Big Bath is among the types of management of results and consists in the reduction of current result aiming at its increase in the future period. Although the results management mechanism may affect profit in some way, its use can not be confused with fraud, since it is possible to flex the accounting of an accounting event so that the postponement of revenue, provided it is expected under the accrual basis. The use of mechanisms that may impact the profit can affect the qualitative characteristic of relevance, compromising the decision making process by the users of the accounting information, which motivates the present research. This study aimed to identify the profile of companies with evidence of the use of Big Bath with shares traded in Brazil, Bolsa, Balcão (B3) using the Pae Model (2005) to measure discretionary accruals. In a sample of 209 publicly traded companies operating in Brazil, which was investigated in the period from 2012 to 2016, the panel data of the Pae Model (2005) with corrected heteroscedasticity (White) was used. The study sought to answer the hypotheses of eight characteristics (size of company, level of corporate governance, degree of indebtedness, control structure, manager change, audit company and audit company change) to form a profile of companies with evidence of Big Bath. Overall, we found 52 evidence of use of Big Bath during the selected interval, and 40 of them were classified as low level and 12 as high level of results management. Most companies with evidence of Big Bath have the following profile: smaller size, not included in the Novo Mercado, high indebtedness, independent of the sector of operation, less concentrated control structure, permanence of the same manager and auditor of the previous period and are audited by Big Four.