Hand Analysis: Dead Money Grab Vs the Stop and Go

The Scenario: You are playing an online $10 buy in $1250 guaranteed tournament. With 40 big blinds after several limpers, you limped in late position with suited connectors figuring to have good enough implied odds. 3time bracelet winner Dutch Boyd on the button has made an isolation raise of around 1/4th his stack, and you no longer are getting good implied odds, but are getting 2.7 to 1 pot odds to call.

The Problem: Without knowing Dutch's exact cards or exact range, is it possible to play 87s profitably at this point?

The Set Up: Leading up to the hand dutch had been powering the table quite well and shortly after started to pick up AA, 44 and AK a few times and was on a pretty nice run of cards. So his image is more wild than usual. Normally he preaches being the bully without looking like the bully. However, sometimes you catch a run of cards and you end up looking like one more than you intended. I think that's the case here.

I believe that may turn out to be important from the perspective of chefward. Since Dutch's image is more wild than usual, chefward can more easily diagnose Dutch's hand range as wide, and determine that he won't have a hand that can call the flop quite as often as he might otherwise have assumed vs your typical player. Even though dutch hadn't made a lot of these big isolation raises, it'd be hard for chefward to give dutch credit for only premium hands here and he'd at least have to include some broadway hands and suited aces and maybe suited connectors and lower pairs in his range. That will translate into dutch finding it slightly more difficult to hit the flop enough to be able to call.

Dutch decides to from the button to isolate his standard $3300 at this level PLUS 1x per limper. The raise size actually should be 8100 if you want to be picky. Aside from a philosophical discussion on whether or not the "standard" should be normal raise plus 1x or 0.8x or 1.2x, or if he should make a multiple of the pot when he's going for the "dead money grab" I am entirely fine with this raise size and don't have any qualms about it.

I analyzed the raise over several limpers for 1.5x and 2x multiples of the pot in this "dead money grab" post. It really depends on the limpers limp range, and their ability to trap. But even with much larger raise size than the 8/9ths of what's in the middle, there's a mathematical argument for raising any two. The argument is entirely mathematical but requires assumptions about opponent's limping range that are speculative. Since the risk of 8/9ths is far less than the 1.5x and since the positional player usually has an advantage, the dead money grab with the J9o is perfectly fine.

I say no problem at all making this play as the raiser, you have to do something to keep up with the rising blinds, and waiting for aces probably isn't going to cut it.

BUT... what about chefward2 who's sitting there with a suited connector? I don't think he can play such that he can make it unprofitable for Dutch, but that doesn't mean he can't fight for his share of the dead money in the pot.

So say you're sitting with 87s after several limps and a seemingly LAG player raising behind you. You're getting 17,000 to 6400 or 2.656 to 1. Certainly if you knew the hands were always checked to the river you'd only need to be 27% to win. But it will almost never check to the river. Especially out of position with shallow stacks, you can't make the usual assumptions. You can't assume postflop that "bets break" postflop so you cannot just assume if you are 27% equity it's a good call. Instead you will be committed on the flop to either calling or folding.

While you may get some implied odds, You most certainly are NOT getting the implied odds to hit 2 pair or a straight or flush. You do have some fold equity on a jam, but probably not enough to jam any flop here. Even if you include 8+outs in addition to any two pair or better hand you are only going to be hitting around 25% of the time. With that range of hands, depending on Dutch's range, you'd probably need to be getting much closer to 3 to one pot odds because you'd probably only be slightly ahead of Dutch's calling range.

So the next question is can chefward just shove every single flop profitably? I say the answers also no.

You're not getting 1:1 on your steal attempts because you're just calling preflop. You're actually risking over 30k to only get the 17k that's in the flop before your decision. You gotta be right close to 66% of the time to break even on your pure bluffs or make enough from your shoves when you do hit to make up for the difference. Probably not happening if you always shove.

We certainly could try to come up with a more precise answer to this in terms of what sort of range our Dutch has to have and has to raise with and call a flop shove with to make a shove with any two profitable, but it probably will involve folding some pairs and when you are getting 2 to 1 it's hard to fold a pair since you will only flop a pair 1/3rd of the time and you probably will have some outs... I just think Chefward's opponent (in this case Dutch) will have a pocket pair or flop a pair or combo draw too often for an autoshove to be correct. And if that's the case, where does that leave you? Does that mean you should fold?

Not so fast.

In order for a call preflop to be profitable for chefward, he has to be creative. The stop and go play with any flop is too reckless and the fit or fold style is too tight. So we have to determine if there is somewhere between where it may work.

How much equity is it going to cost chefward to also push with any pair on no facecard boards and pair with backdoor draws such as 3 to a straight with 87 on a 76J or 89A flop for example? Well, on a board like 56J with a single spade he's still going to be 27.37% to win vs KK. That's not bad at all. Sure he's only 8.25% vs a set but that's a small percentage of his Dutch's range.

While I think it's probably more ideal to take an exploitative solution of checking our made hands far more often, and open shoving our semibluffs and pair with backdoor draws, it's far easier to look at this as more of an all in or fold calculation, and it's already pretty complicated with shove or fold, without adding in a mixture of checks folds and check shoves and shoves.

Chefward may even be able to shove with only a backdoor draw. Sometimes he's going to be against AK, get called, and actually be ahead. If he shoves with a 5 and a spade or a 9 and a spade and no pair he is 10.40% to win against say KK and 7.37% to win vs a set, but he is about 30.6% vs two overcards. Certainly that's not terribly ideal, but when you have a lot of fold equity and also are planning on shoving with straights, flushes, two pair, trips, and 8+ out draws the collection of all your shoves gives you pretty good equity when called.

So let's just look at the equity vs a few basic flops, estimate the probability of a similar flop, and try to estimate the overall equity of the decision to shove certain flops. Then we can look at the dutch's approximate raising range and approximate the probability he is going to find a call there.

So let's start with the pure double backdoor draws.
On a 6s Kh2d flop I don't know that we'll get called as often as we should because of that King, but just so I try to approximate the averages of the same flop with a T,J or Q flop I'm going to include AJ+, 99+, 22,66 plus KT+. When I do this, we have about 15% equity. Certainly could be a lot worse, especially since on these types of flop Dutch will probably have a decent chance of folding. I have a feeling shoving with ALL backdoor draws is too much and will tend to make the AVERAGE equity when called too low, but I think including some of these shoves may end up being right just because we aren't going to flop made hands and strong draws often enough not to include some bluff shoves, nor can we shove on the AK2 type of flops where we have nothing at all and no backdoor draws.

What about a 92J flop. This represents backdoor flush draw with the "idiot end" of the gutshot. If we have 78s and are facing a calling range of 22, 88+, A9, T8,T9,QT, any Jack in his range, some ace kings and some Ace queens (because that's a hand that he may fold, he may not so by only including some we weight for the possibility that he doesn't) Then I come up with about 22.9% equity with 78s. Certainly that's not the best flop, but once we add the fold equity in on a flop like that, it may be okay to shove as part of our entire shoving range since our overall range will be much stronger than the lowest 25% of our shoving range.

If we change the flop to 98K with backdoor flush draw AND now we have a pair and backdoor straight draw, now the equity is up to about 37% given the hand range I come up with. Pretty good.

Now if we add in open ended straight draw with backdoor flush draw we shoot up from the 15% equity we had with only double backdoor draws to 38% equity.

If we have a flushdraw and a backdoor straight draw our equity is nearly 40%. Actually, I forgot to add flush draws in Dutch's range.

If we crush the flop with 2 pair, we are probably far better than 75% up to around 80% or as bad as around 70% depending on how much credit Dutch gives us and depending on whether or not there are flush draws for Dutch or backdoor flush draws for us.

If we flop a straight we are probably above 85% vs Dutch's calling range.
If we flop a flush we are probably about 85% as well since many hands that hit the flop will also have a redraw to either a fullhouse or higher flush.

The difficulty in the math is now subtracting the overlaps. For example, there may be a 16% chance of hitting a gutshot draw on the flop, but that includes when you hit gusthot as part of a combo draw including gutshots, double gutshots, gutshot with flushdraw and/or overcards and when you do the math of backdoor flush draw it is going to include backdoor flush draws plus gutshots or open ender or whatever.

You are not looking at the probability of a double backdoor draw, but the odds that a double backdoor draw isn't also a gutshot or a straight draw, or otherwise you end up counting them twice. I'm not going to do all of the math here, instead I'm going to do a really rough approximation. If anyone knows any tricks or software that can help, I'd like to hear it. I think flopzilla may be the tool I need if I want to be really accurate in this stuff but there are others too, I'm sure.

Based upon my calculations, assuming the equity stays as calculated if Dutch's calling range represents 60% of his holdings, it's not profitable to include double backdoor draws and gutshot draws and if we do, overall the strategy is unprofitable by over 300 chips. If Dutch calls 50% of the time, it's not profitable to shove with double backdoor draws, but it is profitable to shove with gutshots and overall the strategy is profitable. If Dutch calls 40% of the time, it's profitable to shove with double backdoor draws, but probably the worst of flops are unprofitable to shove.

Overall, I think if Dutch's range preflop is wide enough chefward could probably make the call profitable if he was very calculated about which flops he shoved. I also think that if many of the calling hands would be a very difficult call to make, we can justify the call.

Since we are calling 6400, the pot will be 23400 on the flop. The effective stack signified by the shorter stack (in this case Dutch) is 24387 behind. So if you are to shove the flop, you are putting in BOTH the 6400 to call PLUS 24387 or a total of 30787 to add 17000 to your stack from this stage in the decision tree. If you shove and are drawing DEAD when called, you have to get folds nearly 2/3rds of the time. But as we know from the analysis we just did, you are rarely drawing dead, and probably can avoid shoving the worst of the flops.

In Conclusion, chefward almost certainly cannot make 8000 worth of profits which is what's needed to make Dutch's decision unprofitable on his own, but it seems he can probably make a selective stop and go profitable to fight for a small share of the dead money AT the cost of great volatility to his own chipstack, that given ICM factors may not be worth the chips he gains. After working the math I can conclude that my intuition that calling CAN be marginally profitable for chefward2 is correct, but the margins are very thin and seem more thin than I would have thought, and given ICM impact that chips you gain are worth less than the chips you lose, it may be better for chefward to fold anyways. It's one of those really razor thin decisions where either one could be correct. Certainly it's not terrible to gamble a little bit to potentially knock out a talented player, particularly if there's enough dead money that it's positive chip EV and go down swinging at who you perceive to be the biggest threat at the table to you taking the title.

But you can also argue that there's no reason to fight so hard in a go big or go home situation when you have almost 40 big blinds and there are better spots.

With that being said, I'm pretty confident it's possible to get more precise about which flops you can shove with given a lot of in depth analysis of ranges. I very much doubt that chefward has done the kind of work to get there... but who knows, maybe he has. Some flops will be hard enough for your opponent to call, or will miss opponent that even if he does that it will be profitable to bluff shove a lot more, where other spots you will have a decent draw and still  have to fold because opponent's range crushes yours. Like if you had T9 and flop was KQJ, a Ten brings you a straight but probably is almost never good when you get called and hit.

I also still think that you could milk a little bit more out of calling preflop if you check the flops you hit hard enough to be a favorite vs calling range(12+ outs or 2pair+), you will get paid off more, and the result is that you probably will not have to semibluff shove quite as light because of the added equity you get most of the time from checking. You also may be able to get your money in with one pair hands this way and end up ahead if the flop is exactly right (like all 3 flop cards below ten or 3 uncoordinated cards and only one of them being a non ace face card).

I think Dutch probably can find a call 55% of the time chefward shoves. If he starts unpaired, he'll pair up 1/3rd of the time, plus he'll start with a premium pocket pair often enough and hit a big draw himself often enough or have overcards like AK and a chance at having the best hand often enough given the pot odds to make a call. I think it might be closer to 50% in Dutch's case and closer to 60% in the typical opponent's case who doesn't punish the limpers as liberally as they should, but I think 55% is a fair, conservative assumption that chefward should be comfortable making.

Even though it's unprofitable to shove double backdoor draws... if we don't have these in our range we lose too much from check/folding too often such that it is more unprofitable if we don't include them than if we do. So we have to include all gutshots too. The ideal solution to make the most profit is going to only include some of the double backdoor draws, and probably include some more one pair hands without double backdoor draws when the flop misses opponent's range, but I am not going to try to even come close to solving this since I already took some liberties in approximating my assumptions and it would be a nightmare to try to solve every scenario.

So I think if you were VERY VERY precise you could make a selective stop and go play in this spot profitable to the point where it might even be worth it after factoring in ICM, but that would require virtually a photographic memory and a lot of hours of work that are best suited working on core concepts that can help develop better "feel" instead of memorizing precise decisions for exact situations that will almost never arise again.

Nevertheless, it's a very interesting topic for consideration. It would be interesting to look at the stop and go with suited connectors in the same spot when we have a lot more chips behind and thus have better implied odds to hit the better flops. It also would be interesting to compare it where you are even shorter stacked such that you can jam any pair and you get more out of your semibluffs due to not laying as big of a price vs hands that can call you. If we could get a few spots like that, we could develop a much better "feel" for making moves like this profitably and then repeat that with a ragged ace or two offsuited broadway cards or low and medium pairs so that overall we have a good "feel" for when to make a more flop texture based stop and go with around 10 big blinds, 15 big blinds, 20, 25 and the check raise shove variety of the blind defend vs someone who always Cbets as well as vs a more selective Cbetter that Cbets 60% of the time.  If you never limp, you don't probably need to concern yourself too much with the stop and go to defend against a potential dead money grab, but there may still be some spots to use it from the small or big blind.