This paper examines whether the 30-day US dollar futures contract traded on A3Mercados (DLR) is a good predictor of the wholesale spot exchange rate thirty days ahead. Using 5,741 daily observations between January 2003 and September 2026, we build a constant-maturity 30-day futures series and compare it with the realized spotrate across six exchange-rate regimes, from managed floating to the current band scheme. Futures prices are not unbiased predictors regime by regime: they underestimated devaluation under the 2011-15 exchange controls and overestimated it by roughly 1.2%
per month in 2019-25, when the futures price stood above the realized spot on 91% and 96% of days in the last two control sub-periods. Under floating regimes the futures forecast is barely distinguishable from a random walk (Theil's U between 0.90 and 1.02), because discrete jumps in the exchange rate can not be anticipated. Pooled over the full sample the unbiasedness hypothesis is not rejected (β = 1.06), but this result conceals strong regime heterogeneity. We conclude that the DLR futures price measures expected devaluation adjusted for risk and interest rates,not an unbiased forecast.