After each election cycle, Campaigns & Elections presents the win--loss records of political consultants across the nation.

Before you begin your review of the consultants and their clients--winners and losers are all exhibited in stark relief--you should consider the following:

SCOPE: We include political consulting firms that were identified over the past year as being involved during the 2004 elections.

SCORING: We list the reported and identified clients for each firm together with win-loss codes. We do not compile a cumulative box score for each firm because of the comparative nature of consultant-client relationships. For example, a strategically complex U.S. Senate race in a diverse, competitive state should not be evaluated the same as an easy incumbent re-election to the state legislature in a safe seat.

ACCURACY: It is not possible to determine the involvement of every political consulting firm or vendor from available public records, particularly those at the state and local levels. As such, we often rely upon the firms themselves to provide information, although we make efforts to verify what we have received. In most cases, firms have cooperated in the reporting of information. In some cases though, firms either refused to provide information or have submitted only partial client listings; in these instances, we have researched additional information sources to attempt to verify available client listings.

NON-CANDIDATE CLIENTS: Many political consulting firms have association, PAC, party committee and independent expenditure/issue advocacy (I-E) clients in addition to candidates and ballot propositions they directly represent. In some cases, firms have provided us with the names of all or some of these clients; in other cases, they have not. Because many of these non-electoral clients do not appear on public records, and often involve confidential relationships, inclusions vary from firm to firm. It is important to remember that non-electoral and independent committee client lists are possibly incomplete and do not attempt to represent any firm's full corporate, issue and association business roster.

SUMMARIES: In the interest of space, we have combined some state and local elections and reported totals without listing individual candidates.

CORRECTIONS: It is inevitable that after publication of such extensive information that corrections and additions will be reported to us. Once we confirm corrections and changes, they will be published in a subsequent issue.

Research assistance by Erin McPike, Rachel Saffron and Robin T. Reid.