Hi all,

I've been wondering for some time what the (first) raise size should be in PL. Obviously raising to 2BB the bottom of our range and to 3 or 3,5BB the top part is no good. Yet most players seem to chose the bet size according to the hand, so what would be a good sizing system (i.e. better than raising always to the same amount, without giving deadly info to an observant opponent) ?

Jean

I've been playing around with this in the games I've been playing...going from minraise to potsize and have pretty much settled with the stt forum's raising amounts 3x BB + 1BB per limper. Sure there might be other situations where you might change it up depending on the player (does he call potsized bets with come hands?) but what I've seen is that 3xbb gets a lot of people calling in tournaments where a potsized bet thats only slightly larger makes everyone fold.

for my non-short pairs[QQ-AA], i still like 2.25-2.75 BB raise-opens.
Always that risk of getting re-raised pot by a SB or BB who woke up w/ trips.
i also mix this up by raise-opening like this with medium-high 2 pr.
[T's up thru K's up] and baby trips [2's thru 5's].
Granted, most of my P/L draw has been 1.10 sng's as of late, but one of the tradeoffs since the intro of .10/.20 and .25/.50 limit draw @ stars is the decrease in traffic @ .25/.50 P/L draw there... and i can imagine these games have gotten quite a bit tuffer. I tend more to a nit/paranoid style, but it always seems to bag me the 2.10 in the non-turbo sng's [sometimes the 3.90!]

In practice or in theory?
If in practice, what are the limits and what can you say
about your opposition?

I don't use the strength of my hand to determine my bet size predraw, I use my position. UTG I only raise 2x BB because I can fold to a reraise without losing as much. Later positions, I will raise 3x-4x.

I don't want my opponents to gauge my hand strength on my bet size because I think it gives away too much information. UTG, I will raise from .50 to $1 with AA or a pat hand.

3 times the big blind, every time.

OLNY SOLE exception is UTG when I'm pat, so I go in light hoping for a re-raise.

Playing to a raise is a far more interesting question.

In "theory", or in practice vs very tough opponents. My question is about a static system which "does better" than raising to always the same amount against a "perfect" opponent who knows exactly my raise sizing strategy. Now I'm aware that the question formulated as it stands is not correct, as the perfect opponent's strategy also depends on my strategy postdraw, and vs a 3-bet ; yet I hope you get my point.

For practical matters, I'm playing PS 2-4, and this question would mostly interest me for challenging tougher opponents (those who play 3-6 and above).

For instance, I would consider the strategy "always raise to 3BB" better than TomTom's "always raise to 3BB unless I'm pat", as an opponent knowing the latter strategy could dodge our pat hands sytematically.

Cheers

Jean

For cash, the simplest unexploitable would be 3xbb+1bb per limper.

I do remember reading something in one of harrington's books where he talks about a system or bet sizing based on the second hand of his watch to randomly mix up your play and not give anything away. Something like 80% of the time raise 3x, 15% 2x, 3% pot, 2% limp would work or any other combination with a vast majority of the time raising 3x. Basically before you act you look at the second hand on your watch (or I guess get a clock addon for your computer that shows seconds) and portion it off and something like from 0 sec - 45sec raise 3x, etc.

Randomizing the whole range is no good either ; better use always the best amount (2x, or 2,5x, or 3x or whatever it is).

Personnally I think it should be better to use always 2x than 3x, as odds to outdraw are never good at 5CD.

On the other hand, it's very painful to get paid by a lot of pairs in the BB when you semi-bluff raise with flush/str8 draws. So I guess a good bet sizing system would attribute 3x to draws ; but to disguise all these bluff hands one should put most the made 1-card draw hands (two pairs and trips) in the 3x range. But then only remains single pairs and pat hands to go in the 2x block, which is no good, since raising 2x would almost give away our hand.

So I'm stuck so far, keeping the "always raise 2x" strategy ... looking forward to hearing other systems though.

Jean

Why is randomizing the whole range no good? It doesn't give away any information about your hand except that its a hand you raise with, and can confuse your more observant opponents into thinking they've come up with a way to read your bet sizing when in fact they haven't.

The majority of the time you're raising your standard amount, but every once in a while its mixed up in a completely unreadable way.

I like 3x better as you want to try to get as much money in when you have the best hand as you can and since you and your opponent have already seen 5/8 to 5/6 of your hand so the luck factor is greatly reduced.

What I mean is the following : if (say) always raising 3x is better than always raising 2x, then randomly raising one or the other just makes the average of the two, so you'd be better off raising always to 3x.

Ok for the "try that your opponent get a wrong read" argument. My question is quite theoretical though, thinking about an opponent who knows exactly your raising strategy.

I'm quite sure that from utg or the hijack you want to open
raise to only 2xBB against "opponent's that know your exact
strategy" to play all of the profitable hands. Obviously, if
you always open raise for the maximum, i.e., 3.5xBB, you
won't be able to play as many hands compared to the case
where you open with hands at 2xBB. If you decide to always
raise or fold from utg or the hijack, you absolutely need to
keep the 2xBB raises to open with, not just for minimal one
pair hands, but hands such as straight flush draws and very
big hands. If you want to open with a different raise size,
say 3xBB, clearly you want the range for this raise size to
be "on average", better post-draw; otherwise, you're going
to be putting more money with weaker hands which can't be
correct. A "balanced mixed" strategy employing exactly two
different opening raise sizes won't be easy to construct to
maximize EV, and it's unclear that having more than two
raise sizes will perform any better (I doubt it). Also, for
the larger raise size, it wouldn't surprise me that it could
be either exactly or very close to 3.5xBB since normally in
optimization problems, the extreme values give solutions.

I see that many players make a compromise by almost
always open raising for 3xBB, but that is almost certainly
not optimal against "very tough opposition". There are a
few players that sometimes open raise for 2xBB, but I'm
not sure if they regularly open for that amount. From the
button, I think you should open raise for more than 2xBB
and it depends on the range of hands that you think will
play well overall against the two specific opponents in the
blinds. From the button or the cutoff, you probably want
to have your minimum "level" at 2.5xBB or 3xBB to open
raise with.

In practice, I think the 2xBB may not perform as well as a
different raise size because players like to call with subpar
hands and when you only raise to 2xBB, although you may
be able to play more hands, the overall EV in practice may
not be as good as the overall EV in practice if you always
open raise to 2.5xBB since many "mistakes" your opponents
will make will be larger. On the other hand, if the BB and
the SB defend their blinds "correctly", and other players
cold call "correctly", you need to open for 2xBB for most
of your opening range since you should be folding your
worst opening hands to a (near) maximum reraise. So
against very tough opponents, you want to open for 2xBB;
however, as one of my friends might say, "Why would you
bother to play in such a game?"

Interesting answer.

Yah, you have a point there. This said, at the PS tables starting from 2-4 and up, there often is at least one (probably strong) regular each time there is a bad player ; so having a plausible strategy against such opposition seems necessary.