![]() |
#1
|
|||
|
|||
![]()
Okay, this is just hypothetical of course, but let's say you had access to, maybe, 500 million+ hands from Party, Stars and UB 2/4 NL+ from the last few years.
Obviously the number one thing you want to do with the hands is import them into PokerTracker and look at summary statistics for players. Things like winrates, aggression, V%IP, etc. Okay, so now let's assume you've done that, studied the stats, it's all in your HUD, etc. Moving on, now let's say you couldn't replay the hands in PokerTracker; you could only look at summary statistics, but you have another replayer. Only this one you have a developer at your disposal to make any sort of custom queries that you might want. For example, if you want to see every instance where loloTRICKEDu flopped a set in a three way pot first to act after the flop, you could set up a query to do so. Then you could watch those hands with all of the players' stats on the table, HUD-like. So my question is, if you had this at your disposal, what options would you want in the query? Here are some examples: 1. Player name(s): search for one or more players being involved in a hand, or both of them being involved in the same hand 2. Hand strength: define hand strength or hand types for the players being queried. An example might be bottom two pair or better without regards to the texture of the board. Another example might be one pair, but a middle pair held as pocket cards. 3. Flop texture (along with other options for board texture). An example is a flush draw board without a straight draw. Or just any flush draw board. Or a rainbow board without a straight draw and one broadway card. Whatever, as detailed or as simple a category as you'd like. 4. Betting patterns. How someone reacts to being check-minraised on the flop. Situations where player X check-raises the river. Hands where player X calls a raise OOP in the blinds and then fires the flop for Y% or more of the pot. etc. Those types of details. What other fields do you think would be helpful and why? And this is all in anticipation of replaying the hands that you were looking for. Also, no matter how good you are at writing PokerTracker queries, that isn't an option here... This is just about what you'd like to see assuming that it's a user-friendly interface designed for the non-database expert and no SQL. Don't worry about how to actually make the querying interface logical, just pretend it's a wishlist of items that you would want to be able to look for. Thanks, I'm interested to see what HSNL has to say about this. PS - I don't think this type of database or ability exists anywhere, so I'm not concerned about the ethical implications of data-mining or whatever. Please keep this on-topic as a theoretical discussion and not turn it into a debate about whether doing this would be the right or wrong thing to do. |
#2
|
|||
|
|||
![]()
I think some sort of 'volatility index' for various actions such as betting, checking, reraising, etc. for various positions, showdown hand strengths, etc. would be pretty useful, but it will get crazy. But theorhetical so oh well.
|
#3
|
|||
|
|||
![]()
My guess is that depending on my own strengths and weaknesses I would have trouble with different aspects of an opponents game than somebody esle for instance. Based on that I think that I'd want different things for different opponents. Most likely I would find myself in certain situations and note that someone is giving me trouble in X and Y areas, and request information to help me there. As such, I think it impossible for me to answer your question by listing things in general.
Nevertheless, I think it a real clever question. And I am curious as to what the answers will be. Sorry I couldn't be more help. |
#4
|
|||
|
|||
![]()
How many hands do you think you need on a player to actually have any kind of conclusive data for particular custom query? Flopping set multiway doesn't happen that often and you also have to get to showdown. Not to mention the play is affected by the flow of the game. In this thread, for example, consensus seems to be that stats are more useful to identify things that stand out, especially when multitabling. I guess that this kind of tool would be more useful for limit games (but keep in mind, i have no clue how high fixed limit games play.)
I still think this might be useful for reviewing/analyzing play from opponent tendencies and game theory perspective. PT by default lacks these capabilities. I, for one, wouldn't mind having something like this to analyze my own hands. Plus queries like Select ?oppCR% when opp=akshawnd & street=flop & flop=Axx & opp.position = (sb or bb) & PFR=true might actually produce some interesting results. I want hands-on experience. =) |
#5
|
|||
|
|||
![]()
this is unethecial and makes poker no longer a true art
|
#6
|
|||
|
|||
![]()
[ QUOTE ]
let's say you had access to, maybe, 500 million+ hands from Party, Stars and UB 2/4 NL+ from the last few years. [/ QUOTE ] You really don't want or need more than just the last three months. Year old stats will do you more harm than good. [ QUOTE ] For example, if you want to see every instance where loloTRICKEDu flopped a set in a three way pot first to act after the flop, you could set up a query to do so. Then you could watch those hands with all of the players' stats on the table, HUD-like. [/ QUOTE ] You are leaving out way too many variables. This won't work very well against lolo. [ QUOTE ] PS - I don't think this type of database or ability exists anywhere [/ QUOTE ] Yes it does, several people have them. [ QUOTE ] this is unethecial and makes poker no longer a true art [/ QUOTE ] No it's not and no it doesn't. [ QUOTE ] might actually produce some interesting results. I want hands-on experience. =) [/ QUOTE ] It works extremely well up to about 600NL. Above that it is not as useful. Regulars at the higher limits that you are able to get enough hands on usually adjust fairly quickly. |
#7
|
|||
|
|||
![]()
[ QUOTE ]
How many hands do you think you need on a player to actually have any kind of conclusive data for particular custom query? [/ QUOTE ] It really isn't very useful for specific players. However, it works extremely well for identifying generalized +EV spots. This is where it is still marginally useful at the higher limits. [ QUOTE ] this is unethecial and makes poker no longer a true art [/ QUOTE ] I made my last response before I noticed what site the OP runs. I was referring to personal use. As far as a public DB, I completely agree on both accounts. It's only a matter of time before some putz makes this information publicly available. Fortunately the information is only effective up to a certain limit. It is the same thing as a starting hand chart, just more advanced, getting you higher up in limits, but still not very high. It will mean more straight forward players making their way to 1000NL and above who think they're pros. And that's a good thing. |
#8
|
|||
|
|||
![]()
[ QUOTE ]
Select ?oppCR% when opp=akshawnd & street=flop & flop=Axx & opp.position = (sb or bb) & PFR=true [/ QUOTE ] I am not fluent in pokertracker-ese. what does this do? Is there a good guide somewhere as to how to write these queries? |
#9
|
|||
|
|||
![]()
[ QUOTE ]
PS - I don't think this type of database or ability exists anywhere, so I'm not concerned about the ethical implications of data-mining or whatever. Please keep this on-topic as a theoretical discussion and not turn it into a debate about whether doing this would be the right or wrong thing to do. [/ QUOTE ] I could be misinterpreting OP's last paragraph here, but as I understand it he is not concerned with the literal implementation of this system. Basically it's really clever way of asking: "what do you look for when you are searching weaknesses in your oppponents style?", or "what information do you think in general is important in looking for weaknesses?" I could be offbase here, so OP correct me if I'm wrong. However I think that this thread widing up in a discussion on ethics would be a true waste and not the OP's intention. |
#10
|
|||
|
|||
![]()
[ QUOTE ]
I could be misinterpreting OP's last paragraph here, but as I understand it he is not concerned with the literal implementation of this system. Basically it's really clever way of asking: "what do you look for when you are searching weaknesses in your oppponents style?", or "what information do you think in general is important in looking for weaknesses?" I could be offbase here, so OP correct me if I'm wrong. However I think that this thread widing up in a discussion on ethics would be a true waste and not the OP's intention. [/ QUOTE ] He runs the website thepokerdb.com "Your free source for tracking online tournaments and players." |
![]() |
|
|