Poker Tracker Stud

It is a program that is being sold to poker players. Should you consider buying it? I have to say NO! While many players love to quote their VP$IP and so on, the information is cursory and somewhat inaccurate.  It has important and useful data that comes wrapped in frustration.

I programmed database application for NYSE listed companies. I did them from design upward. Sucking data into a program is the trivial component. You do want to make sure the data is accurately imported but the structure of the database determines how and what can be referenced easily and efficiently.

This application is written using PowerBuilder from Sybase. This development tool is in the category of enterprise software. Such applications are kitchen sink tools. They try to be all things for all people. What they end up being is very structured design tools that force themselves on the design. They tend to be bloated and sluggish with a user interface that limits user interaction. SAP is a competing product line and every bit as frustrating to end users.

So, from the git-go, the application has sever limitations that affect the user experience. That is annoying but not insurmountable. The database used is Microsoft’s distribution database. It isn’t the best product but it carries no added cost to the developers. If you do a good design of the structure and keys, it can be adequate.

Documentation is a key component of any program development. The product is a black box product with limited documentation. A black box application is one that does not disclose what it does; it merely displays results. I see things like a player being profitable in overall play with a negative BB/100 statistic. I was told this was the preferred method and the anomaly related to table levels. That is one way to skin the cat but that method isn’t part of the documentation and, even on inquiry, the response was very limited.

On of the stats that is provided and extolled by many is VP$IP. (Voluntarily Put $ In Pot) That is more straightforward in a flop game. Stud has a forced bring in. It is possible that the hand will play to late streets and even the river without further betting. Or the bringin may fold to the first bet. However, that first bet may be several streets in.

I had an exchange with the programmer shortly after my purchase. I was referencing the data provided by PokerStars. They provide a statistic for participation through various streets. If you are the bring in, that is not a part of their statistic other than it shows that you were active on 4^th street. That happens when nobody completes on 3^rd street to force a decision from the bring in player. On a full table a player will have a bring-in 12.5% of the hands. Just how many are hands he wishes to continue placing money in the pot will vary but it isn’t a high percentage normally. At times and after playing for a significant period, the VP$IP and 4^th street data from PokerStars are often very close. I’ve even seen it where the report by the application is marginally higher. So, I don’t trust that info but it retains marginal usefulness.

The application accesses the hand histories being saved by the site. The application inspects typically text in the form of hand histories for the sites it supports. That worked on PokerStars for several months and then would lose a random group of hands reporting that it could not find the completion of the hand. This distorted all information for the session and took some time to fix with an update release. Not too acceptable.

Inspecting text and converting it to one’s data is annoying programming. It is called parsing. A well written parser has some flexibility. Hand histories very from site to site but only in minor syntax. It might be: raise, Raise, RAISED, Raises etc. for the same action. These are key words that will trigger how information is collected. A flexible tool might work with the majority of sites. One that isn’t well written or where the programming environment limits the approach can be brittle. This seems brittle.

Another part of parsing is efficiency. You can use a lot of processor if the algorithm is inefficient. I wrote a parser that I placed in the public domain that was more efficient than the common methods in use. It was well received by those using the languages I provided it for. (several BASIC variants and PL/I) PT is very inefficient. When it cycles the data collection phase, not only does the program seem to hang but the other application active are sluggish. That’s really chintzy. It may be attributable to the enterprise nature of the development environment. But, that just means the programming sucked from the beginning poor choices.

One thing that could be handy is the notes. It can selectively export information about the other player. It is nice to know just how profitable a person is along with other stats and you can design what the program will export to the site’s notes file. That worked swimmingly for the first few tries. For the last 3 months or so, the export is about the same number. It exports within a couple around 240. In that period I played around 10,000 additional hands. The program is somehow stuck. So, I have to depend on the info on the other useful tool that they call the Game Time Window that has a subset of the data available with the export notes function.

I can’t say the application is horrid. Certainly parts of it are. Others are marginally useful and I continue to load the application. What is annoying is that it could be so much more. Other components like grading players just aren’t useful in their present form. Collection of data is a fairly trivial task with a decent database manager in hand. Manipulating that data to resolve questions or display in a flexible format is the key to having a good program or just a big bucket of junk. This program sadly falls toward the lower end of the scale.

I could have documented more stuff. But, the darn thing is so annoying that I hate to use any more of it than I have to. So, I have to say to those contemplating its purchase: probably, don’t bother. There are bits that help but its a pretty annoying way to get to some really decent info. I guess I can recommend it to someone who want to keep a record of their winnings and doesn’t want the scut work of maintaining a spreadsheet. I’m in that category and that the main reason I continue with it.

P.S. I only use it on PokerStars. When I’d switch to FullTilt and then back, the Stars data often goes missing for a day or so. I magically reappears at some point but I just don’t bother with it anymore. And telling their support guy would be another frustrating experience I can do without.

ADDENDUM:

The bad beat jackpot at PartyPoker was hit with a balance at over 1-million dollars. Here’s the hand results from a .50-1 limit table:

ShoulderGuy balance $94.28, lost $8

[Full House -   Kd,   Kh   Kc   9h   9d]

judith75 balance $59, lost $8

[Four of a kind -   9c,   9s   9h   9d   Kc]

Hiyaall balance $44.77, bet $8, collected $24, net +$16

[Royal Flush -   Ac,   Kc,   Qc,   Jc    Tc ]

Judith75 wins Bad Beat Jackpot - $354,683.57

Hiyall wins - $177,341.79 + $24 from the POT!!!!

All other players win 22,167.72

This entry was posted on Friday, August 15th, 2008 at 3:08 pm and is filed under Journal.