This is part of the plan - saved ranges will be coming this year.
I had originally planned to do this earlier, but surprisingly it was pretty low on the requested features list.

Any idea when there will be more 'ask the oracle' questions?

Especially the Q:

How often does - player X - have - top pair - or better - by flop/turn/river

Odds Oracle 1.15 is out! (http://propokertools.com/odds_oracle).
It includes the Advanced Range Constructor (http://propokertools.com/oracle_help...ge_constructor) for Omaha. If people like this approach I will be adding something similar for the other games.

Your feedback is most appreciated.

- bachfan

Some screens:

Does it have a feature like in ev slice whereby you can view the distribution of equity from flop to turn on each possible turn card (with a graph and a table)? I konw you can view equity distribution for xxxx vs xxxx on the ranges of flops but what i'm looking for is given xxx flop and xxxx vs xxxx how will each turn card change the equity matchup. I dled the trial and all i can see is the standard equity distribution given a random flop and a random turn

Thank you for the question - I will add this to the todo list.

Price going up!

Fair warning - I intend to bump the Odds Oracle price in one week. If you were planning on buying a license, now would be the time (yes, all upgrades are free forever).

http://propokertools.com/odds_oracle

Cheers,
bachfan

Odds Oracle 1.2 is out! (http://propokertools.com/odds_oracle).

New this release: saved ranges/macros, saved simulations (so you can see the last few sims you ran), and a PQL command-line interface.

Cheers,
bachfan

Maybe im reading this wrong, why to both the shover and the caller show a net profit?

The pot started with 4 bets in it. That's why the opponent net and shover net totals to 4.

Hmmm... I don't understand this...

But lets say he raises the Button and we jamm our 12.5 bets AI from the small blind. He will call the shove with 50%... The call can never be profitable for him, even though he already invested 2.5 BBs with his preflop raise (pre ante..).

The way I would calculate it, the result would look like this... What am i doing wrong ?

It appears we are not computing the same thing.

Without looking at this too closely, it appears that in your spreadsheet villain has negative equity when he folds.

In the Shove Equity Tool, by contrast, it is assumed that all bets already in the pot are dead money, thus when shover shoves and opponent folds, opponent has net chips of 0 by definition. Put another way, the Shove Equity Tool evaluates net chips for each player _after the shove occurs_.

I hope this helps,
bachfan

ok, i understand, thanks

I'm really impressed with Odds Oracle and am especially happy that, as of 1.2, I can now run PQL queries from the command-line.

One enhancement to this that I would really appreciate would be the ability to run multiple queries without firing up a new JVM for each. This way the overhead of creating a new OS process and re-loading all the classes for each query is eliminated. Also, I can get more of the performance benefits of JIT compilation and the generational heap. There are two ways I could see this happening -- and, both have benefits...

1) RunPQL.main() could parse multiple queries out of it's command-line arg or stdin.

2) There could be a query parsing method that could be invoked from other java/jython code. This is already possible with some hackery. I can call RunPQL.main() from my own java/jython code. The only issue is that RunPQL calls System.exit(). I can get around that by installing a SecurityManager that blocks VM exits, but that's a bit ugly as I need to catch and throw away an exception for each query. Also, the output is always going to System.out, so I'd probably have to replace that with my own PrintStream if i wanted to do anything else with it.

I think I like #2 (with the need for hackery eliminated) better because I could see doing a lot of cool stuff with PQL queries from jython scripts (or my own Java).

Anyway, thanks again for the great piece of software.

One final question -- Is this thread the place you'd like to see these sorts of requests? Or is there somewhere more preferred?

This is a good place for requests.

I had similar thoughts on making batches of queries fast. It is on the todo list

- bachfan

WOW!! Just saw you added the history in the program! SWEET! props for u bachfan!

More things on my PQL wishlist
If there's already a good way to do any of this, I apologize for missing
it and would love to hear it.

1) Low-draw checking functions ie:
fourLow(player,street) : True if player has 4 cards to a qualifying low.
threeLow(player,street) : True if a player has 3 cards to a qualify low.

2) Some sort of THandTypes for low ie:
anylow
nutlow
nthlow(n)

Or some other way to do with low hands what minHandType, outsToHandType, etc..
do for high hands.

3) A way to see the players' hands and boards used in the simulations. My
desire here is to answer questions like "Show me what it looks like when I flop
nothing but am a favorite over a hand with the nuts" with a query like:

select p1, p2, board
from game='omaha8', p1='*', p2='*'
where (nutHi(p2,flop))
and (exactHandType(p1,flop,highcard))
and (equity(p1,flop)>0.5)
and (not madeLo(p1,flop))

4) A way to take a list of hands and write it as efficiently as possible using the generic range syntax. So given game='omaha8' and the following list:
ASAH3S2H ASAD3S2D ASAC3S2C ASAH3H2S
ASAD3D2S ASAC3C2S AHAD3H2D AHAC3H2C
AHAD3D2H AHAC3C2H ADAC3D2C ADAC3C2D

the tool would produce
Ax2xAy3y



Thanks again for the great tool.

Thanks for the suggestions!

- bachfan

Any idea on the ETA of selecting only hearts or x% of a hand combo ??

Thanks

Can you clarify what you mean?

If you are asking for "a hand with two hearts or in the top x% of hands", you can do this quite easily with "hh,15%" for example (using generic syntax).

- bachfan

im talking about two different things.

1: I select a hand but only want it to count for the combos with hearts. For example villain will stack off with 22 66 TT (pairs) and AKhh. If in the hand selecter i select AK suited it will count four combos. For one hand its very easy to mark yourself, but when selecting several, it makes it a lot more work.

2: When selecting villains range i assume he raises them only 50% of the time, so it should count for only 50% of the hand combos I select.

I hope you understand what i mean.

Odds Oracle 1.25 is out! (http://propokertools.com/odds_oracle).

New this release: range support for the board/dead cards (http://propokertools.com/simulations...board_examples)

Cheers,
bachfan

I have a suggestion for the "what hand type player has on flop/turn/", can you add to it to also include draws?

FYI, there is now a dedicated ProPokerTools forum:

http://forum.propokertools.com

- bachfan

P.S. I will continue to post general updates here, but if you have a suggestion or question about the Odds Oracle or propokertools.com, the new forum is the first place I will look.

Odds Oracle 1.3 is out (http://propokertools.com/odds_oracle)

New this release:

• Hand vs. Range Equity graphs for all games
• Hand History Importer (PokerStars and Merge (Carbon))

- bachfan

bach, whens the eta on the "step through hands" ala CRev for the Monte Carlo sims?