way off with "almost always". Sorry about the misinformation, it has been a while since I delved deeply into UTH strategy. In fact, board texture come into play heavily, and thus it gets really complicated.
Quote: CanyoneroTo get rid of a line I would recommend combining the fist two rules to "any made pair or better", unless "made pair" is not understood outside the poker community, then you might go with "any hidden pair or better (except pocket deuces)".
I'd like to combine lines wherever I can. My original post said, "Raise with any hidden pair or better." I don't mind adding the pocket deuces exception.
One hand that changes is a three of a kind on the board. Grosjean says to only make the 2x raise on that with a king kicker. It probably doesn't matter much one way or the other how to handle a trips on the board, since it happens so seldom (1 in 425 hands), and the strategy comes down to the kicker.
Another issue is "any hidden pair or better" ins't totally clear if it means the hand better than a pair must be hidden. In other words, what takes precedent, the "any" or the "or."
So, I'm still debating with myself which way to go. Brevity or clarity?
Quote: CanyoneroYeah, I was way off with "almost always". Sorry about the misinformation, it has been a while since I delved deeply into UTH strategy. In fact, board texture come into play heavily, and thus it gets really complicated.
No worries, believe me, I apologize if I sound like a hound dog on this. And the far more important board texture is "paired / unpaired", and the straight strategy still holds I believe (i.e. T9 is correct on a J88 or JJ8 or 887 board).
Quote: WizardI'd like to combine lines wherever I can. My original post said, "Raise with any hidden pair or better." I don't mind adding the pocket deuces exception.
One hand that changes is a three of a kind on the board. Grosjean says to only make the 2x raise on that with a king kicker. It probably doesn't matter much one way or the other how to handle a trips on the board, since it happens so seldom (1 in 425 hands), and the strategy comes down to the kicker.
Another issue is "any hidden pair or better" ins't totally clear if it means the hand better than a pair must be hidden. In other words, what takes precedent, the "any" or the "or."
So, I'm still debating with myself which way to go. Brevity or clarity?
Didn't catch the trips thing - the "two pair or better" wording included that one all the same. It might be a concern though, since almost all king kicker hands were raised 4x already - that means, if the trips do come on the board, the 2x raise is almost always wrong. I guess it all comes down to how much the numbers change.
[btw after some more testing, forget my "straight" suggestion, it only applies to J9o and T9 anyway...]
I read it through several times, and interpreting "hidden pair" as both cards being hidden, was utterly baffled why you'd recommend not raising when you flop top pair. The explanation of "outs" would have cleared it up - or at least made me realize something was wrong - but I skimmed past it, since I'm quite familiar with how outs work! While "pocket pair" is a more common term in holdem, I've heard "hidden pair" to describe a pair both face-down in stud games. "Board pair" (holdem) or "open pair" (stud) for both face-up, "split pair" (usually stud) for one up and one down. I don't think I've heard "hidden pair" used to describe a pair with one face-down card in any poker context before.
I would like to suggest a FOLD option...
FOLD if you have a Bottom pair 4-4 or less (include pocket 2-2) AND there is an open-ended 4-straight OR 4-Flush on the board.
EXAMPLE 1: 4-10 pocket with board 4-5-6-7-Q (This totals 20 strikes, plus Dealer pocket pairs 33 or better)
EXAMPLE 2: 4-10 off suit pocket with Suited board of 4-7-8-Q plus off suit K (This is 21 strikes including the flush draw)
The rule may not be obvious to most Players, and re-emphasises that an Open-ended straight adds 8 strikes, and the 4-flush adds 9 strikes.
A clarification on the 2x Flush rule: I tend to raise with a Suited Jack or better... if the board has a Suited Jack or better I raise with a Suited 10. ONE CARD or TWO CARDS in pocket. (Therefore Suited J + off-suit 9 or less, or Suited J-2 thru J-7. If the Jack or better is on board, then Any Suited 10 + off suit, or Suited 10-2 thru 10-9).
I personally have always found Stephen How's concept of using kickers to be easier than counting outs, but his presentation is a little confusing.
For example, if the board isn't paired you need the 2nd best kicker to play (Q in your example).
I'm sure the outs counting gets quicker as you get used to it though.
What has been updated lately.... you're not one of those people thaat blindly upgrade IE8/9, AV/No-Script without knowing consequences are you? If it ain't broke don't upgrade.
I would think its an IE problem, but you say all browsers are affected. So it must be a security app or SW that got updated.
Seriously, though, this sounds like a Java-Script setting, especially the security level. Adjust at your own risk.
FWIW I'm running Linux with Firefox, and no problems. Included apps are AdBlock, Better privacy, and Redirect Cleaner.
Okay, it worked now after I changed my Java security settings.Quote: 98ClubsToss windoze. Sorry, couldn't resist.
What has been updated lately.... you're not one of those people thaat blindly upgrade IE8/9, AV/No-Script without knowing consequences are you? If it ain't broke don't upgrade.
I would think its an IE problem, but you say all browsers are affected. So it must be a security app or SW that got updated.
Seriously, though, this sounds like a Java-Script setting, especially the security level. Adjust at your own risk.
FWIW I'm running Linux with Firefox, and no problems. Included apps are AdBlock, Better privacy, and Redirect Cleaner.
I think it changed when I blindly updated Java...
This is a curious difference. Statistically it should not make any difference, but it does go against the official rules literature put out by Shufflemaster.
Quote: Okienst111Hello I have a question my buddy's and I had a boys weekend to Vegas on Super Bowl weekend. We all enjoy playing the ultimate Texas hold'em game well we were staying at Paris so we played there and I noticed that they deal the 5 community cards first then there 2 cards to the dealer and so on to the players. My question is this the new way the game is dealt and does it change the odds being dealt this way?
Not a game protection specialist, but from a player standpoint, I like that they're doing it as you describe, because people coming in and out of the game will not affect the cards on the board or in the dealer's hand. (It still will effect what you're dealt unless you're sitting first base; it might even have been in response to heavy play for first base that they do this there.) It also would make it slightly less of an issue if players touch their cards before the full deal is complete, because if they're going to foul the hand through exposing or miscounting either the dealer's or the board's cards, they will have done it before they have given out cards that now have to be forfeited back.
Quote:For each rank on the board there are three more that can beat you. For example, if the dealer has the jack of hearts, diamonds, or spades he will pair up and beat you. So, 5×3=15 cards will give the dealer a hidden pair. Also, all four queens and aces will beat you. So, 15+8=23 cards will beat you. If the dealer has one of the three remaining tens, then you'll lose lose. That brings us to 26. You need to have less than 21 outs to stay in, so fold.
This states that if the dealer has a Ten, you "lose lose". First, I think you have an extra "lose" in there.
But that isn't necessarily a true statement. Sometimes you'll push. Also, in the original post, you use that hand as an example and report 23 outs.
What is the correct technique? I've always just counted single card outs, ignoring kickers.
Quote: TriathlonToddJust a nitpick thing, Wiz, but you never specifically address pairs in your 4X section. When people read the rest of the rest of the strategy, they might figure out to raise with pocket pairs of 3 or higher, but they also might not come to that conclusion.
It is there in the strategy, just text below the colorful chart. Easy to miss.
https://wizardofodds.com/games/ultimate-texas-hold-em/
Parse error: syntax error, unexpected T_STRING in /home/wizardofodds/wizardofodds.com/games/ultimate-texas-hold-em/index.php on line 8
But in the past month, my sessions have been -$400, -$350, -$200, -$200, and -$130. I can honestly say that in all of that play, I didn't even have one upswing of more than $75. I've heard people say that they like UTH because it's less volatile than blackjack. They're right. Instead of going up and down like in blackjack, you just go straight down. There you go, less volatile. It seems to be the common theme at the table. It doesn't seem that anybody is ever coming out ahead. Sure they don't play correctly but it seems to save them money cause that ace doesn't do you any good when the dealer always seems to have a 5 to connect a straight. Last night I won 4 out of 10 4X raises, the time before, probably only 4 out of 12. Everytime you lose, you lose $30, everytime you win, you only win $20-$25. I feel the blind bet is the real killer. It's just at such a disadvantage.
So I did some testing. I took a single deck and shuffled and dealt them just like they do at the casino. Even with perfect hole card info, I can't get the game to turn positive. After 600 test hands, I wound up down about $200. The high was $200 and the low was -$500. I realize this isn't a great sample, but it gives you a pretty decent example. Any comments would be appreciated. Thanks
I don't play without the Trips bet. Which may cause you to disregard anything else I say, but that's where I make my money, along with a very aggressive betting strategy that freaks most people out, but IMO is the only way to win at this game. If you play this strategy, it gets the Element of Risk down around .5. The Las Vegas Advisor sells the strategy card for $5. I highly recommend it.
I'm too tall to get away with hole-carding, so I don't even try. But adding hole carding to the aggressive strategy would give the player a big advantage, if the info is used correctly. I think the web site apheat has a pretty good discussion on just how much it adds, and I would suggest you check there. Be sure to read the blog entry in the link for UTH in that alphabetical listing.
UTH strategy is also discussed in great detail at WizardofOdds.com , a sister site to this one. Look especially hard at the advice on how to play kickers.
Also be sure you're looking for opportunities to play the board for a push, because it can be hard to see, but boards like 2 pair + a 10, or 3OAK plus 2 face cards, are times to bet with nothing in your hand. This will become apparent if you start counting your kickers per WoO advice, but most players miss these and fold.
Based on your description of your play, I would guess you're folding too early on a few hands, combined with not being aggressive enough on your 4x and 2x bets. But that's totally my SWAG, because that's what I see a lot; this game is different from nearly all the rest in that you have to bet exactly as the strategy card says, with the refinement of playing the kickers correctly, and let the math do the work for you. No psyche bets, no trend bets, no backing off your aces when the dealer's filling gutshot straights.
Good luck.
