diff --git a/R/flatline_forecaster.R b/R/flatline_forecaster.R
index e437f50ea..197c8cca5 100644
--- a/R/flatline_forecaster.R
+++ b/R/flatline_forecaster.R
@@ -94,6 +94,12 @@ flatline_forecaster <- function(
 #' Constructs a list of arguments for [flatline_forecaster()].
 #'
 #' @inheritParams arx_args_list
+#' @param ahead Integer. Unlike [arx_forecaster()], this doesn't have any effect
+#'   on the predicted values. Predictions are always the most recent observation.
+#'   However, this _does_ impact the residuals stored in the object. Residuals
+#'   are calculated based on this number to mimic how badly you would have done.
+#'   So for example, `ahead = 7` will create residuals by comparing values
+#'   7 days apart.
 #'
 #' @return A list containing updated parameter choices with class `flatline_alist`.
 #' @export
diff --git a/R/layer_cdc_flatline_quantiles.R b/R/layer_cdc_flatline_quantiles.R
index c64b96c2d..3f178f6da 100644
--- a/R/layer_cdc_flatline_quantiles.R
+++ b/R/layer_cdc_flatline_quantiles.R
@@ -149,7 +149,7 @@ propogate_samples <- function(
     for (iter in 2:max_ahead) {
       samp <- shuffle(samp)
       raw <- raw + samp
-      if (symmetrize) symmetric <- raw - (median(raw) + p)
+      if (symmetrize) symmetric <- raw - (median(raw) - p)
       else symmetric <- raw
       if (nonneg) symmetric <- pmax(0, symmetric)
       res[[iter]] <- symmetric
@@ -157,7 +157,7 @@ propogate_samples <- function(
   }
   res <- res[aheads]
   list(tibble::tibble(
-    aheads = aheads,
+    ahead = aheads,
     .pred_distn = map_vec(
       res, ~ dist_quantiles(quantile(.x, quantiles), tau = quantiles)
     )