New Beta - RängeMeister - Hand Analysis Facilitator
11-30-2010
, 02:01 PM
Often times when playing poker, you'll run into situations where you're not sure what to do. What range of hands does your opponent have? How will he respond to a check or bet? What are the possibilities for how the hand could play out?
RängeMeister is based on the theory that good players are effectively able to leverage previous experience to come up with good estimates to answer these questions. The objective of RängeMeister is to help players evaluate situations that they've found themselves in objectively by listing similar hand histories as well as the relevant data from these hands.
Using RängeMeister is simple; you just paste a partial hand history into the input screen just like you would if you were about to post the hand in a forum. Then you reduce search results by applying filters. You can create any number of "what if" scenarios and see results conveniently in a replayer.
Here is an 8.5 minute video demonstrating usage http://www.youtube.com/user/AssaultW.../0/tLUrbWhH0pg
Screenshots
setup wizard
A simulation in action
Data extracted from similar hands
setup wizard
A simulation in action
Data extracted from similar hands
RängeMeister works by indexing the hands that are in your HEM/PT3 databases and storing that information in a proprietary database. Your HEM/PT3 databases won't be written to at all. Once some hands are indexed, you can search for hands by inputting partial hand histories and go from there. Please note that you do not need to wait for your whole database to get indexed before performing searches.
RängeMeister currently works with the following poker rooms:
RängeMeister is currently an early beta. That means there are likely to be bugs. Additionally, many of the features have not been fully fleshed out yet. Currently, RängeMeister works only for NLHE/LHE cash games.PokerStars
Some of the things that will be coming up
-tournament/SNG support
-better equity calculation based on similar hands
-filtering based on more stats as well as "caddy" notes
You can download RängeMeister from the download page
11-30-2010
, 02:48 PM
12-01-2010
, 11:51 AM
12-01-2010
, 07:04 PM
12-01-2010
, 07:20 PM
12-02-2010
, 01:54 PM
12-02-2010
, 02:26 PM
12-02-2010
, 02:36 PM
12-02-2010
, 02:40 PM
12-02-2010
, 03:12 PM
12-03-2010
, 05:30 AM
Install and index fine with W7. Your site says Net 4 needed for NC but I had it working with without it. Net 4 WAS needed for RM
Pasting partial HH from HEM has the same annoyance as with NC (not your fault
)
If we select a hand in HEM; right click and ask to view HH we see an output in HEM's own format. If we copy and paste THAT we get an error. However, if we export the HH and then copy and paste from the txt file it produces, RM has no problem.
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format
I tried one test hand I CB and the opponent check/raised and my initial observations are:-
a) Historical Range for alpha player chart has the player name of the check/raiser - clearly it is not HIS chart but the results of all alpha players
b) The chart gave 7 hand descriptions 1 each of 7 types and each one was at the 1 level on the chart. If there had been 8 showdowns and "mid-set" had been 2 would that have been at 2 and the others @ 1? If so, is that the best way to do it? Raw numbers probably have to be converted to % in our mind (well they do for me). I would prefer to see the graoh as 14% for each and, in the 8 results mid set=2 example, I would like 25% for mid-set and 12.5% for the remaining 6 results
c) The requirement to SD still seems to be a weakness to me. The implied weakness as introduced in NC would once again be useful here. There were 22 hands that RM found that started like this; 7 saw a SD as explained above; in 3 hands the alpha player folded before SD. It is not unreasonable to think that he did not have a monster when he made the XR. The omega player folded 7 times to the raise; that is useful if we are investigating the likely success of the check/raise but it tells us nothing about the alpha player's hand range does it?
That leaves us 15 hands from which we could learn something about alpha's range; 7 we know because he went to SD; 3 he folded before SD and that leaves 5 hands that he won before SD. These 5 are what I would call "implied strength" hands because to win before SD having got past the flop, the alpha player needed to fire again.
IMO a more useful chart of those results for hand range would be 1 each for the known hands; 3 of implied weakness and 5 for implied strength. That would give 33% for implied strength; 20% for implied weakness and about 7% each for known hand types.
If we will be forced to fold later because in a particular scenario we often face additional bets from implied strength hands, we are probably as well to fold now if we are marginal
d) Reaction chart also shows numbers rather than % and has the name of the player in just one hand showing whereas this is the reaction from ALL players.
Cheers
T
Pasting partial HH from HEM has the same annoyance as with NC (not your fault
)If we select a hand in HEM; right click and ask to view HH we see an output in HEM's own format. If we copy and paste THAT we get an error. However, if we export the HH and then copy and paste from the txt file it produces, RM has no problem.
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format
I tried one test hand I CB and the opponent check/raised and my initial observations are:-
a) Historical Range for alpha player chart has the player name of the check/raiser - clearly it is not HIS chart but the results of all alpha players
b) The chart gave 7 hand descriptions 1 each of 7 types and each one was at the 1 level on the chart. If there had been 8 showdowns and "mid-set" had been 2 would that have been at 2 and the others @ 1? If so, is that the best way to do it? Raw numbers probably have to be converted to % in our mind (well they do for me). I would prefer to see the graoh as 14% for each and, in the 8 results mid set=2 example, I would like 25% for mid-set and 12.5% for the remaining 6 results
c) The requirement to SD still seems to be a weakness to me. The implied weakness as introduced in NC would once again be useful here. There were 22 hands that RM found that started like this; 7 saw a SD as explained above; in 3 hands the alpha player folded before SD. It is not unreasonable to think that he did not have a monster when he made the XR. The omega player folded 7 times to the raise; that is useful if we are investigating the likely success of the check/raise but it tells us nothing about the alpha player's hand range does it?
That leaves us 15 hands from which we could learn something about alpha's range; 7 we know because he went to SD; 3 he folded before SD and that leaves 5 hands that he won before SD. These 5 are what I would call "implied strength" hands because to win before SD having got past the flop, the alpha player needed to fire again.
IMO a more useful chart of those results for hand range would be 1 each for the known hands; 3 of implied weakness and 5 for implied strength. That would give 33% for implied strength; 20% for implied weakness and about 7% each for known hand types.
If we will be forced to fold later because in a particular scenario we often face additional bets from implied strength hands, we are probably as well to fold now if we are marginal
d) Reaction chart also shows numbers rather than % and has the name of the player in just one hand showing whereas this is the reaction from ALL players.
Cheers
T
12-03-2010
, 08:28 AM
Quote:
.
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format........
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format
........
12-03-2010
, 08:31 AM
but now I have released a patch 0.1.0.4 that should allow for full hand histories to be entered. The only thing is there won't be an omega
Quote:
Install and index fine with W7. Your site says Net 4 needed for NC but I had it working with without it. Net 4 WAS needed for RM
Pasting partial HH from HEM has the same annoyance as with NC (not your fault)
If we select a hand in HEM; right click and ask to view HH we see an output in HEM's own format. If we copy and paste THAT we get an error. However, if we export the HH and then copy and paste from the txt file it produces, RM has no problem.
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format
Pasting partial HH from HEM has the same annoyance as with NC (not your fault
)If we select a hand in HEM; right click and ask to view HH we see an output in HEM's own format. If we copy and paste THAT we get an error. However, if we export the HH and then copy and paste from the txt file it produces, RM has no problem.
Obviously it is natural for users of RM and HEM to want to copy/paste from the HEM format because usually that is where they see the hand they want to investigate. It would be useful if you include alternative parsing to be able to read that format or for you to convince HEM to show the actual HH rather than their own format
Quote:
I tried one test hand I CB and the opponent check/raised and my initial observations are:-
a) Historical Range for alpha player chart has the player name of the check/raiser - clearly it is not HIS chart but the results of all alpha players
I tried one test hand I CB and the opponent check/raised and my initial observations are:-
a) Historical Range for alpha player chart has the player name of the check/raiser - clearly it is not HIS chart but the results of all alpha players
Quote:
b) The chart gave 7 hand descriptions 1 each of 7 types and each one was at the 1 level on the chart. If there had been 8 showdowns and "mid-set" had been 2 would that have been at 2 and the others @ 1? If so, is that the best way to do it? Raw numbers probably have to be converted to % in our mind (well they do for me). I would prefer to see the graoh as 14% for each and, in the 8 results mid set=2 example, I would like 25% for mid-set and 12.5% for the remaining 6 results
Quote:
c) The requirement to SD still seems to be a weakness to me. The implied weakness as introduced in NC would once again be useful here. There were 22 hands that RM found that started like this; 7 saw a SD as explained above; in 3 hands the alpha player folded before SD. It is not unreasonable to think that he did not have a monster when he made the XR. The omega player folded 7 times to the raise; that is useful if we are investigating the likely success of the check/raise but it tells us nothing about the alpha player's hand range does it?
Quote:
That leaves us 15 hands from which we could learn something about alpha's range; 7 we know because he went to SD; 3 he folded before SD and that leaves 5 hands that he won before SD. These 5 are what I would call "implied strength" hands because to win before SD having got past the flop, the alpha player needed to fire again.
IMO a more useful chart of those results for hand range would be 1 each for the known hands; 3 of implied weakness and 5 for implied strength. That would give 33% for implied strength; 20% for implied weakness and about 7% each for known hand types.
If we will be forced to fold later because in a particular scenario we often face additional bets from implied strength hands, we are probably as well to fold now if we are marginal
d) Reaction chart also shows numbers rather than % and has the name of the player in just one hand showing whereas this is the reaction from ALL players.
Cheers
T
IMO a more useful chart of those results for hand range would be 1 each for the known hands; 3 of implied weakness and 5 for implied strength. That would give 33% for implied strength; 20% for implied weakness and about 7% each for known hand types.
If we will be forced to fold later because in a particular scenario we often face additional bets from implied strength hands, we are probably as well to fold now if we are marginal
d) Reaction chart also shows numbers rather than % and has the name of the player in just one hand showing whereas this is the reaction from ALL players.
Cheers
T
12-03-2010
, 08:33 AM
People should be entering partial hands for the most part though so they can see different possible outcomes for the hand. If they enter the full hand they can only see possible ranges and that kills most of the value of using it
That said, you can still do what fozzy says and then cut off the end of the HH in the RM entry field
12-03-2010
, 08:45 AM
Quote:
OMG Fozzy ITT
People should be entering partial hands for the most part though so they can see different possible outcomes for the hand. If they enter the full hand they can only see possible ranges and that kills most of the value of using it
That said, you can still do what fozzy says and then cut off the end of the HH in the RM entry field
People should be entering partial hands for the most part though so they can see different possible outcomes for the hand. If they enter the full hand they can only see possible ranges and that kills most of the value of using it
That said, you can still do what fozzy says and then cut off the end of the HH in the RM entry field
12-03-2010
, 08:46 AM
12-03-2010
, 08:47 AM
12-03-2010
, 08:49 AM
12-03-2010
, 08:54 AM
12-03-2010
, 08:58 AM
12-03-2010
, 09:18 AM
12-03-2010
, 09:21 AM
SreticentV
Fozzy solved the copy/paste HH problem; we can get it from HEM. I had just not spotted how - my bad!
His solution is GREAT - my alternatives are crap
I think whether it gives the player name or says alpha or is totally blank on the range screen is pretty irrelevant really. I just mentioned it in case it was not what you intended
Fozzy solved the copy/paste HH problem; we can get it from HEM. I had just not spotted how - my bad!
His solution is GREAT - my alternatives are crap
I think whether it gives the player name or says alpha or is totally blank on the range screen is pretty irrelevant really. I just mentioned it in case it was not what you intended
12-07-2010
, 08:26 AM
Quote:
TBH I dont think thats the best way to do it.
I think RM should look at a HH and depending on what street we are on RM should ignore everything from future streets untill we get to the next street.
I can see what you are saying though.
BTW when you say it looks for similar hands.. how similar do bet sizes need to be? Whats the % tollerence between a bet being classed as similar and being considered different?
I think RM should look at a HH and depending on what street we are on RM should ignore everything from future streets untill we get to the next street.
I can see what you are saying though.
BTW when you say it looks for similar hands.. how similar do bet sizes need to be? Whats the % tollerence between a bet being classed as similar and being considered different?
As far as hand similarity, it works like with the NoteCaddy bet size filters. So it's by ranges of pot size etc
Quote:
Picking up on the ideas in this thread
http://forumserver.twoplustwo.com/45...ftware-933030/
and also in another thread in which Handcrawler was being discussed.
It seems to me that the HM DB structure and filters are geared toward hero involvement in a hand or the showdown. When we fold pre or during a multiway hand, a lot of the useful info becomes extremely difficult/impossible to access.
http://forumserver.twoplustwo.com/45...ftware-933030/
and also in another thread in which Handcrawler was being discussed.
It seems to me that the HM DB structure and filters are geared toward hero involvement in a hand or the showdown. When we fold pre or during a multiway hand, a lot of the useful info becomes extremely difficult/impossible to access.
Quote:
Handcrawler seemed to be heading in a good direction to be able to access similar scenarios already posted.
Just wondering if there is any intent for this wealth of information contained within our own dbs/hand histories to be accessed?
Just wondering if there is any intent for this wealth of information contained within our own dbs/hand histories to be accessed?
12-07-2010
, 01:25 PM
First, thanks a lot for writing this software. It's been really helpful for the 2 days I've used it, and it has tons of potential.
My HM suggestion to allow us to copy original hand histories from the replayer was due to RangeMeister. If you'd like this feature, please vote for it on their forums: http://holdemmanager.uservoice.com/f...he-site-specif
But since this probably won't ever be on their priority list, I second the request to have RM be able to parse HM's Txt output.
My other RM suggestions/requests:
- Does RM recognize draws? When looking at some ranges, I don't think I've seen them, but I don't think they should just be grouped into Nothing. It would be nice to get the various draws separated.
- For post-flop hand values and actions, I think a pie chart would be a better representation than a looooooong bar graph. For actions, also showing percentages would be nice than just straight counts.
- For holecard ranges, I think a matrix like the one in HM would be best:
- ability to export results to a CSV file
My HM suggestion to allow us to copy original hand histories from the replayer was due to RangeMeister. If you'd like this feature, please vote for it on their forums: http://holdemmanager.uservoice.com/f...he-site-specif
But since this probably won't ever be on their priority list, I second the request to have RM be able to parse HM's Txt output.
My other RM suggestions/requests:
- Does RM recognize draws? When looking at some ranges, I don't think I've seen them, but I don't think they should just be grouped into Nothing. It would be nice to get the various draws separated.
- For post-flop hand values and actions, I think a pie chart would be a better representation than a looooooong bar graph. For actions, also showing percentages would be nice than just straight counts.
- For holecard ranges, I think a matrix like the one in HM would be best:
- ability to export results to a CSV file