Another big draw, how cautious should I be when I hit?

[JTG51]

Big draw hands seem to be in vogue here lately, here's mine. It's from a typical loose passive 5/10 Foxwoods game. The aggressor in this hand (LP) just sat down at the table. He's a little loose, and a litlte aggressive, but no maniac.

UTG limps, I limp UTG+1 with T /forums/images/icons/diamond.gif 9 /forums/images/icons/diamond.gif . MP limps, LP raises. Button calls, BB calls. 6 ways for 2 bets.

Flop: J /forums/images/icons/diamond.gif 8 /forums/images/icons/spade.gif 3 /forums/images/icons/diamond.gif

I like it. BB checks and UTG bets. I call, MP calls, LP raises, Button calls 2 cold, BB folds, UTG calls. I 3-bet, MP calls and LP caps it. Button calls 2 cold again, everyone else calls. 5 ways for 16 BBs.

Turn: [J /forums/images/icons/diamond.gif 8 /forums/images/icons/spade.gif 3 /forums/images/icons/diamond.gif ] 3 /forums/images/icons/club.gif

It's checked to LP who bets, everyone calls. 5 ways for 21 BBs.

River: [J /forums/images/icons/diamond.gif 8 /forums/images/icons/spade.gif 3 /forums/images/icons/diamond.gif 3 /forums/images/icons/club.gif ] 2 /forums/images/icons/diamond.gif

What's my best action?

[slavic]

Was it Lee Jones who wrote "Bet the flop"? Umm I hate to give different advice but I think you check, call. I'm not sure it was the flush you wanted to make, and a reraise means your dead in the water.

[JTG51]

Was it Lee Jones who wrote "Bet the flop"? Umm I hate to give different advice but I think you check, call.

I don't really understand what you are saying here. I should have bet the flop? It was already bet when it got to me, do you think I should have raised? Is the check and call advice for the river?

Thanks.

[slavic]

sorry bet the river, too much coffee on my part.

ok so what I said is check call the river.

As far as your flop play, I probably would have raised, after the first bet, but you played it fine and got a lot more information than I would have from my raise.

[Bob T.]

I think that I am going to bet, and call one raise, but if there are more raises than that, I will fold. If my flush is good, I think that there is too much danger in it being checked through, and if my hand isn't good, I think that there is too much danger in going for a checkraise, so in either case, I think that the better play will be to bet out on the river.

[Bob T.]

As far as your flop play, I probably would have raised...

Because the bettor was on his immediate right, I wouldn't have raised here, even with this big draw, because the raise would tend to shut out players, and with your draw, you want as many player in as possible, to give you better odds on your draw. If I had the same hand after several callers on the flop, this would be an easy raise.

[Ulysses]

What's my best action?

Bet and call a raise.

[rkiray]

I hate to say this, but I would check and fold on the river. I almost never do that in a pot this big so check/calling is ok for a single bet. But with this action and this many players someone must be on a bigger flush draw. I think the best card you could have gotten was an off-suit 7.

[Homer]

Wouldn't you fold on the turn then, since you plan to check-fold no matter what card comes on the river?

-- Homer

[rkiray]

If I make my striaght I'm betting it, particularly if the 7 off-suit hits.

[rkiray]

I decided my last answer needed explanation. This also got me thinking and looking in my database where I found something odd.

In general in a hand like this I fear the flush (especially a large one) much more than a straight. I've noticed over time that I generally see more flushes than straights on hold'em tables. I've discussed this with my brother a couple of times. He a very bright guy who enjoys poker (he teaches at the IU medical school). He has noticed this also. Our theory was that people just play more suited cards than cards that make straights. Some people will play any Axs or Kxs. Some maniacs seem to play any two suited cards. Even most maniacs don't play hands like 36o. So that why we both thought we observed more flushes.

I decided to dive into my database to see if I could confirm this. What I found is really strange. I can't see to find stats on exactly what hands I ended up with that I actually played but there are numbers on cards dealt. Here are the stats I find interesting.

There have been 20431 hands where I was dealt cards that went all the way to the river. of those i got <ul type="square">
333 flushes (6.80%) expected 318 (6.51%)
261 straights (5.33%) expected 223 (4.56%)
[/list]

This appears very odd to me. Since flushes beat straights should flushes occur less. Why does the stats program expect more? Note, these stats arn't hands that I played all the way to the river. I'm sure I folded lots of 26s preflop and missed straights and flushes.

Investigating a little more I found these stats. In 34572 hands dealt to me I received :
<ul type="square">
suited starter 8061 (23.32% ) expected 8135 (23.5%)
connectors 5338 (15.44%) expected 5423 (15.7)
suited connectors 1329 ( 3.84) expected (3.92)
[/list]


If these numbers are true the first set makes more sense. Apparantly suited cards are more common than connectors. You combine this with my belief that people generally play a higher % of their suited cards than their connectors that would explain why I see so many more flushes. If all these numbers are true, why doesn't a straight beat a flush (that's another reason why I think people play more suited cards, because they know the flush is the bigger draw). Am I crazy here? Is my database goofy? Is there something odd about the structure of hold'em? Any insight greatly appreciated.

[SoBeDude]

So you're calling a turn-raise to hit a 3 outer?

I'm not sure thats a +EV play.

-Scott

[SoBeDude]

Let me try to help.

First, you're making a very easy mistake to make.

You're going on the assumption that it takes 2 flush cards to make a flush AND it takes connectors to make a straight.

Of course you DO need two suiteds to make the flush, but you can make a straight with any one card in your hand (ignoring a board flush or a board straight)

So you're accidently comparing apples to oranges here. If you simply look at all starting cards played to the river, you'd see more straights than flushes.

So, put another way, if you looked at all starting cards, you'd make more straights.

So looked at from another angle, Given decent starting hand requirements, you're going to make more flushes than straights. (Remember 9 cards fill a flush and only 8 make a straight.

Just a curious thought. I wonder if Fish make more straights than good players...? I'd bet they do.

-Scott

[JTG51]

So you're calling a turn-raise to hit a 3 outer?

I'm not sure thats a +EV play.

Sure it is. There was no raise, just a bet. I was getting 19-1 to call on the turn, which is more than enough for a 3 outer. If I saw everyone's cards and knew that my only outs were the 3 non-diamond 7's, I'd have an easy call. The only way I can have exactly 3 outs though is if someone has QQ and someone has a bigger flush draw. It's far more likely that I have 15 outs, 6 outs, or 0 outs. Which one of those three is most likely is certainly debatable.

[JTG51]

My thoughts exactly, Bob.

[Homer]

Sorry, I blanked out and forgot about the straight draw.

-- Homer

[JTG51]

I'm not wild about my chances of winning with the flush, but I think checking and folding for one bet would be a very big mistake.

[rkiray]

Actuall I believe you are making a couple of mistakes here. But I admit I find this subject very confusing and may be wrong. You state :
Of course you DO need two suiteds to make the flush, but you can make a straight with any one card in your hand (ignoring a board flush or a board straight)

This seems incorrect to me. You can make both straights and flushes with one of your hand cards.

Second of all if I understand the pokerstats. It was looking at every hand I was dealt that the river was also dealt (whether I stayed in or not). It showed more flushes did occur over 20,000+ hands and even odder it expected about 50% more flushes than straights. This is what confused me. Are there any other pokerstat uses who can check their databases. The screen is under the tools menu, hands dealt.

[JTG51]

I think those of you that said bet and call a raise on the river were probably right.

I didn't do that though. I thought my chances of being in third place were as good as my chances of winning, so I checked planning on calling 1 bet. LP also checked, and the button bet. I called and LP called.

Button showed A /forums/images/icons/diamond.gif x /forums/images/icons/diamond.gif for the winner. LP had black aces. The other two players mentioned that they each had a J.

Thanks for the comments.

[rkiray]

I actually thought you had 6 outs (both ends of the non-flush straight, I would just be happier with the low end). I would still call if I thought it was only 3.

[Joe Tall]

River: I be afraid of everyone freezing and checking this through, I'd bet out and call 1-more bet on the rebound. But you've seen my posts where I bet out for info often, so I've been revaluating this play and I'm curious to know your results here. But I'd be cautious if it was 3-bet here if I bet out, however, and if I was confindent with it, I'd lay it down.

Flop: I like you call/3-bet on the flop, it's a great way to Marry everyone to the pot and jam-up you draw without shutting everyone out!

Pre-flop: I'm not sure if I'd limp here, call me a tight-a$$ but I seem to lay this one down here in UTG+1. I guess if you are confident that it is a 'limping is contagious' type of game the limp is fine.

Peace,
JT

[Zag]

I wonder if this anomaly occurs because a flush beats a straight, and would be even worse if we declared them to be the other way around. That is, all the hands that contain both a straight and a flush (but not a straight-flush) are currently being counted as flushes, creating more hands that have a flush than there are hands with a straight but no flush. If we reversed the order of these two hands, then there would be more straights because all those hands with both would be counted as straights.

But I think that more likely there is a bug in the software.

I am trying to come up with a simplified example: Suppose we dealt from just a 10-card deck, just 23456 of the black suits. If you deal a 5 card hand, your chance of getting a flush is vastly less than your chance of getting a straight. However, I suspect that if you deal out 7 cards and choose the best 5, you will have a more flushes that straights, because much of the time you will have both to choose from, so you will choose the flush. Certainly if you deal out 9 cards and choose 5, you will always have both to choose, so you will choose the flush. Therefore, in the 9-card hand, there are no straights!

But again, I bet it's a software bug.

[SoBeDude]

You are correct. where is my mind?

you can make a straight OR a flush with one card in your hand.

But the probability of making a straight is still higher (hence less valuable) than the probability of making a flush.

And we play more flush cards than straight cards. Do you play 59o very often? I bet you play 59s more...

Grab a program like TTH and run sims to the showdown with every hand. I think you'll see more straights than flushes.

-Scott

[rkiray]

Does anyone know the odds of getting a flush with 2 random hole card and 5 random board cards? What about a straight?
SW bug sounds plausible, but I think it would have to be two bugs since both the expected and the actual show more flushes than straights.

[rkiray]

Actually I would only play 59o or 59s in the blinds. But I do play hands like 89s much more than 89, KTs than KT, QTs than QT, etc. I have TTH, but never used to run sims. Don't know how. I could probably figure it out easily enough.

Oh yeah, I forgot to answer a question from a previous post. I agree that fish get more straights than good players.

[MaxPower]

I just did a quick calculation. The probability of being dealt a straight in 7 cards is 8.3%. The probability of being dealt a flush in 7 cards is 4.2%.

This might be slightly off, but its close enough.

[rkiray]

Weird, maybe pokerstat has them backwards and the reason the straight number is higher in your calculation than the flush numbers shown in pokerstat is that hands that have both straights and flushs are counted as flushes.

[Ulysses]

I hate to say this, but I would check and fold on the river.

Ugh. You don't think a smaller flush or trips will bet if checked to? You don't think an overpair could bet? You don't think top two could bet? Checking and folding for one bet on the river with a flush here w/ 22BB in the pot is a terrible, terrible mistake.

Check out two current posts of mine in the mid-stakes forum "Not taking the free card" and "What if he had raised the turn" for a couple of similar situations.

But with this action and this many players someone must be on a bigger flush draw.

Always making this type of assumption will definitely cost you a lot of money in the long term.

[MaxPower]

I knew that I made a mistake in the above post. I counted the number of straights and flushes wrong.

Here is the corrected answer:

The probability of a flush in 7 cards is 3%
The probability of a straight in 7 cards is 4.6%

[rkiray]

Wow,

If these numbers are correct pokerstat is seriously screwed up (or I am). I assume these are upper bounds on the numbers that pokerstat should be showing. This is just the probability of a flush or straight, not that there isn't also a bigger hand present also. So straight would be counterfitted by flushes, full houses, and striaght flushes. Flushes by full houses and straight flushes.

[rkiray]

I figured out what went wrong. This was a page I was not used to using. I was reading results for hands where I had two suited cards preflop. For all hands that I saw a river pokerstat expects 4.62 % straights and 3.03 % flushes, which matches very well with your calculations. Sorry for the confusion.

[SoBeDude]

Could you show the math on how to do this please?

Thanks!

-Scott

[rkiray]

I understand what both you and JTG are saying, but I think this is one of the rare cases where I make a tough lay down unless all these guys are maniacs or this is an extremely loose table. With four other players in for this much action and the way some of them go back and forth between calling and raising it is just screaming big flush draw to me. I would not have been surprized that there were both a K and A high flush out there (and I know three flushes are very rare). I thinkg the guys playing the Js both should have gotten out earlier. Top pair doesn't win very often with this many players and this much action. Did they both have overcards? I must admit that at most of the 5/5 tables I play at in CO I would play it exactly the way you guys say. But those games are often very wild and the structures rewards drawing to lots of crazy hands.

BTW, I don't ususally advocate this. About a month ago David Ross and I debated what the showdown win % of a good player should be. David thought a good player will win more than 55% of the showdown. I said using your guys logic that it is much worse to fold for one bet than to lose an entire pot and not all show downs are heads up that 40-50% is optimal. My % is 44, so obviously I don't lay them down that often. But still if these are even decent players, the action was screaming big flush, so I make a rare good play here.

[Ulysses]

I think this is one of the rare cases where I make a tough lay down

Well, when I'm the guy stuck in there in the BB w/ 5d6d, I'll be hoping you're the opponent w/ Td9d! /forums/images/icons/cool.gif

[rkiray]

There is also a pair on the board, and with this much action someone could easily have a full house. I don't do it often but I really thought there was a big flush out there, so I think I would do it. Of course, in the heat of battle, I might have called since as stated in the last post I might tend to do that too much. It's easier to think about it when you have time and don't have your money on the line.

[Ulysses]

There is also a pair on the board, and with this much action someone could easily have a full house.

That's why you fold if it gets raised and 3-bet after you bet out.

[MD_]

One reason why flushes come up more than straights when the river card is dealt is that more rivers are dealt when flushes are possible than when straights are possible. People will play any two suiteds, and always stay to the river with a four-flush. Many of the times the straight would've gotten there, nobody got that far.

-MD

[MaxPower]

Rather than send you my spreadsheet, this link explains it well. Even though this is talking about 7-card stud, it applies to holdem as well.

http://www.math.sfu.ca/mast/people/faculty/alspach/comp20/