I have used a few probe bets most by not counting the pot size enough (live) and I don't think they work in live games. It give too good odds for someone to call you. I think 80-100% of the pot is a good bet, but I am pretty new at the game.

I think the bet has to be at least 75% to be taken seriously.

Anyone else have comments.

In a tournament it might work because there at not many chasers (you lose your chips and you are out).

Hmm... I think it depends on your opponent. If I know the way someone raises overs and I have a smallish pair I may lead into them with a half pot bet if the flop looks favorable to me. I am leading with sets and straights here too against more aggressive opponents the exact same way. If I made an error they will generally raise, and I can make up my mind if they are right or wrong.

I don't do this with overly aggressive players that love AK unimproved to the river, or good players that may raise me for the heck of it because my bet is weak. Generally this is reserved for weak opponents in their raised pots in or out of position and isn't something I make a habit of doing unless I'm pounding someone that likes raising 3x out of his raised pots consistently for half pot bets.

This one way that you can use consistency to vary your game against the better opposition. The better players will see you running this play against weaker players, when you run it against them they will be more likely to raise, but you will have the goods. It is a pattern that they can follow and you can manipulate and exploit against the aware as you will be feeding them false and incomplete information about the way you intend to play against them.

Speaking of that style of betting someone just ran it against my AK with what I suspect was 77 or 88. I opted to take one off the top and hope a scare card of some sort fell or a helper card for me. Another undercard rag paired the board and I gave up for another half pot bet.

I'm also very interested in this. Some randoms thoughts

I have tried to make probe bets / underbet, but I get into too many tough situations for my liking.

To make probe bets you need to be very good at reading your opponents.
Its also important that you know how observant your opponents are.

I think its better to do in live poker than online.

I think its more a tournament play than a cash game play.

I love this play when you're getting deeper in tournies.

People are going to be a lot tighter at the FT of a tourney than they will in a cash game. That's just how it is. I'll give you an example from my lsat tourney.

400/800 blinds w/antees, i have 20k. BB has me covered.

[9d] OTB, i open raie to 2400. BB flat calls.

about T5500 in the pot. [Jd][5d].BB checks my, instinct might normally be to check here, but i want a pot, so i bet 1600. he calls.

T8800 in the pot. Turns the he checks, i bet 2400. he calls.

T13600 in the pot. rivers the , he checks. i bet 5000, he calls and mucks his [Ac]., i win a T23600 pot

Assuming i bet hard on any street, there's almost no way he can call.

In a normal game, the 30 BB pot i pick up in this hand would equate to a $118 pot at my normal 2/4nl game - only slightly above average. But at this stage in the tournement, a 30BB pot is huge.

Hey, thanks for the comments!

I really tend to agree with ice on this one under most circumstances. I'm still wondering about the exceptions, which imo still just has to be a weak-tight game (tourney situations also, but I haven't been messing with those as much lately).

Anyhow, just as empirical result of trying out this loose strategy the other night online: The lowest stakes tight game I could find was at the $100 level, so I just went there. Tried the loose strategy with probe bets and lost $50 pretty fast... I won a fair amount of it back with TAG play once I decided it was time to quit this stuff, but, anyhow, bad experiment... I really should have known, as the table had a few bluffy spots and, in general, despite being pretty tight, wasn't afraid to put money out there.

Anyhow, Dan Harrington's example from later in the book: You call a raise with 88, and the flop comes Q62 (something like that), so you make a probe bet. Ok, I can see this in principle, but it feels to me again like asking for trouble--at least unless you're betting sets the same way. But on sets I hate to do it because it doesn't give you enough leverage to stack them. AK presumably lays down either way to a half-pot-ish or weak bet (although at least for me, if it's significantly less than half the pot, I do consider going ahead and representing AA), but AA still has to raise whether your bet is half pot or pot, so I'd rather it be as big as possible.

I have had some opponents who get scared on overpairs and just lay down to a pot-sized bet into them. That type of opponent might be worth working on with probe bets, but that again speaks, I think, more for the idea that its greatest value is against weak-tight opposition.

I'll continue to keep it in mind if a table configuration comes together where I think bets like that might make sense in a cash game--and, please, anyone else, do post in great detail any situations where you've found this to work. At the moment, I'm still just not really seeing it...