Preflop raise sizing:

So, what is preflop raise sizing? Well it's the amount that you raise preflop ldo!

/thread


Well, I do have some things to say about it as it turns out. But, before we begin I have a few terms to define (if you know terms scroll down to, CONTENT STARTS HERE:

SPR: stack-to pot-ratio; a pretty self explanatory concept that is very critical, a lower spr makes pots easier and more straightforward to play.

3bet: the 3rd bet. the blinds are the first, someone opens is the 2nd, someone who then makes another raise is said to be 3betting.

4bet: see above, the 2nd better then reraises the pot, thus putting in a 4th bet

cold 4bet: after a 3bettor has put in their 3rd bet another different person puts in the cold 4th bet

open: put this in front of a word if they are the first person in the pot e.g. open raise or open limp

iso: raising for the purpose of playing hu, usually we try to isolate limpers because they are weak

steal: a raise made in lp (late position) for the purpose of taking down the blinds.

squeeze: a raise made when there are several people in the pot, usually a raise and a few calls, with the intent to "squeeze out" the people in the pot and take it down.

CONTENT STARTS HERE

So one of the most contentious issues in poker has been the sizing of preflop open raises. That is, how big do we make our bets when we are first into the pot? I am here to present my view on this issue. The two most commonly sized opening that I saw from regs as I moved up through the limits was 4x and 3x the big blind from all positions. Let's compare the advantages of these two sizings:

Value extraction from loose passive occurs by putting in large (potsized) bets street by street by of course using the classical b/b/s. Our first comparison of 3x and 4x sizing will be looking at the geometric growth of the pot if we vary our open raise size by only one (1) big blind. In both scenarios we are going to assume villian is a loose passive fish in the BB who will c/c down with A2o on AQT53 unflushed while we hold AKo and that we both have 100bb stacks.

if we start with 3x:
preflop: raise to 3x, fish calls, pot is 6bb
flop: bet pot (6bb,ignore sb), fish calls, pot is 18bb
turn: bet pot (18bb), fish calls, pot is 54bb
on the river now we have 73bb behind and can't really shove it in because it looks like such a gross overbet that the might might fold, so we bet about 60bb and the fish snaps us off and we profit about 87bb-rake.

however, look at what magically happens if we open to 4x:
preflop: raise to 4x, fish calls, pot is 8bb
flop: bet pot (8bb), fish calls, pot is 24bb
turn: bet pot (24bb), fish calls, pot is 72bb
this time on the river we have 64bb behind and we can easily shove river and not have it look like an overbet and get snapped off no problem.

so at first glance, it seems like 4bb is the way to go and I believe that if your game is loose passive and filled with fish with a lack of 3betting that this is true. thus, for general play at the micros, 25nl and below imo, opening to 4x bb is correct. however, as you move up in the limits, players begin to 3bet you lighter causing you to fold some of your raised hands preflop and raises don't only start to mean the nuts. thus, 3x becomes a better option (at least imo). this way we can still play large pots against the fish (which are a bit less plentiful) due to the geometric increasing nature of the pot without having to make large cbets at regs which fail more often and without killing too much of our bottomline because of how much we have to fold. however, as everything in poker is dynamic, sometime you must ADJUST, if someone is 3betting you light at 25nl, open smaller, or a much better option is to LEAVE THE TABLE.

Note: some very successful regs at my current limit (100nl) open to 4x but I would say at least 95% if not 95%+ open to 3x.

The last thing that I have to say about preflop open raise sizing deals with stealing. When I am stealing OTB (on the button) I will minraise any two cards. Since we have the button we have ultimate position and we will be able to use that to our advantage to hopefully outplay our opponents postflop. Additionally, since it is only 2bb, you will find that some of your fishier opponents will call waaaaay wider than they should. I personally minraise everything including AA and 72o for disguise and because of the fact that it is very rare for me not to have a shortstacker or a reg in the blinds who will occasionally 3bet me light or shove on me light. However, if you have only fish, feel free to raise up that AA and if you have someone with 100% vpip, don’t worry about folding your 72o. Like I said above, ADJUST, leave if there are 2 shortstackers to your left unless you have position on a fish. The last, and most important, reason that we open to 2bb is this: If we open to 2BB, we risk 2 to win 1.5, so we need our steal to work 57% of the time. That’s it, do this and you should show instant profit.

The last position I have to address opening from is the small blind. Due to the fact that we are oop against the bb I suggest raising to 3bb to dissuade the big blind from calling too wide because it will be extremely difficult to play. Don’t be ashamed to open fold and give the big blind a walk. Before this turns into a dissertation on stealing let me refer you to Happy Pixel’s excellent, excellent article on blind stealing written way back in CotW #4: http://forumserver.twoplustwo.com/78...ealing-419666/

And of course, Pokey’s article on blind stealing, if you haven’t read everything Pokey has ever written I highly suggest you do: http://archives1.twoplustwo.com/show...Number=5348855

Ok, so that about covers open raising, so now let’s move on to limpers and iso’ing. So as it turns out this too has been covered during the CotW series very well by KurtSF here: http://forumserver.twoplustwo.com/78...impers-431379/

Kurt pretty much owned limpers hard in his post so the only thing I truly have to add is this; look at the stack size of your limpers before you decide how much to isolate. If they have a smaller stack (say 35bb) then maybe you should iso to 3bb instead of 4 or 5 bb. The reason for this is that you will have no problems getting stack sizes in if you hit well due to only needing 2 streets to do so and for those times that you do lose where they call your cbet and pot the turn or shove over your cbet or donk out, you will lose less (not to mention you can cbet smaller, say ½ pot). As for what ranges to iso, this depends very much on your opponents’ tendencies/stack size as well as your comfort in playing things like TPNK or MPGK. In general thought, isoing any two broadway cards and a bunch of suited connectors/one gapers is the way to go against the bigger stacks, and the broadway cards against the smaller stacks.

3betting, 4betting, cold 4 betting, and squeezing:
Ok, so now we aren’t the ones opening the pot. Someone, or a few someone’s, have shown that they like their hands a lot and have invested some monies. So you have picked up your hand and decided that they are weak or your hand is great so you have decided to do one of the above possibilities. A few general rules to keep in mind before you 3bet, 4bet, cold 4bet, or squeeze either light or for value:

1. You want to keep your value raises and your bluff raises the same size against thinking regs or else you become too easy to read. Against fish who aren’t paying attention feel free to adjust your raises.

2. You want to bluff the smallest amount that will work the highest % of the time and raise the largest amount that will get called when for value.

3. You don’t want to be playing these large pots multiway and thus should probably squeeze larger than your standard 3bet size (say there is one raiser and two callers I might squeeze to 15 or 18 rather than 9 or 10). Where a standard 3bet might be 2.5x or 3x ip or 3.5-4x oop.

4. You want to be 3betting bigger oop because it is going to be much harder to play postflop otherwise. You can shove your big hands in much simpler fashion if you just bet bigger

5. If you put in 1/3 or more of your stack preflop you have to stack ATC (note: I lost the link to this math and would greatly appreciate it if anyone found it), thus I generally make all my 4bets 25% of my stack no matter what.
6. You generally want to polarize your range when you do any of the above.

A Note About Shortstackers: If a shortstacker has 20bb and raises to 3 and then I 3bet to 12 and get a fish and reg calling behind me and it then folds to the shortstacker who shoves, I can only call and not raise. The rule is this; if there is an all-in raise that is smaller than the last previous raise (in this case an 8bb raise coming after a 9bb raise), then no more raises are permitted. Beware of this fact! I learned this the hard way when I held AA and got set mined by JJ. In the above example, it would be much better to raise to 10 or 11bb because we know the shortstacker is going to shove and we also know that we don’t want to see a possible multiway flop and would like to reraise to set up a lower spr and a nice side pot or just shove preflop.

For more about crushing our sworn enemies you can see CMAR’s incredible post: http://archives1.twoplustwo.com/show...umber=12157664

Preflop raise sizing ties DIRECTLY into betsizing which is what I think is the biggest leak of all micro players. A mandatory read is Split’s article on best sizing that is one of the best articles dealing with poker that I have come across ever. http://forumserver.twoplustwo.com/78...-value-325157/

1st!

tl;dr, yay scary post number

Very tired, but good post! Will talk later (or someone can pick it up for me) about adjusting your raise sizes based on range strength. The stuff about position and the general type of table you are at is VERY good.

I can adress this too but it is something I don't really like tbh. I would be happy to have a discussion with ppl about it itt though. I couldnt really identify too many strengths or advantages in that manner and would think that you would need to keep up a truly random sequence of when you raise what hands bigger or smaller which is impossible afaik. I might not completely understand the concept though.

There is a fairly long section in NLTAP about preflop raise sizing and they suggest raising bigger (4-5 times) with your hands that are apt to win big pots, this would include small pairs, suited Aces/connectors, basically hands that are easy to play and can make well concealed monsters.

They suggest raising smaller (min or 3x) with hands are apt to make top pair type hands and usually don't want to see the pot get very big post flop, this includes basically all of your high card hands even as strong as AK.

This seems to be the only credible source I've come across that explicitly tells readers to vary their raise sizes based more on the strength of their cards, as opposed to situational factors such as what type of opponents they are playing, their position, limpers in front, etc.

Thoughts on this approach?

Also, I think the min raise in steal position is really interesting and I've been experimenting with that myself lately with good results. It is absolutely incredible how often you steal the blinds with a min raise. Also, since most of your raises in other spots are the more typical 3-4 times variety, players are not very likely to play back at you or view these raises as weakness. If anything I think people probably think they smell a rat and lean more towards giving you credit for a monster hand that wants to entice them in.

nice work.

if we ignore the fact that this is EXTREMLY exploitable if anyone realizes what you are doing, and what with us playing a million hands an hour with session reviews and stuff this is a very big if,

i find that most of my money comes from fish, and the best way to stack these fish is to make any hand that is basically TPGK+. the easiest of these hands to make is TPTKish kind of things which require me to get usually about 60bb each into pot cause that is a fish type stack. hmm, another way to put this. i raise 55 a large amount. fish don't care either way. they will call with KTo 5x or 2x whether i am raising 55 or AK. so i cbet my 55 on JT4r board and get called by a fish. that's 5+6ish or so bb for the cbet that i am losing. and i am now in a WA/WB situation that is nigh on impossible to play. where as if i have AK, i can potentially actually stack with two overs + gutshot straight cause will be happy to get it in with KTo or whatever.

ok so that rambles but i hope you get the idea? i mean i play Axs from almost all positions in an attempt to outcooler the regs and pick up more outs as i fire more barrels, but against a fish type game i would rather raise the same with everything. actually, in a fishy type game i would rather raise as small as possible with pocket pairs and as large as possible with AK. in a reggy game, i think the exploitability is paramount to a reason that approach is bad.

so it's both sides of the possible coin (maybe there's a 3rd) in a reg type game i think it's dumb to do anything to telegraph your hand. and in a fishy game...

so that kinda rambles but i hope it gets my point across? sorry if it doesnt it seems clear in my head but then again, im tired and i just tilted :/

good post, fml got to read a lot if i follow every link

great post sir...and thnx for the shoutout on my old betsizing article.

(for those that read my article, know that some of the intricate details (like my 3b size for example) are different than i use now, and imo, wrong for today's game conditions. that article was written about a year ago, and much has changed in that time.)

Was looking forward to this COTW. Was not disappointed. Excellent post!!

That sounds very counter productive. Logically you'd like to raise BIGGER with your BIG hands (AA, AK) to set up SMALLER SPRs, and thus be able to stack off with TPTK without feeling bad about it.

Simararly you'd like to raise small with SCs to set up LARGER SPRs.

Right??

First, Vanguard, great stuff, thanks!

As far as RE's comment, I could be wrong but I think RE is talking about adjusting your raise sizes to your RANGE not your specific hand. In the Ed Miller ebook SSNL there is a sizeable section that discusses this. The general gist is that if in a certain positions (say UTG) your range is strong (say TT+, AQ+) your preflop raises from this position should be a bit higher since you are more often than not going to have a playable hand on the flop. Conversely if your UTG range is 22+, AJ+ (or whatever) you should tend to size your bets a bit smaller because in the long run you will be giving up more on the flop. It's not really exploitable because you will still be consistent from each position, and your range is pretty obvious to anyone with a HUD. A bb or half a bb will make a big difference over time.

I think this is sort of intuitive for most players. Personally, I open UTG/UTG+1 for 4x, MP-CO for 3.5x, and BUT/SB for 3x. Obviously this changes if their are limpers but rather than 4x+1, it's yx+1 where y is your standard open from that position and then any adjustments depending upon the nature of the limpers.

Anway, I think that's what he was getting at, but if not, just ignore me.

This is true of any play. No matter what you do, it's exploitable. ABC is exploitable. TAg is exploitable. LAg is exploitable. That's why mixing it up is so important. The good books always preface individual maneuvers by saying that these should only be done from time to time. If you try too many fancy plays, you'll be demolished. I don't think anyone is saying that you should raise 3x with suited connectors every time. Often, it's profitable to just step into the slipstream and let the other players do the work.

Well, it's almost true. Theoretically there's a perfectly balanced GTO strategy that is unexploitable... however no-one plays perfectly or even close to it, and so exploitative strategies (while themselves exploitable) are always going to achieve a higher winrate than a perfectly balanced one. If you open up a leak in your game in order to exploit someone else's leak, you only lose out if they are actually going to adjust and take advantage of that - this is why flexing your bet sizing against weak players is super important, especially at micros where a lot of players fall into the weak/unobservant category. You can open to 12bbs with AA if the table dynamics call for it.

In short varying your raise size based on hand strength is very exploitable but doesn't necessarily mean you shouldn't do it.

oh and excellent post OP

What is your 3 bet sizing now? What are the reasons for the change?

Thanks

3x for the 3b...with slight variation dependent on hand/opponent/position

more light 3b-ing = smaller size is better for metarange and also makes a favorable SPR regardless

Worth the wait!

RE and Vanguard, imo the only players against which you should be varying your bet size based on hand strength are big fish. Balancing, exploitability, telegraphing, shania... all that goes straight out the window and just pot (or overbet) for fat value with your big hands.

cjk, interesting stuff there. So theoretically your could change your "standard" betsizes when switching gears.

I don't like the varrying raise size by position there though. I've talked to some folks recently that do simmilar - 4bb from EP, 3bb from MP, 2bb from LP/BU. Why would you want to play your smallest pots from the best (winningest) positions on that table? Of course, I haven't worked on balancing their ranges like I have mine, so maybe I'm off track. I guess this would be good for someone like a 9/6 who has a narrow range from all positions to get anyone to play a pot vs you in LP, but it doesn't strike me as a good way to size preflop bets for a exploitative style of play.

Read DiggerTheDog's "nit boot camp" for a good primer on non-exploitative play.

Agreed in theory but at least at nl50 and I'd suspect it's still pretty true at nl100, people are folding wayyyy too often to reraises and are not really adjusting their range based on raise size (ie, they're not going "omg he only 3b me 2.5x instead of 3.5x, I must turn 44 into a bluff and 4b!")

I agree with this one. This is sort of a function of someone's fold to 3b %. If they're folding like 25% or something, then obviously you can raise much bigger for value (consequently, you should be 3b'ing almost never as a bluff vs these guys) and if someone is folding to 3b's 75%+, then you can raise much smaller with your whole range.

I 3b more OOP not because it's harder to play (3b pots are basically a function of "do I have at least top pair - or big combo - or not?") but to cut down on IO, to win a bigger pot when bluffing, and to create more FE.


Right on! I'd like to add/point out that this means you're often making just slightly larger than a min 4b, but you still have awesome FE and put your LAG opponents in a tough RIO spot when they resteal.

Hell yeah. I think it changes based on position though. I like to keep a super polarized range (big hands for pure value which I will stack off with pre, complete trash that I'm not put in a tough spot if I get 4b) in the blinds, and a wider but easily playable range IP - stuff like Ax and Kx suited in addition to all my other stuff. Probably comes from playing 6-max.

Great post. =)

excellent post OP, will probably read again.

Honestly I'm still experimenting with it and am not sure myself at this point. The other thing is I think the spread has to be fairly narrow, say x, x+.5, and x+1 across your various ranges. I do agree that since the button is most profitable you generally want to play bigger pots so making smaller raises seems counterintuitive. The question becomes does the theoretic benefit of using range to size bets outweigh this. I really don't know.

One thing I realized though after I hit post is that using UTG as the example doesn't illustrate the point well. Ed and Co were really getting at the fact that since in positions like the button where your range will be weaker, your raises should just be on average smaller than in those positions where your range is strongest. They go on to say that it is imperative that regardless of the actual hand in your range, you make the standard raise, so if you are open raising the button with AA or K4s you make it 3bb either way. I think it is this final point that drives their whole concept in that you are masking your great hands by "hiding" them in a range.

Great CotW!

Here are my comments:
I would only do this if the pot is unopened. If it has been open limped, you are just asking for a limp fest. I would default to the 3/4bb+1x per limper. Oh, and I would only do this with my standard Btn range and not ATC. If it is unopened, I believe that min raising with ATC on the BTN (and maybe the CO), is a +EV move.

I tend to make pot sized 3bet raises. This will make any call a 2/1. Usually this folds back to the raiser because it usually conveys to the table I am willing to play for stacks. Which I may play for stacks, if there is a call. If they 4bet me, or there is 4bet out of left field (not original raiser), then I can re-evaluate the move.

I have also been experimenting with a min 3bet to iso the raiser with a premium hand. (Example: on btn w/KK, MP raises to 4bb, I raise to 7bb) This scares most people out of the pot and is very difficult for the original raiser to fold since they are getting the pot odd to call. This seems to work very well on the aggressive folks (TAG/LAG/sLAG). Of course, after they call, I punish them on the rest of the streets also.

if you potsize your 3b it is generally about 3.5x

(raise to 4x, we pot so 3b to 13.5 which is ~3.5x)

its not bad...but you could still go smaller. you apply the same pressure while giving yourself a better price

Split,

You are saying that I should give better than 2/1 odds? By going smaller, you are increasing the equity the villian has in the pot and decreasing the equity I have. Sure, you can save a BB when the villian has you dominate but that is not as common from my experience. How much lower are you suggesting?