Unhappy with the software
I'm finding this website more and more useless. Although I like the features that allow me to choose the hands, the difficulty and the positions, I find the reports utterly useless. First, I don't believe I'm playing the same hands as the other players, except maybe in the case of the weekly challenge. Therefore, being told that I played better than 60% of the field is completely meaningless. Second, this piece of information would only be useful if I could find out what I did right and wrong compared to others. Knowing that I called/raised 5 times when I had a worse hand doesn't say squat to me because I could be in that situation and still win the hand. As an example, I recently had a run where I tripled my buy-in and got an 88 poker IQ although I had 0 case of folding better hands either PF or FTR. I had 11 cases however where I called or raised with a worse hand, which didn't prevent me from winning a majority of those hands, not because of luck but because I bullied my way to get bots' folds when they had a better hand.
So, I think the reports need a serious makeover, otherwise they don't provide any valuable information for us to improve. Food for thoughts...
Welcome
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
I think the software is more geared towards a GTO style. If you play LAG then the software will definitely score you lower. As far as playing the same hands... We are all playing the same hands since there are something like 169 hands lol. If you go to the reports there are ranges of stats that tell you if you are not raising preflop enough, calling to many 3 bets, or playing too many hands over a large sample size.
And as for winning and scoring 88, poker is a game where you can make a good call and loose but also make a horrible call and win. The program is meant to make you make the correct EV decisions so you can win over the long term instead of making calls that you dont have pot odds for and getting lucky.
Im by no means saying that the software is perfect but when i was new to the site i thought it was not worth the money as well but i have come back multiple times.
The site is not GTO, in fact the bots are intended to be quite exploitative like people. There is actually a post that compares this site to Poker Snowie that goes into this discussion further.
The reports are only guidelines, and are there to help you look at areas where you might be doing things differently from OTHER WINNING Players. It is likely that you are NOT playing the same hands, although maybe certain scenarios do come up. What the reports are telling when you are calling on the turn when you are behind, is that this is generally not a good thing, Sure you could win your hand but it doesn't mean that this was wrong or even right. In the long run however, if you are calling a wide range of hands when you are behind on the turn you are likely spewing a lot of money. The bots should be raising more given you chase draws and bluff catchers.
the examples you listed where you raised and got bots to fold better hands demonstrates both the exploitative play of the bots and the effectiveness of things like CBET and check raises vs certain bots at certain levels. If you increase the difficultly the bots should be adjusting. Also the bots would have to take into account what hands you are betting and compare this to calling with a marginal hand. This takes more significance in MTTs and less so in cash where you are more than likely to get called down.
Hope this helps. Not an expert about the site, but sharing things I think you might not be noticing.
If you want a lot of data, upload your hands into PT4. They work for cash, I have not got them to work for MTTs smoothly.
I dont understand how anyone can say he isnt playing the same hands. Over a large sample size of RNG hands he is going to have the same percentage of hands as everyone else.
There are a total of 133,784,560 distinct 7-card combinations possible in NLHE. If you were dealt 1 hand per second and no hand was ever repeated, it would take you ~4 1/4 years without stopping to see them all. If you ignore suits, the number of possible 7-card combinations becomes more manageable at a mere 6,009,159. That would only take you ~69 1/2 days.
These numbers are just to see all the possible combinations. Now add in all the variations of position and the numbers balloon to the point where you wouldn't be able to see them all in decades of non-stop viewing. Add in the time it takes to actually play a hand and we are talking several lifetimes without repeat.
Times when you called behind on the turn are losses in the long run, and you don't want to see many of those unless you are planning to bluff your opponent off on the river. But you can only do that so much before you get yourself into trouble. The IQ scores in any given session don't mean a lot, but if you find yourself typically below 100, it means other APT users are on average making fewer mistakes than you.
It's not just about the cards you're being dealt but also the cards that the other players are dealt. If the cards are randomly distributed then the probability that you're playing the exact same hand as any other player is ridiculously tiny. Yes, you may both be dealt AQs but the bots at the table will certainly not be dealt the exact same cards, therefore you're not playing the same hand. So unless the cards are NOT randomly distributed but the same for every human player and all the bots at the table then the hands played cannot be compared. So, in the weekly challenge, unless everyone and all the bots get the same exact cards throughout the 50 hands dealt, then results cannot be compared between players.
Another issue is the issue of luck. The measurement of luck should be going two ways: the amount of luck you're getting (e.g. How many times you flop a set vs. the theoretical percentage) but also the amount of luck the bots get. If you hit a set on the flop 15% of the time then sure you're overly lucky, but if a bot flops a straight 5% of the time, then are you really lucky?
So now how is that parameter taken into account, if at all, in the luck factor calculation? I don't think it is.
I have read David Sklansky and I understand these concepts very well. The example I was describing in fact show that I can get the bots to fold even when they have the best hand. If I raise while behind and get a bot to fold, then that should count as a positive in my poker IQ, not a negative which seems to happen here. If I fold a better hand then, yes, it should be a negative, but not the other way round. Furthermore, I would argue that calling behind is not always the wrong play if there's a purpose to that play, for example if you plan on raising the turn if a specific card comes. Of course if all you do is call in the hope of picking up a card that matches your whole cards then of sure that can be a mistake in the long run (unless of course the odds are right)
Getting them to fold on the river would increase your IQ score - and by more than the cost of calling on the turn. There are a bunch of threads on the forum that explain the calculation of the IQ score - I think one is titled ABC poker, but if you search for IQ you will find several. If your score is still showing up low, you are making other mistakes. IQ scores are very volatile session by session, but overall they are accurate in terms of comparing you to other players on the site (and only that - compared to other players on the site). You have to look at everything to really get a feel for where you are costing yourself money. Personally, I think if you have a lot of hands where you are calling on the turn behind, that is not an indication of good play. You may be able to chase the bots off hands sometimes, but that is not something you can do on a regular basis in the long run.
@AllenBlay said:
"Personally, I think if you have a lot of hands where you are calling on the turn behind, that is not an indication of good play"
Are you saying that based on the actual analysis of my data or off the top of your head? Because my average poker IQ is around 110, so I might be calling behind on the turn sometimes, which of course is a flaw, but that doesn't happen as often as you seem to believe.
I was actually talking about a specific report where I got an 88 IQ which didn't make sense. I also pointed out that the reports were not very helpful in the sense that they're not telling me any specific mistake I've made. Just calling behind on the flop is not always a mistake, it depends on the opponent, the odds, the position etc... If that were the case, then nobody should ever call a bet on a missed flop with AKs, right?
Anyway, I was trying to point out that I think that the lack of specifics is a flaw in the reports if one is really trying to improve.
Ok, this was just a general comment, not specific to you. If you are getting 110 on average, you are doing well. The 88 is probably an anomaly because there is a significant amount of variance within a single session just based on the cards the opponent had.
I agree that it would be awesome if we could give a lot more specifics, but that really isn't possible to do because everything is so situation-specific. In my opinion the best way to use the software to improve is to spend most of the time working on getting better at assessing odds and opponent ranges using the brain button. I don't use the advice much, but I do use the brain button a lot to work on my odds against potential relevant hand ranges. Once you know what your odds are, everything else becomes much easier. Just the sheer volume of hands you can play on here is the big selling point to me. As Doyle Brunson always said, poker is all about repetition and pattern recognition. The more hands you play, the better you get at recognizing patterns and what your opponent might have.
The reports are good at identifying specific hands that might have been an issue -but it definitely is not close to 100% accurate because it has no way of knowing exactly what you were trying to do. But the hands flagged do sometimes help identify potential issues. Also, I think just looking at the 169 hand chart is valuable because sometimes you can identify specific hands where you aren't maxing out your winnings. Of course, this is only meaningful after you have been playing a bunch of hands on the site.
Thanks for the comments and this discussion.
Thank you for following up and your input