Consider this game:

- 100 players.
- Each has a 1% chance of scoring 100, 1% chance of scoring 99, etc.

Each of these players has an average score of 50. But, what chance do you stand of winning, if you can consistently score 90?

Answer: Our odds of winning is something like 100,000-1.

Our strategy enables us to EASILY beat any one of the players. However, as the other players have a higher profit ceiling, we’re 1000 times less likely to win.

Show your math.

I didn’t do the math, someone on Reddit did lol… is it wrong?

Depends on the assumptions.

I’ll let you assume anything you like lol.. (I don’t speak math, not sure what you mean)

The guy on Reddit said that every player has to score less than 90. So, it’s 0.89 to the power of 100. (He actually wrote 0.89 next to a little 100, but my device doesn’t speak math either)

Anyway, he said it gives us a 0.000868961% chance of winning. Which sounds exactly right to me!

GTO in tournaments is not trying to maximize chips. In very simplified terms, GTO in tournaments attempts to get in the money as often as possible more so than try to win as often as possible since this averages out to higher ROI.

If the prize structure was very top heavy then GTO would shift to "shooting for first" a bit more and score worse on average, again, to increase ROI.

GTO is not a static strategy for every environment and it doesn't care if you score well very often or not, it cares about how much real world money you generate on average. The way this is currently modeled is with ICM

That’s really interesting, well written etc, but it’s completely beside the point lol.

I’m not sure if you agree with me or not???

You're saying GTO is really bad in tournaments because of a mathematical formula inherent to poker. That's dumb. GTO is whatever strategy maximises money using poker math.

Any valid point against using GTO on tournaments would have to do with human tendencies, not math inherent to the "base" game.

What I said before is not besides the point. You can score 1st less than someone else on average and still make more money than him in the long run. In fact, that's usually the norm in tourney math. Shooting for the highest likelihood of scoring 1st leads to getting eliminated too often in spots where you could've cashed out some money. This is why in the bubble you often can and should forfeit some chips (which hurts your chances of winning the entire thing) in order to jump prizes.
There is of course a middle ground though, and that exact point is calculated with solvers and ICM calculators

You heard it here first people. GTO is whatever makes the most money.

So, why do people exploit?

I guess I need to explain it more clearly for you.

GTO is whatever makes the most money outside of opponent mistakes. That's why in my previous comment I said there are valid points against GTO that have to do with human tendencies. English is not your first language I guess.

People exploit to make the most money possible in real life.

You used a mathematical model (whether it's correctly calculated or not is another issue) that doesn't take into account human tendencies to prove GTO is bad. That's as stupid as it gets.

Refusing to exploit your opponents is silly and inefficient. That doesn't make your point good or smart.

So, in real life, exploiting potentially makes more money than GTO.

In a real life field of players, each of the players can make way more money than the GTO player.

In a large field of players, in real life, the GTO players lower profit ceiling obliterates his chances of finishing in the top flight.

I recommend you re-read what I said until you understand it.
Also, learn a bit about ICM https://blog.gtowizard.com/icm-basics/

I know all about ICM. It relates to the prize distribution and how to adapt your strategy to maximise your profit from it.

You clearly don’t realise that exploitative players can cater for that too.

I’ve been nice. I’ve used clear and perfectly accurate math. You’ve been very rude, you used jumbled up, aggressive arguments that make absolutely no sense.

Man, don't even try to talk to yadula, its pointless.

You know what’s funny. You just said:

“don’t even try to talk to Hands of God, pointless talking to that guy about Poker”

That’s what my name means. I was born for this s***. But, anyway, the theory. The basic math. That proves you shouldn’t use GTO in tourneys. It’s what this thread was about… up at the top there…

I don’t even play tourneys. I was just saying it for you. If you want to lose then keep going with your GTO. Maybe you’ll beat the math. Good luck with that.

All players could individually, but the majority won't. The winners need to be exploiting somebody.

Nice, but you have it backwards.

Each of the exploitative players do exploit each other. In a large field, it’s easy to see that whoever wins that battle will, on average, have a far bigger stack than the GTO bot.

The GTO bot can always beat 1 exploitative player, on average. In my first example, our player represented the bot and he had a average score of 90, far bigger than the other players who each individually had only an average score of 50. I gave him a huge advantage over any one of the other players.

It’s only combined when they beat GTO. Some of them must fall, but it’s their sacrifice that’s our salvation.

Here’s some more points..

- The height of the profit ceilings for both the bot and the exploitative players will rise as the exploitative players play further from GTO. The bots will rise slower, which will decrease the bots chances of winning.

- The number of players obviously has a massive effect on the chances for the GTO bots chances. The more players, the less chance of finishing the the top tier.

These two figures can be combined and we can say: the overall distance from equilibrium of all the players combined, constantly minimises the bots chances of finishing in the highest tier.

There are other factors…

- Higher variance strategies than GTO will increase the bots demise.

- Lower variance strategies than GTO will reverse the bots demise. I see no reason why the bot couldn’t become favourite against a fair number of exploitative players through the passivity of their exploitative plays.

I think that’s exactly right. I’m a bit tired, and I’ve been given a lot of abuse today.

It always amuses me that anti theory/gto types always assume theory based players don’t exploit.

At best they don’t deviate as far away from balance with too little info. Or their exploits may not be as wide which there’s many reasons for.


I’m not sure where all these people who stick as close to equilibrium as possible are. I’ve never found or met one. Just heard about them from anti theory folks.



What is definitely true is that a well studied theorist will be able to articulate why/what he deviated from balance far better than non studied. And that usually makes the studied player sound like he’s “playing GTO.”


Also, sometimes when your opponent is playing so far away from balance….you can make plenty without taking the risk of deviating from balance until you have more info.

For example, if someone is overfolding to cbets, you’re already making more EV cbetting at a normal frequency since you’re supposed to get called or raised on some of those cbets.

But there’s no need to do much else since the opponent is doing the work for you. Obviously you could probably cbet wider for more EV…but likely don’t need to. As they might create some unwanted attention that gets V to become more aggressive.

This is just another thread where someone is stating the obvious……that a very good exploitative strategy can make more than an equilibrium strategy. Duh.

The question is, can you pick the correct exploits and avoid the wrong ones enough to make more EV than playing balanced?

Link the reddit thread

A GTO strategy will have 0 ROI in a tournament if all other players play GTO, and always positive if at least one player deviates (while playing a hand with the GTO bot).

And that’s why we have reddit.

I’m really surprised none of you get this. I used math for a change!! The effect I’ve described is undeniable and none of you have understood it yet.

I think didace might be close.

Yogurt Daddy… There are bots, this effect proves they won’t do well in tourneys. There are players like Polk, his army at upswing, and many similar groups too. And, there is an entire army of solver trained players who can barely exploit… You seem to be only level 2 yourself mate, as you don’t realise you can exploit unknowns by manipulating their perception of you. (Hustling is one of infinite examples).

But my most important point to you is this: if exploiting can make more money, and there are multiple exploitative players, the chance of one of them making more than the bot is increased as the number of exploitative players increases. This is undeniable. In tourneys this sane logic translates to finishing positions.

It’s yet another HUGE discovery I’ve made.

Tombos, not much was said on Reddit. The same kind of stuff as you guys said here. Nobody can accept that GTO might not bad in tourneys. On Twitter thousands of people saw the same post, 1 replied, a half famous player. He insulted me loads, I kept repeating myself, he kept insulting me.

Johnny, that is wrong, the math here proves it without any shadow of a doubt. It shows that a GTO player in a pool of only exploitative players will be guaranteed to finish in the top half (on average), but, he his chances of finishing in the top flight are tiny.

Wait wait, it’s apparent that I am the only person in the poker world who understands this effect. It seems I discovered it, so I get to name it again!

I dub thee: “unintentional collaboration”

Boring, I know, but it makes sense. People already realise that players collaborating can beat GTO, and what the exploitative players are doing here is accidentally collaborating against the bots.

Eg. “Due to the unintentional collaboration of all exploitative players, GTO don’t stand no chance in a large field tourney”