I purchased Leak Buster recently - I find it a great tool to assist in looking for trouble spots. I thought it also might prove as a jumping off point for various discussions. For the record, I have nothing to do with the people that write the software - I'm not attempting to shill for the product or anything of that nature.

My most recent Leak Buster overall grade came back as a "B" - a nice warm and fuzzy that reinforces that I'm not way off in my game. But there were seven leaks found, and three of these graded out with a big red "F". Here is one of those three F grades, my BB fold to steal is too high.



The upper left of the image shows that my BB fold to steal is 85%, and my BB reraise steal is 7.6%, so I am employing a fair amount of light 3betting (I use a polarized range, Axs, pairs that I can't call for set value, and sometimes wider if the villain has ridiculous steal and F23B%). That leaves my flatting range at around 7.2% I currently choose to flat hands I feel are ahead of the stealer's range but behind his 3bet continuance range - KQ/KJ/AJ/AQ/88-TT.

So now, some discussion - do you agree with LB's assesment that my Fold BB to steal is too high? If so, how would you adjust? Are you moving more hands into the light 3betting range, or into the flatting range? If so, what types of hands would you choose for each?

well I dont think there should be a magical number, I think it should align with your other stats. I mean if u r playing a 10/8 with 2% 3bet, u cant fold to BB 70% or else it will look like u r mindlessly defending and will get u alot of light 4bets. BUT, if u r playing a 18/15 with 5% 3bet game, ur BB fold to steal should align with ur game. It also changes dramatically with the stakes, i'm not sure if LB accounts for all of that (does it have a formula to align the fold BB to steal to ur game? and does it have different threshold settings for different stakes?) fwiw, i play a fairly aggressive game and my fold BB to steal is 78%

Should also check your w/r at BB.

I currently run 15/11, and my redline is a bit too cliff-like for my tastes. My win rates at CO/BTN do not offset my losses in the blinds, which is my minimum goal.

When you mean pairs that you can't call for set value. Do you mean like 22-77? Maybe I should try that more.

Calling 22-77 for setmining purposes against a raiser with a wide range is a bad idea because the villain is unlikely to hit anything strong enough to pay you off if you hit. Example: BTN raises with K7s, you call with 33 in order to hit your set, and you succeed. Consider how few flops exist with a 3 in them that BTN will also enter a 50BB+ pot with.

It's a bit high but not that bad. Also depends how much players are stealing in you're game. How much is you're Fold BB to SB steal stat. This should be alot less than over all.

Right. I like it. I usually just throw those away to a steal but that is probably bad practice. You raise w/ a speculative hand w/ alot of focus, that is if it hits the flop you are almost certainly going to get villains stack but if not it doesn't really have SDV. Most of the time you will just take it down PF and if they 4-bet it's an easy fold. If they do call it'll be with a stronger range to pay you off if you hit. Makes sense.

also have leak buster but i think stats don't tell the whole story. i have A- overall (mostly 5NL and little 10NL...lol)
but TONS of leaks in my game
as for the blind play. some things might affect your stats:
-what about you're resteal %
-and steal from SB? pfr from SB seems a little low (i'm not an expert though, but it should be close to CO pfr% imo) some people don't like stealing from SB too much (being oop), but i find a lot of nits folding 90% of the times and find it profitable.
-fold BB vs SB steal?
i also think that at micros, if you must have a leak from the blinds, it's way better to fold too much.

um what?

Just play the hand better postflop than villain ez game.

what? I think you missunderstood... mtagliaf meant, if we call with small PPs vs a stealing range, we'll never get enough value when we hit our set to justify the preflop call. With a hand like 88 instead, we have often enough some SDV which justifies calling pre, because we are not calling to set-mine only.

I understand that we're not calling to set mine but this doesn't mean we shouldn't call and play the hand on the flop.

I dont think there is much to play with 22 or 33 on the flop

Sorry for calling your button steal w/ Ace rag off the other day and calling down

Just to clarify, you are still turning these hands into a bluff, you're advocating that doing so on the flop is more effective than preflop.

If so this is somewhat tough because we are giving up our preflop positional advantage completely. Postflop we are completely reliant on our skill advantage, with a hand that doesn't play well 88% of the time first to act. This means our skill advantage needs to be huge. I must confess that my skill advantage (if it exists) is likely not anywhere in the range that this would be a reasonable play. The op's balancing range seems quite reasonable to me, and would limit some of the spewy spots we tend to get in.

In fairness it is cheapest to learn at the micros and if one could ever learn to handle these situations well, then we are talking about a scary good player.

As to the original question, 85% is a pure leak, the stealer can raise every time and fold every time you raise or call and still make money. You have to know your opponent though, if you fold this much to a 50% opener from the button this is terrible. If it's a 7% nit then there is no harm.

I like that

I've done some more digging on my blind defense - I decided to break it out by the position of the stealer. What I found was very interesting.

CO, 298 times, fold to steal 89%
BTN, 350 times, fold to steal 85.4%
SB, 282 times, fold to steal 78.2%

While it's good that I'm playing more against a small blind steal, I don't think it's enough. Venice's blind stealing COTW gives the example that you can play your top 35% of hands against a SB raising top 40%, which gives you a card advantage as well as position.

I re-worked my steal numbers by position, substituting a 35% defend from a SB steal, and my overall fold to steal becomes 80.4% instead of 84.5%. I've shaved 4% off my number right there, and brought my overall fold to steal much closer to the line suggested by Leak Buster.

Making this change, as far as identifying a few more defends from CO or BTN steals, and I should be able to get my numbers in line.

To help effect this change, I have also altered my HUD to show steal%s by position, instead of a composite number. In my first session, I have already noticed that some players will steal liberally from the button but much closer to their hand strength in the cutoff. I think this HUD change is going to help greatly.

@mtagliaf

I don't think it's too bad to play small PP versus a steal (I personally don't do it) but if you have a great skill advantage then it's no problem. If you don't hit a set, then try to rep it. Of course this is dangerous because you have no equity when called. Also, make sure you know villain is capable of folding.

I don't completely agree with your point about not getting paid off with a set bvb. Yes, villain's range is wide but often enough people stack off lighter 'because it's bvb'. For example you can check raise with a set on K32r and try to level himself into stacking of. If he folds, start check raising more.

I've been considering Leak Buster. Does it only compare stats to acceptable ranges of winning players, or does it also dig into how one might attain such an acceptable range? You should be careful adjusting your play to hit a stat range without being very sure that the change you're making is profitable. Maybe that's where the coaching sessions & DB walkthrough part comes into play, no clue.

If memory serves the authors of Leak Buster like a summed loss rate in both blinds of -0.3ptbb/100 and you're not that far off at about -0.37. The first thing I notice is that your SB winrate is quite good. In that spot, you are 21/10 with your current strategy, and have fairly equal flop & turn c-bets. I wouldn't change a thing right now with your SB play. However your BB winrate seems pretty bad and you are playing quite a bit different at 10/5 with very exploitable flop & turn c-bet percentables. If you don't consciously try to play differently from these two positions, it appears you are doing so sub-consciously. From the BB, I think we should actually be defending (with calls & resteals) more than we do from the SB and it doesn't appear that you do this.

The second thing I think is appropriate to consider is exactly how you plan to reduce your fold-to-steal stats. You should find a profitable situation you can implement in the blinds, and then your stats will follow. One thing I see is that your BB reraise steal of 7.6% seems way too low for someone that plays 15/11 overall. I recommend comparing this to your SB reraise steal %, which based on your win-rates and VPIP differences, I expect to be much higher and possibly even double (oops, didn't see it in the original picture...see final edit*). If so, then that's the only adjustment I would make right now. If you're able to find profitable resteals against 2 villains from the SB, then you should be able to find at least as many profitable resteals against 1 villain from the BB, at a cost of one 1/2 bb less.

Since when changing strategies it's helpful to change one thing at a time, I don't think I would be so quick to expand my calling ranges if I were you (unless you perhaps shrink it to add more 3-bets ... depends on what you do with what). Since playing OOP is hard and lacking initiative makes it even harder (which you're guaranteeing when calling), I would just recommend increasing your 3-bet resteals first.
The final thing I can't directly comment on, but suspect is true, is that you are not winning back your share of your BB in pots that limp around to you. I would be interested in seeing a couple things there:

1. Overall loss rate in hands where you have the option to check in the BB. My suspicion is that this is lower than your overall loss rate in the BB.
2. Overall rate in hands where you decided to raise, and the frequency with which you did. My suspicion is that you are breakeven or even have a winrate here, but that the frequency is very low.
3. Overall loss rate in hands where you decided to check. I don't expect it to be pretty since hitting flops is hard, though based on your COTW on the matter I suspect you could be winning back slightly more than the average person. I wonder if your comfort level in this situation is leading you to just check in the BB way more than you should?

I should add that anyone that has noticed my posts in the last week or so may know that I've been trying to improve my blind loss rates too, especially in this last situation. So I suppose I stand to obtain help in this conversation as much as you are hoping to receive help.

*edit; noticed SB resteal is only marginally higher than BB. Sorry for the reading fail... I still posit BB resteal should be at least as high, if not a couple % higher than in the SB. I also now think the BB/check case might be more of a contributor to your worse loss rate in the BB than in the SB than I did before since the resteal disparity is lower than I thought it might be.

** After googling leak buster, I see I've confused it with Leak Finder. Apologies to our creaters of Leak Finder.

wow, great post djames, exactly the type of discussion I was hoping to get going. Let me look up the stats you requested.

I selected "BB" under "Position", and "Unopened, 1 Limper, 2+ Limpers" under "Preflop action facing player. Both are on the "Main Filters" tab. You are correct, my loss rate is -13.39 bb/100, much better than the -52.55 rate I incur in the BB overall. This is 1132 hands. (32000 total hands played this year, overall sample is a bit small but usable I think).

90 times. +256.09 bb/100 win rate (!). Much of this is raising the SB as the only player in the hand ("Hit em with the crowbar. Every. Single. Time." as *Split* would tell us. I don't do this every time, but I'm working on it). With more limpers or limpers in other positions, I am much more likely to take the free flop.

One thing I know for sure about my game is that I'm much less likely to isolate limpers from the blinds. Hands I would happily iso from BTN/CO (KJ/A9), I often simply check. I suspect this is a leak, but I'm following the theory that I want to keep pots big in position and small out, especially with the "trouble" hands.

-36.66 bb/100. This is most of the hands.

Didn't read all of thread, and I think you figured it out, but the easiest way to reduce your losses from BB is to defend super wide to SB opens. It's a big leak of a lot of micro and small stakes players. 3 bet a really wide bluff range and flat and raise a lot of flops. It's so hard to play from the SB against an aggressive BB, so people will mostly start folding a lot. People tend to play really badly when they flat 3 bets OOP, and their range will be really face up, so experiment with barrelling a ton when they check call the flop.

It appears that BB option hands are not the source of your high BB loss rate. I actually expected your BB option hands to turn up a worse loss rate than your overall BB hands, though I see now that by using "lower" I wasn't so clear.

I'd hate to just spam you and ask for more stats, but if it isn't your BB option hands, then it's got to be either the hands where you're calling raises or hands where you're 3-betting. Given your BB option hands were better than your averages, obviously at least one of these call/3-bet sets needs to turn up a really bad loss rate. Of course this also means that for you defending more may actually not be the wisest choice. That is, if one of the calling or stealing set is really bad, and both of these are already in the steal defense camp, maybe you should defend less and not more. But I realize this isn't what you're trying to do, so sorry I haven't been able to help you...

calling raises. 205 hands, -166.49 bb/100. Ow.

Stacked twice in those 205 hands (was a 2:1 favorite in both when the money went in, but villain had huge draws and played correctly by getting it in).

If you remove those two hands, loss rate goes down to -68.92 bb/100. Still poor.

3betting is not a problem - 89 hands, +268.54 bb/100 win rate.

No need to apologize for offering to help me!

So I think your (tiny) leak is in calling raises then.

Sure, those 2 hands clearly bias your bb/100 downwards, but it's not correct to simply remove them altogether though because even if you were a 2:1 favorite in each you still expect to lose 1 in 3. So it looks like you lost about 341bb when calling raises, and you lost 140bb in all but those 2 hands. This leaves 201bb in those 2 hands where you got stacked. Let's keep 67bb of them for a 2-hand EV adjusted loss of about 207bb in 205 hands. Since this is basically 1bb/hand, you should be ambivalent between calling the raise or simply folding to it.

This is a problem. If you are electing to call raises, you should be doing so when the value you expect to gain is more than both folding and 3-betting. You mentioned you flat with a range similar to KQ/KJ/AJ/AQ/88-TT. Sure, it may seem counterintuitive to raise a hand like KJ or 88 when it's ahead of the stealer's range, but until you can make profits with these hands in these spots that equate to more than what you would make by 3-betting, you probably shouldn't call with them. I know the debate of what to call with and what to 3-bet with is old, and sure, we don't want our calling range to be too small so we are not an open book, but whatever, your results are what they are.

Unfortunately though, shifting hands from your call raise bucket into your 3-bet bucket isn't going to improve your leak buster grade. But, if by 3-betting a touch more you could reduce your BB loss rate, would you really care?

Agreed that it seems pointless to change our game only because LeakBuster tells us to. Your analysis shows that me moving hands from fold to call would currently yield no benefit to my bottom line (I would lose 1bb/hand, the same as folding), and the benefit should always be improving the bottom line.

It appears to me (see post 17 above) that the easiest way to correctly shift hands into my call bucket is in SB vs. BB situations. I am folding 78% of hands vs. SB steals - this is too big a number. This weekend, I added SB Steal% to my HUD, and I think it will help me identify better spots to defend. This alone should bring my stats down closer to the LB recommended grade (at least out of big red "F") territory.

I think we can consider this leak busted. (at least the plan is in place for busting it).