As of November 2021, the APT Forum is closed to new posts. Like with many online forums, usage has decreased in recent years. All previous posts are still available.
The Rake
Sign In to comment.
Welcome
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
Hi,
It is possible. We have it on a list of possible features. We never have considered it that important because win rates on our site aren't overly realistic anyway. However, I do see the benefit of learning to measure the effect of rake.
That might actually be a fun stat - winrate with and without rake.
Thanks for the post.
Multi-hand NL holdem is a very difficult game to program - this is one of the reasons why most of the top bots (e.g., U. of Alberta bots) are heads-up bots and play without rake. Every tweak you make makes it all the more difficult to simulate appropriate bot play, which is another reason we don't have rake in the game. Pretty much any book that talks poker theory and play ignores rake in most discussions because there are so many different types of rake and levels of rake that it is practically impossible to address all the differences. My personal belief is that if you get good enough to play outstanding poker without rake, you will be a good enough player that you can deal with rake and adjust to it, or get help from a human coach who can better adjust to it. So if we did ever add rake in, it would most likely be just a live cash-game style rake, and it would be an option, not something that people have to use.
As far as win rates not being overly realistic - what I meant is that our goal isn't to perfectly mimic a real game. There are no bots in existence outside of heads-up that can compete well with the best humans. Multi-player NL Hold'em is not a game that is even close to being solved. Good humans will beat computers (crush them) in multi-player games, which is why the win rates in our game can't be perfectly realistic. However, the purpose of our training is to allow humans to play a whole bunch of hands really fast - and at the same time learn odds, estimating what their opponents ranges might be, and have their weaknesses diagnosed (the weekly training plans). Used appropriately, our bots are good enough and they come close enough to simulating typical human players that they enable people to get better at the game of NL Holdem much, much faster than you otherwise could.
To me, Allen's point is the reason I love this site. The ability to play a bunch of hands in a short period of time helps me to identify my strengths and weaknesses and see them in the context of my overall game. I can get a high level overview of my game in much less time than would be possible, if it even is possible, by playing a real game.
Excellent response and much appreciated. Makes perfect sense to learn the theory and then adjust to distortions or conditions as applicable afterwards. I also appreciate the clarification about winrates. I read the initial comment to mean that the rates were pretty disconnected from real play and that didn't sound all that appealing. As long as they are good predictors in general, they don't need to perfectly mimic reality.
Thanks again. I am down to 2 choices for a training site (APT and PokerSnowie). I plan to make my decision today or tomorrow and this info helped a great deal.
Just a note on our site vs. snowie.
Snowie's bots are better than ours (much better). They play much closer to game-theory optimal poker (GTO). I personally think that is less realistic and tougher to adapt to a game vs. humans. But for learning poker theory and GTO, Snowie is better. Our game has many more options and the opponents are closer to humans. I think our game is more flexible and provides a better assessment of your game vs. human players. Plus, I think our site has some things that Snowie doesn't have anything similar to - like the Beat the Pro Challenges, Steve's hand reviews with Scott Blumstein, etc. It definitely comes down to what you are looking for. I've used Snowie and it definitely serves a solid purpose.
What a fantastic answer! Honest, self-reflective and informative - exactly the combination I want when looking for a place to train. We all have to recognize our strengths and weaknesses if we are ever going to improve, at poker or at anything else. As a result, I have just bought my 1 year membership and look forward to trying it out. To be fair, I am also joining Snowie. It wasn't a matter of costs but more about trying to get into a proper training regiment without getting conflicting advice and having limited time. So I'll give both a whirl for a bit and see whether it will be workable using 1, both or neither.
Thanks again for the comments and I look forward to starting to become a less horrible player
To 1warlock - any updates how do you combine Snowie and APT. I play live 2/5 and small/ medium tourneys(200 to 1000). I am looking for more training resources but I feel that majority of training sites geared towards online.
Hey @stumbras - I don't think I've really combined the 2 but I've used both for different purposes and feel that I am benefiting from both. Snowie is great for theory and I believe that is important. Once you have the theory down, its a matter of deviating to fit the particular field you are playing against. It makes it easier for me to spot the leaks in other players games and gives me the tools to exploit the heck out of them.
I haven't played those stakes (cash) live in a very long time but some of the people here who do have posted some good comments on how to adjust the setting here to replicate those games. I don't recall who off the top of my head but if you search the threads I think you will find them. @highfive @apt_gs and @dhirigoy seem pretty tied in to the forums so maybe they can help point to the proper threads or give you more help on this?
In general, I like using flopzilla and equilab to help practice my hand v range and range v range estimates. For tournaments having ICMizer (or something similar) is pretty much a must. Any practice you can get here with the MTT trainer is great, along with playing lots of SnG's for final table practice.
I'll invoke the names of @SteveBlay and @AllenBlay to draw their attention to your question. Since they know this place inside and out, I think they could give you the best insight on how to use the site for your purposes. If I think of anything else, I'll post another comment as well. Always good to keep looking wherever you can. Lots of poker forums out there with very talented people willing to help other players.
GL
+1 for rake
Got you @stumbras. It is definitely on our list.
I'd go a step farther and say that the APT advice is a TAG style, but it is not one that I would use directly. The advice is worthwhile for tough situations, but on an overall basis, the advisors can be exploited (because they are designed to play like typical humans). I don't think the direct strategy could beat most online games if you did exactly what the advisors say at every move - but I do think modified versions of it (meaning modified by human instinct when the computers clearly get it wrong - which they do) would be pretty good.