|
|
本帖最后由 吹牛无罪 于 2020-2-4 13:15 编辑
这位朋友所说的跟我看到的事实不符合。校长原文花了极大篇幅讲GTO的泛滥给业界带来的坏处：
3) The harm this GTO has done to the growth of poker
I would not have written this post for just pure discussion of a game model. What has bothered me is its damaging to the growth of poker.
A) The power of the GTO theory/tools are extremely exaggerated
“GTO solver”,”GTO trainer” are really hot these days. If you don’t use one of these tools in your gameplay discussion, you are automatically classified as an old school guy. And nobody bothers to listen to what you have to say.
The poker population has been brainwashed to such an extent that no many seems to care that none of these solver/trainer is a true GTO solver/trainer.
Quite often you see someone claim his software is the closest one to GTO. That may be true, just like someone says Mercury is the closest planet to the Sun. And we all know, Mercury is not Sun. There are some simplified situations these solvers can be used. Beyond that, which are the vast majority of cases/situations, it quickly becomes useless.
Don’t blame the engineers for these tools. A GTO solver demands so much resources, the algorithm used in these software have to have a lot of shortcuts to make the tools “working” possible.
B) Mislead players to believe GTO is the way the game to be played in the future, if not already.
There is no doubt that AI will beat human players very soon. But that does not mean the game will be played the way AI does.
Deep Blue beat the best human chess player in 1997, twenty three years later, human beings still play the game the human way.
In 2016, Google’s Alphago shocked both the AI world and the Go world by defeating top human Go player 4:1, and followed up with a 60:0 sweep against top pro Go players. Just a year ago, it was common belief that it would need at least another decade for AI to beat human pro player due to the game’s complexity.
Yet, human Go players still play the game the human way. In fact, the team AlphaGo acknowledged that AlphaGo is like a blackbox to them.
If human Chess/Go players can not be trained to play like AI, what makes anyone believe poker can be played like AI by human player? Unlike chess and Go, where the goal is simple (to win), and path is relatively clear (tree search) with perfect information, Poker is much harder for humans to learn to play like AI. There is really no such thing as human GTO players, not until human beings can implant powerful computer chips in their brain. Human brains are just not fit to that kind of storing and processing information.
It is kind of like, although we know cars run faster and more efficiently than humans, marathon players just can not try to move faster by rolling on the ground. Humans are not built that way.
Or, we can think the easier way: poker as we know it will die long before human beings can play like AI should that eventually happen.
Button line is, GTO was not, is not, and will never be the “right way” poker is played.
C) Its learning curve is bad for the players, producing mediocre players in volume
Beside model limitation and algorithm “shortcuts”, there is another issue for the players to use any modeling tools: they have to understand the model/algorithm/parameters they use. Due to lack of GTO understanding, starting from its definition, it is not surprising that not many players know what they are doing with these tools.
When players get a traditional coach and not happy about their progress, they blame the coaches for not being good enough. When players try the GTO way and feel stuck in the learning curve, they blame themselves for not trying harder.
The reality is, they quickly reach the bottleneck with little hope to get through. Combined with the reluctance to try an alternative way, they are bound to become mediocre players, except for a few really talented players.
D) The GTO superiority makes games unenjoyable
This mentality produces a volume of arrogant GTO players.The arrogance of some GTO players can really make a game unenjoyable.
Many players follow the GTO/software religiously. I have seen many times some 5/10rmb(less than $½) commentators criticize those high roller players with confidence. Where does this confidence come from? The “GTO solver”. It is ironic when some of the high rollers keep talking about how those solvers help to improve their game, only to see some lowest level players use the same software and criticize them like “it is well known that live high roller’s playing strategy deviates significantly from GTO strategy”.
You can not blame them. They believe what you want them to believe. They do what you want them to do. It is just natural when your followers find out you are not playing the ultimate strategy, they get worried. They consider their critique is from the truth generating machine, not from them playing at the lowest level.
And when this kind of players talk at a cash table with no big name players presented, it can be super annoying.
E) GTO scares off players
I have heard numerous times players, including some established players, saying: I am just not a math guy. What they mean is, they kind of give up on improving their games. I always answer with something like: poker does not need a lot of math. Observation, reasoning, execution are all ahead of math.
Still, they feel they are at an inherent disadvantage to their opponent, although it is almost never a key factor for most players in most situations.
So, what we really have about this GTO thing for so many years is the brainwashing the whole poker world with a deified learning system, which was based on a pseudo algorithm of a model with serious limitations for the real world games, and the theory was never fully presented to the mass, not even the definition of the theory.
As a result, it is not surprising that this systems produce a large number of arrogant mediocre players which make the game worse on all aspects. And it also turns away potential players with the fake powers that is believed the non believers can not possess, which put them into inferior group with no way out.
并且校长并没有反对大家用软件。他甚至把“促使大家用软件“列为GTO不多的好处之一：
B) GTO helps to get players attention to use computer tools
In this modern era, a lot of tools are out there to help players improve their games. GTO certainly help to gain the attention from players to try these tools as a by product, although many of them can not even tell if a software should be classified as a GTO software.
校长这些年有没有用solver学习过一个小时，我不知道。从这些年的战绩看，似乎也没太大必要。
他对目前的高额无限德州发展成什么样有了解吗？我也不知道。但是我保守估计一下，目前的高额无限德州发展成啥样子，对99.99%的牌手来说都是看个热闹。对校长这样层级的牌手也许有意义，对我对您一定毫无意义。
校长原帖，您也说了关键点都对，这就干货不少了吧。思考观察+用solvers会比思考观察更差吗？有可能。给八岁孩子一辆坦克去敲核桃剥核桃仁吃，后果一定比让他自己找砖头更糟糕。
我没看到校长对反对他的人有啥”必须牛逼且业界有名气“的要求。校长只是希望大家至少认真看完他的主帖。这一点您似乎没做到。至于”不能有任何利益相关（比如卖软件，或者做教学）“，这要求不过分吧？其实就算谁有利益相关，只要能就校长文章的内容发表回应，只要说的有理，公道自在人心。可惜，目前为止，只发现校长词儿用的真准确：”Many players follow the GTO/software religiously“。
|
|
