In poker parlance, 'data mining' has come to mean the collection of hand history data for poker games, stakes and sites of interest. The uses of this data are many and varied, and will be discussed in more detail in future articles. For now, though, I will concentrate on the ways in which you might go about obtaining this data repository.

When it comes to obtaining hand history data, poker sites can be grouped into three main classes:

• Those that record hand histories in a way that can be incorporated into a database without requiring additional software.
• Those that do not automatically record observed hand history data, but for which this data may be obtained using a separate software program (commonly referred to as a 'handgrabber').
• Those for which observed hand histories cannot at this point in time be obtained (though software to permit this may well in time be developed).

As increasing numbers of poker players become aware of the uses of mined data in preparing for play (e.g. by observing the statistical differences between winning and losing players at the stakes of interest), in table selection, and in making decisions based on more informed player 'reads' during play, so the debate on the legitimacy of data mining intensifies. So far PokerStars has set out its anti-datamining stance most clearly. As per its 'Prohibited Online Poker Software FAQ', software enabling the collection of hand histories from open tables at which the user is not currently playing (handgrabbers) are prohibited.

Clearly, PokerStars recognizes the power of data mining and the advantage it offers to a skilled player who can make full use of the data. As do the authors and users of software programs such as PokerEdge and Poker Sherlock, both of which offer access to a large centralized database of hand history data collected by a(n at least partly user-supplied) data farm. Again, however, these programs are not permitted by PartyPoker or PokerStars, and their legitimacy at supported smaller sites is at best unconfirmed. The debate will run on as to whether data mining should be an important part of the online poker world, but to borrow a phrase from the promotion of one of the aforementioned software tools--the fact remains that, in online poker as in the live game, information is power.

Hand History Exchange and the programs described below offer a route to a large hand history database that does not require a monthly subscription as do the centralized tools described in the previous paragraph. At sites in categories 1 and 2, histories can be recorded to the hard drive automatically or using the programs. To make use of them in game selection and player reading, they should then be imported into a poker database analysis program. The most popular of these are Poker Tracker and Poker Office, and others are in development. Either of these costs around $50, and most poker players (including of course myself) regard this as an essential investment for any semi-serious online poker player. Along with one of these programs you will require a HUD (heads-up display) to view the essential statistics as you play. GameTime Plus (GT+) is a free HUD program.

Data Mining Tools