I'm trying to work on ranges to 3bet the villains, but I'm in a tough situation that I don't know where I'm making a mistake. I know that in river spots where the bettor has a polarized range and the caller only has a range of bluff catchers you can use gto frequencies, if you can't exploit or don't know villain's tendecies. I'm trying to use the formula : alfa = s/(s+1) or alfa = risking / (risking + gain) in a preflop spots (I don't know if I can but I think so) to know ranges. (Ok this concept of gto is probably wrong I read now that it isn't gto.)
Example: Blinds are 150/300 and 150 antes on the table. A villain with 7271 chips open from HJ a raise of 624 chips. The hero has 3492 chips in his hand and 300 on the table because he is on BB.
POT IS : 624 + 300 + 150 + 150 = 1224
The hero decided to shove 3492 chips so : alfa = 3492/(1224+3492) = 0,74
so the villain calls 1-0,74 of the time = 26% for me don't make autoprofit.
This villain is oppening 14 % of hands from HJ I chose: 55+, A7s+, KTs+, QJs, A9o+, KJo+. IF he calls only 26% of the time I put him nearly calling with : 99+,AK (3,92%). But when I put it on icmizer, chip ev. The answer is shove any2 cards will be profitable, I know it seems obvious because he will call too nity but what is the mistake on the calculation, Why is it not applicable ?

I think I got some idea. The odds of the villain is ODDS : 3168/(3168+4716) = 40 % , so if I shove any2 his range to call will be more than 3,9% to have 40 % of equity so he could open his calling range to adapt his self on the situation. So I have to build a range that give him this odds based on his 14% PRF and it will not be all my range. is it ?

Confirmed: MDF isn't the same as GTO.

is this minimum defense fold true only when the villain has enough bluff to balance his value ranges, if he doesn't have enough bluff I shouldn't never call, because I will not have the odds to call if the villain doesn't have enough bluffs ?

why is not gto ?

You can click the button 86o detailed result to see why it is profitable.
Generally from what I understand your assumption is that if you are called you magically lose 100% of the time.
However in real world you have some equity if called preflop, so for this raise/ call resteal range setup from the screenshot we can push ATC.
You can have 0% if called if you are bluffing on the river however.

yes that is exactly what I thought that was the reason of the difference. I did it , CBluff/CValue = 0,74 , So I put in a program the range of the villain and I saw all cards that had equity bigger than 50% against his range and called by "Value Cards" and the result was like 12 cards so I calculated the amount of "bluffs" and the result was 8,88 so I took 9 cards with a equity less than 50% against his range of 14 % and called by bluff (It would be better If I did it with number of combos instead of number of cards) . But in my mind the difference is that my "value" hands doens't have 100 % chance to win and my "bluff" hands doens't have 0 % to win, thats why I think is the difference. so is not useful ? hehehe

Pre-flop, MDF is a pretty useless concept. If you're contemplating a 3-bet jam, then the EV math is all about pot odds and equity. If you jam with an optimal range, and villain calls with an optimal range, you both get a share of the dead money, and the bottom of each range will break even. If villain folds hands that would actually break even or make a profit vs your jamming range, he's giving away money, and you can exploit him by jamming even wider than GTO.
To put it another way, if you jam ATC on villain, he's not going to call to "prevent" you from jamming on him. He's going to call with hands that have enough equity to call profitably against your range.

I don't know but for me one thing depends on another, His calling range will depend on my jam range, and my jam range will depend on my fold equity, so it seems to be a circle I don't know how to explain. doesn't it exist an equilibrium ?

Yeah, it's quite circular.
An equilibrium arises when neither player can increase his EV by changing his strategy. This equilibrium might arise through a series of iterations.
e.g. You start by jamming ATC, because you know villain folds too often. You can print money with 32o from fold equity alone. When he sees you're shoving ATC, he realises he could profitably call with a wider range because he has enough equity to do so, so he starts calling more often. Villain will keep adding more and more hands to his calling range until they stop being profitable for him. When villain calls wider, you lose much of the EV you gained from fold equity. 32o stops making money if villain calls often enough. If his new calling range stops the bottom of your range being +EV shoves, you change your strat and tighten your 3b range. With a tighter shoving range, some of villain's calls start losing money (as they don't have enough equity), so he has to fold the bottom of his range. When his calling range tightens, you adjust your strategy again.
By repeating the process ad infinitum, you eventually settle on the perfect/optimal shoving range, and villain has the perfect calling range, where it makes no sense for either of you to deviate. That's the equilibrium, where you jam X% of hands because they make money from a mixture of fold equity and hand equity against villain's optimal calling range, and villain calls with Y% of hands that are profitable calls against your optimal shoving range.