Minimum defense frequency is an aspect of 'balanced play' that I am trying to study and figure out.

In my last thread 'balance' RobertUTK, ARtyMcfly and thegodson brought up some things about MDF:

1. that MDF is most relevant on the river, optimal play converges on the river (no more streets of action/variables to consider)

2. that MDF can be potentially harmful when misapplied

I want to focus on this second point and pose a question to the forum: does MDF have "practical utility" to a player that wants to incorporate balance into their strategy or at least understand how to implement a balanced-strat and deviate accordingly? Is it a concept which should be loosely adhered to or avoided altogether in games where balance is not relevant?

I read a simple analogy in this article that I think is great: MDF is the shield, pot odds are the sword: https://upswingpoker.com/analyze-poker-hand-techniques/

I want to add this:

There seems to be value in understanding how to utilize a balanced/optimal strategy, even if you play stakes/games where this kind of strategy is unnecessary (where I do). Understanding what balance is and how to create it enables a player to recognize and make adjustments effectively. This is my impression at least.

There is a big difference between understanding these concepts on paper and using them in real life, though. I want to figure out if MDF is something that is even worth thinking about/using when playing in real life.

I know this language of GTO, exploitative, balanced-merged-polarized-capped etc is worn out in modern poker theory.

A few things come to mind here:

a) the river is the street that allows the most room for exploitive adjustments; closing the action heads up on the river leaves zero room for counter exploitation by the opponent.

b) deviation from the Nash equilibrium strategy on an early street leads to non equilibrium ranges on later streets; using mdf while playing with a non equilibrium river range will result in either making -ev calls or missing +ev calls on the river.

I gotta go do some stuff but I'll be back in a bit to ramble about the effects of these facts on mdf.

As Bob alluded to in another thread, balance arises out of playing "good" ranges. It's not the be-all-and-end-all. i.e. The aim of the game is not to be "balanced" and "unexploitable"; the aim of the game is to play your range in a way that maximises EV.

It's been established that a certain amount of balance increases your profits (it also makes you a tough player to beat), but I don't think you should focus on MDF numbers.
The basic concepts are quite trivial though and it basically boils down to this:
If you face a big bet (which creates worse pot odds), less of your range will be profitable to continue with, so you should fold more often. If you face a small bet, you can stick around with more hands, because better pot odds mean you don't need to win so often to break even.
You just have to learn that some boards and situations give the other player a range advantage, and on those you might have to "overfold" (i.e. fold more often than the MDF number), while the converse is also true. If your range has the advantage, you can "overbluff", expecting villain to fold a lot. If you analyse millions of hands, then your overall betting/folding frequencies might end up close to the MDF numbers, but in the vacuum of one hand the numbers can be radically different to what the MDF math says.
e.g. If you defend in the BB vs UTG and the flop comes AK7r, you'd have to fold to a 2/3 pot bet very often, as only a small proportion of your range will be +EV against the range that villain is repping. (You'd basically have to fold all your underpairs, and just continue with Ax, Kx, 7x and Broadway draws).
If the flop comes 874 instead, villain can't have many strong hands, while you do, so you'll continue with them not because MDF says so, but because so many of your hands have a positive expectation vs villain's range.
By the time the river comes, ranges will be narrower (you'll have folded your air, and villain will give up with some of his bluffs, for example) and at that point, the continuance frequencies will usually be much closer to the MDF numbers*. That said, sometimes the river card is so "ugly" for one player's range that he still has to check back or check-fold at a very high frequency.

* With HUD stats and post-game reviews, you can spot opponents that still "bet too often" or "don't bluff enough" on the river, by looking at their river frequencies. At low stakes in particular, you'll find that most players are extremely value-heavy when they bet the river, so bluff-catching according to the MDF number would be suicidal. For beginners, I think it's better to just look at the pot odds and ask "Am I winning here often enough to break even by calling?". At 2NL or 5NL, you definitely shouldn't be thinking "Oh, villain has potted the river. I need to call 50% of the time to prevent him exploiting me by over-bluffing". At microstakes, hardly anyone bluffs anywhere near as often as theory suggests. Their big bets represent big hands, so you shouldn't call down light.

^^^ This is exactly what i wanted to know, thanks for taking the time to break this down.

There is obviously huge value in these concepts about 'optimal' play, but sometimes I feel like the way the poker world approaches "GTO" ideas (including myself) is misguided.

To make a comparison it reminds me of the weightlifting industry where there is a heavy emphasis on advanced periodization/programs, many of which are mostly relevant to people who have been lifting for years. Novice lifters become engrossed in the programs, but all they really need to be doing is putting in quality sessions consistently. These people may benefit from advanced programming and the ideas inside, but ultimately they are trying to implement concepts which are unnecessary and even harmful at their level.


But dont get me wrong I love the theory/math of poker, I just think its interesting how GTO has exploded and affected poker as a whole.

https://forumserver.twoplustwo.com/1...22/?highlight=

Here's a thread where I explore some asymmetric (0,1) game river situations for fun.

Note that when the bottom of the in position players range never wins a showdown, strict minimum defense frequency works on the river.

I was always skeptical of the value of the extreme precision Applications/pre solver "GTO" range modelling but the concept of a defense frequency and a funneling of ranges helped me ask the question "if I'm not folding this hand, what would I fold?" and visa-versa.

Hands in the bottom/top few percentile of hands you can show up with on the turn for example must surely be correct to fold/continue in theory with weight given to bet sizing.

Yes.

It is important in all games.

I tried this thingy with an EP opener with the range 77+, suited broadway, AKo-AQo, A5s-A2s. The button player calls with QQ-55, AQs-ATs. Flop comes Q72r. In this situation the caller defends to the river with roughly QQ, 77, and AQs assuming turn and river cards are bricks. Opener's only value hand is then QQ (3 combos). Bluff hands could include KJs and JTs, but only with the backdoor flush draw (6 combos). BU calls with QQ-99, AQs, and 3 combos of 88.

Turn is a 2 rainbow. EP bets 3/4 with QQ (3 combos) and KJs (3 combos). BU calls with QQ, 77, AQs, JJ and 2 combos of TT.

River is a 3. EP bets 3/4 with QQ (3 combos) and 1 combo of KJs. BU calls with QQ, 77, and AQs.


This model did not account for the opponent raising with strong hands or blockers. It seems kind of weird to me that you wouldn't bet 3 streets with 77 since it is about 50% equity versus the final calling range. I guess you would go 2 streets with it according to the article.

If you end up checking the flop, I suppose the situation would then be reversed. Your opponent starts betting with a range where at the river you end up calling 3 streets with 77, KK+, AQs, and some combos of AQo. The value hands for BU player would then be QQ and 77 with the appropriate number of bluffs.

If BU player ends up checking then perhaps you can go to value town with AQs. The article doesn't go into detail with partial value betting. It says it checks back the flop with A9 when it is 56%ish, but it doesn't discuss value betting with a hand that is less than 50% to the river, but more than 50% for flop and turn calls. I would think betting flop and checking turn would be an acceptable play with AQ on this board run out.

Seems like continuation betting is at a very small frequency. Only 8% of the time for EP player. Seems quite small, but could be correct due to being OOP. IP player bets 34% of the time if checked to. Personally, I find myself continuation betting much more frequently.

My brain breaks as soon as the BTN flats queens pre. :/

It's almost certainly mixed as a flat vs. that UTG rfi in the pseudo gto universe, and likely at a reasonable frequency (25%+) if not even higher, considering the range he is giving UTG is like 10%....

I respectfully disagree with the premise here.

MDF is what the game is about. Why dont you only play As pre? Because if you do so, you don't defend the pot enough. Poker exists in the first place because you have to defend the pot a minimum.

Now to answer the initial question, what is the practical utility of MDF? This simply allows you to play against loose aggressive players. A good loose aggressive player will bet any two cards in some spots knowing you will fold too much in those spots, making betting any two cards profitable.

For example, if a player call on the BB only with AK/T+, you can raise any two cards on the BTN for example, and it will be profitable. (Well I think it is, havent made the maths of that on typical game structures.) So the BB players need to have a bigger MDF.

Another example. Flops come Q23. If you fold every time you dont have a Q in your hand (or better) AND if you dont have a Q in your hand (or better) some significant amount of the time, a loose aggressive player could bet any two cards profitably.

MDF isnt practical against a tight nit player though who bets only for value (that's the harmful part of it). On the other side, you probably do use MDF already against regular low limits fishes who bluff all the time. If the flops come Q23, you will probably continue till the showdown against a "fish" with your pocket Js if nothing too bad happens on the turn and river. When you do that, you are using practically MDF.

That being said, as I mentioned in some of your other threads, I think what your looking for to understand balanced play is this: try to beat the players who bets only with strong value. Because when you do so, you understand why playing only with strong value is far from optimal, then you understand the premise of balanced play. Adding a balanced MDF to your strategy isnt gonna help you beat those players. Exploiting their MDF is. Then to not get exploited yourself, you will start having a balanced play and decent MDF.