Exploitation and the FTC Theorem

Progression through the micros naturally trends us toward unbalanced lines. We start playing basic TAGG, then learn to steal blinds more, to cbet more. Hey, it works. When we face better competition at 100nl or higher, the lack of balance is very exploitable. Trust me on this. I lost 237 buy-ins with my unbalanced lines until I realized what was happening.

Example: Resteals

Villain steals with 40% of his hands from the BTN with a 3.5 BB raise, but when facing a 3bet from the BB, he’s only willing to continue with 8% of his hands. He’ll 4bet 4% (AK, QQ+, KQs, ATs+) and call with 4% (JJ-88, AJo, AQo, KQo). Let’s make a ratio of fold to continue frequencies: Villain continues 1 out of 5 times, so his FTC ratio is 4 to 1.

Our BB Hero faces a steal with 5 BB’s in the pot, so he’ll win those chips 80% of the time he 3bets. Suppose that Hero 3bets to 11 BB (puts in 10 BB’s + his blind), and refuses to commit any more chips when villain continues. The math is assuredly in Hero’s favor: 4 out of 5 times he wins 5 BB, and 1 out of 5 he loses 10 BB. That play has EV = 2 BB / hand for Hero. Stealing with 40% and only continuing with 8% is extremely unbalanced.

I’m not saying you shouldn’t be stealing a ton on the button. You should. When your opponent learns to 3bet light, however, you have to adjust in one of two ways. You have to fix your fold-to-continue ratio, so your options are to steal less or continue more often. Typically, we do both, 4betting a bit lighter, calling a bit wider and narrowing our stealing range.

FTC Theorem

A player with a high FTC ratio is easily exploited with bets and raises.

Turns out, most common lines have a fairly narrow band of FTC ratios that are not exploitable. For 3bets and most flop bets, if we have a low FTC ratio, say 1 to 3, our range is too tight. If it’s high, say 3 to 2, our range is too bluffy. Optimal FTC ratio bands depend on the dead money in the pot, the bet size, fold equity based on villain’s range and what percentage of pots we can contest successfully when called. But we can make some simplifications to get us close enough for poker usefulness. For example, flop check/raises have an optiomally balanced band between FTC ratios of 1:1 and 1:2. Find an opponent whose band is outside it and profit.

I climbed this mountain of ratios, ranges and calculations only to realize IowaSkinsFan had gotten to much some peak without all the math. In some ways, my FTC theorem just restates the ISF theorem. But there are a few original morsels ISF can’t claim to have written down before I got to them. The only problem is that they pretty much all fit into Renton’s ABCD theorem. So I am reluctant to post my FTC work like it’s brand spanking new. I can only claim it’s a way of looking at poker I developed that has helped me find exploitable weaknesses in agro-fest 6max opponents. I’ve plugged some leaks. I have developed coherent adjustments. Maybe it will help you too.

Finding Exploitable Weaknesses in our Opponents

To understand FTC ratios without all the math, just start asking yourself this question when facing a bet or raise: will this villain fold about half his range to a raise? If the answer is yes, you’re generally well-advised to attack with a wider-than-value range. Lighten it up. In our simplified resteal example, Hero should keep a wide (light) 3betting range until Villain’s FTC is below 2 to 1. At that point, Hero has to start winning some postflop contests to keep the 3betting profitable. Probably it still is, but some analysis and adjustments are required.

You’re at the table looking at a player whose steal stats are through the roof, but you know he doesn’t continue all that often. Put him on a range. What will definitely call with? What will he raise with? Do some work away from the table. You will find that most TAGG players at 50nl and below can’t steal more than about 25% of the time without losing their balance. As soon as they get above a 30%, I’m coming at ’em. They’re just stacking free chips in the middle, and when they jack up the steals to 40% or 45%? Nirvana.

The interesting game play tidbit I discovered is that you don’t have win all that many postflop battles to balance things out. Say you 3bet a serial stealer with KJo, and you get flat called. The flop is Q43 with a flush draw you don’t have, so you check and get bet into. What’s the problem? Fold! You’ll get at a Kxx or Jxx flop about 1/3 of the time, and out of position you can just wait for it. You can stab at some Axx flops and other textures, depending on your villain and his range, but you don’t have to chase every pot to keep even your very bluffiest 3bets profitable. You can check/fold half the time, play carefully the rest, and really profit from wide weak range of your opponents, much of which they’ll just fold.

What are some other ways to exploit others and fix your own lines?

Check Raise the Flop

Being out of position sucks. As soon as you graduate from the micros to small stakes, you’ll start running into what I call the "auto-floaters." Everyone knows about the cbet. So now they call lots of cbets. And since most folks cbet a ton, any preflop raiser who checks the flop must have absolute dog poop for a hand, right? So you find a smaller but still sizable group of low stakes players who stab in a flop bet most of the time when checked to. In the games I play in, cbetting air sucks since cbets have lost a good bit of their fold equity, but consider an alternative: check a lot of air and medium strength hands, then look for spots to raise the agro-donk who splashes chips in the middle on almost every flop where’s he checked to by the preflop raiser.

That agro-donk is surely easy to find. I know because I was the agro-donk, and all the TAGG-regs I played figured me out pretty quick. But I learned fast. I only lost 97 buy-ins before I realized what was going on.

Think about the range of guy who flats your preflop raise in late position and then bets out when you check the flop. He’s much more likely to have to 44 than KK, and much more likely to KT than AK. So unless he’s just been offered sexual favors by the flop gods and hit two pair or better, he’s about as excited about your check raise as he would be about a kick in the groin. He can continue on the 3.5% of flops where the sexual favors occur, and if he bets draws aggressively can manage another 6-10% of flops – maybe. But his range is pitiful. If I weren’t making so much money off it, I might even feel sorry for him. But then I remember the 97 buy-ins my dumb ass spewed before I learned, and I quit feeling sorry pretty damn fast.

The great thing about this particular FTC theorem money-maker is that at 50nl and below, most players react to check raises like they would to being robbed at gun point. The cough up the money unless they’re bulletproof. So their perception that "check raises mean business" means it takes them a lot longer that it should to adjust to our extreme bluff-heavy check raise line. We also gain the advantage of being able to check more flops, which leads to more delayed cbets and strengthens our check/call line as well. We have to be careful with board textures, but most small stakes players should check a lot more flops out of position.

"Seriously?!" you’re thinking. "I’m going to check raise air?" Sure you are. But not just any time. Then you’d be like me when I discovered this profit center and proceeded to lose 41 buy-ins while I fine-tuned everything. You, grasshopper, you will instead pick extremely aggressive opponents who habitually raise in this spot and boards with the right texture. What’s the right texture? That’s a whole other strategy post. Mainly, just put your well-chosen villain on a range and attack flops that completely miss it. You’ll take a bit care when picking your spots. And then you’ll profit.

But think about this: a balanced check raise FTC ratio is between 1 to 1 and 1 to 2. Hero needs to have somewhere between a third to half his check raises be bluffy just to keep things in balance. If you do some homework, you’ll find most TAGG-regs at small stakes play tighter than the 1:2 ratio, some of them way tighter. You’ve got to add some bluffs in there, something to boost it up near 50% bluffy, or you’re leaving chips on the table. And you can play a 2 to 1 FTC ratio against most villains since their flop bet FTC is so vulnerable. Until they adjust. You gotta be aware when you adjust or it’ll be like the time I lost — oh, well, you get the picture.

Flop Raises in Position

The same math applies to Hero in LP who flats a preflop raise and then faces a cbet. When Villain is cbetting 75% of the time, or more, Hero has the opportunity to raise with a wide range of weak hands and steal a nice pot. This is a different animal than check raising, however, since Villain’s range is very different and includes the premium hands. Generally, I’m looking for a preflop raise percentage of 20% or more, but you find that’s true for lots of 6max villains in MP or the CO. When a serial cbettor has a wide preflop range, he’s got to continue with total bluffs or surrender too often.

Be careful here, however. Most 6max TAGG’s UTG range is around 12% and can sustain a 70% cbet rate or higher without serious vulnerabilities. Make sure the preflop range is wide enough. And board texture is vital. This is one of those once-an-hour type of actions that applies to maybe 5% of the players I face, and I have to wait for the right time and board texture to attack.

Theory and Ranges

I will give some recommendations in a moment, but first I have to warn you about a couple of things.

1. We don’t need balance against the unaware.

If no one’s restealing, steal away. Steal with 50% of your hands if you’d like. If you’re getting plenty of folds to your cbets, cbet like a monkey. Don’t worry about FTC balance until your opponents force you to.

2. We don’t play ratios, we play ranges.

Each time we look at an FTC ratio, we have take time to form ranges for raising and calling, and analyze how those ranges will play against villain’s likely range. The whole shania deal comes into play, too, as changes in each line alters the expected value of other lines.

3. Hurt but don’t alert.

I generally don’t steal much more than 35 – 40% even against complete nits. My motto is "hurt but don’t alert." Even the most ridiculously unaware players notice if you steal 5 or 6 times in a row. As soon as they notice they (might) adjust. I try to get the maximum profit from my opponents’ lack of balance while keeping my opponent blissfully unaware of the need to change it up. I do check raise a good bit more than average, and raise a good bit of flops from late position. Still, I try to rape and pillage in only the most juicy spots so I don’t wake up too many sleeping donators.

* * * *

Let’s work through some of the math and ranges for the flop examples. Suppose we have a villain who open raises 20% of his hands in the CO (22+, Axs, A9+, all Broadways except QTo and JTo, 65s+). The flop comes Q 8 3 rainbow. He cbets. You check, find you have 1.5k hand histories and that he’s cbet 80% of the time. What can he continue with? 20% of the 1326 starting combos is ~265 combos. If you raise, he’s continuing with what? One key feature of this is the unlikely event he hit two pair given his range. He’s obviously continuing with sets (9 combos), AQ (12 combos), KK+ (12 combos) and possibly 99-JJ (18 combos). He might continue with A8 or A3 (3 combos each), playing it like a 5-out draw. That’s a maximum of 58 combos out of 265.

If we’ve picked our mark successfully, and size our bet properly, his FTC ratio is almost 4 to 1. Say the pot after preflop is 8.5 BB, and he bets 6 BB. A raise of 15 BB wins 14.5 BB almost 80% of the time.

I’ve got to stop here and mention the problem. Usually we don’t have a great number of hand histories. Folks who haven’t watched it happen underestimate the impact a heater can have on your postflop stats. They can get way off fst. Folks don’t play all that many postflop pots, so even if we have 200 hands on a player who picked up QQ+ 12 times during those 200 hands is going to have some very big errors. So be careful putting too much weight on an 80% cbet stat until you have several hundred HH’s.

A better method for estimating ranges is watching showdowns after certain actions. If I find a certain villain whose stats make him look like a target, I focus on that table trying to get a showdown read. How does he react to flop raises on certain boards? Will he shove a draw? If so, how many outs? Will he continue with 2nd pair ace kicker? How about TPNK? If his showdowns are TP2K hands or better when continuing and I feel his preflop and cbet stats are pretty accurate and wide, I know I can raise him pretty light and make tons of profit on his folds.

Some final words of caution. If you’re 3betting light, or raising the flop light, you have to be able to fold to pressure. The worst pressure is the monster hand that just flats you, because you’re tempted to think: "He’s weak. One more big bet and I’ve got this pot." That’s usually not true. If someone continues after you raise the flop with a weak draw and not much else, shut it down. You’ll make enough to money to cover some failures, but not if you get sucked into huge pots with hopeless hands. If you don’t have the discipline to put in 20 BB of your stack and then fold, then don’t try raising the flop light. You’ll be the one with the "free chips and beer money" sign hanging above the pot.