diff --git a/R/predict.R b/R/predict.R
index 72a0733d0..ab27c3450 100644
--- a/R/predict.R
+++ b/R/predict.R
@@ -4,45 +4,42 @@
 #'  `predict()` can be used for all types of models and uses the
 #'  "type" argument for more specificity.
 #'
-#' @param object An object of class `model_fit`
+#' @param object An object of class `model_fit`.
 #' @param new_data A rectangular data object, such as a data frame.
 #' @param type A single character value or `NULL`. Possible values
-#'   are "numeric", "class", "prob", "conf_int", "pred_int", "quantile", "time",
-#'  "hazard", "survival", or "raw". When `NULL`, `predict()` will choose an
-#'  appropriate value based on the model's mode.
+#'   are `"numeric"`, `"class"`, `"prob"`, `"conf_int"`, `"pred_int"`,
+#'   `"quantile"`, `"time"`, `"hazard"`, `"survival"`, or `"raw"`. When `NULL`,
+#'  `predict()` will choose an appropriate value based on the model's mode.
 #' @param opts A list of optional arguments to the underlying
 #'  predict function that will be used when `type = "raw"`. The
 #'  list should not include options for the model object or the
 #'  new data being predicted.
-#' @param ... Arguments to the underlying model's prediction
-#'  function cannot be passed here (see `opts`). There are some
-#'  `parsnip` related options that can be passed, depending on the
-#'  value of `type`. Possible arguments are:
+#' @param ... Additional `parsnip`-related options, depending on the
+#'  value of `type`. Arguments to the underlying model's prediction
+#'  function cannot be passed here (use the `opts` argument instead).
+#'  Possible arguments are:
 #'  \itemize{
-#'     \item `interval`: for `type`s of "survival" and "quantile", should
+#'     \item `interval`: for `type` equal to `"survival"` or `"quantile"`, should
 #'            interval estimates be added, if available? Options are `"none"`
 #'            and `"confidence"`.
-#'     \item `level`: for `type`s of "conf_int", "pred_int", and "survival"
+#'     \item `level`: for `type` equal to `"conf_int"`, `"pred_int"`, or `"survival"`,
 #'            this is the parameter for the tail area of the intervals
 #'            (e.g. confidence level for confidence intervals).
-#'            Default value is 0.95.
-#'     \item `std_error`: add the standard error of fit or prediction (on
-#'            the scale of the linear predictors) for `type`s of "conf_int"
-#'            and "pred_int". Default value is `FALSE`.
-#'     \item `quantile`: the quantile(s) for quantile regression
-#'            (not implemented yet)
-#'     \item `time`: the time(s) for hazard and survival probability estimates.
+#'            Default value is `0.95`.
+#'     \item `std_error`: for `type` equal to `"conf_int"` or `"pred_int"`, add
+#'            the standard error of fit or prediction (on the scale of the
+#'            linear predictors). Default value is `FALSE`.
+#'     \item `quantile`: for `type` equal to `quantile`, the quantiles of the
+#'            distribution. Default is `(1:9)/10`.
+#'     \item `time`: for `type` equal to `"survival"` or `"hazard"`, the
+#'            time points at which the survival probability or hazard is estimated.
 #'  }
-#' @details If "type" is not supplied to `predict()`, then a choice
-#'  is made:
+#' @details For `type = NULL`, `predict()` uses
 #'
 #'   * `type = "numeric"` for regression models,
 #'   * `type = "class"` for classification, and
 #'   * `type = "time"` for censored regression.
 #'
-#' `predict()` is designed to provide a tidy result (see "Value"
-#'  section below) in a tibble output format.
-#'
 #'  ## Interval predictions
 #'
 #'  When using `type = "conf_int"` and `type = "pred_int"`, the options
@@ -58,37 +55,42 @@
 #' have the opposite sign as what the underlying model's `predict()` method
 #' produces. Set `increasing = FALSE` to suppress this behavior.
 #'
-#' @return With the exception of `type = "raw"`, the results of
-#'  `predict.model_fit()` will be a tibble as many rows in the output
-#'  as there are rows in `new_data` and the column names will be
-#'  predictable.
+#' @return With the exception of `type = "raw"`, the result of
+#'  `predict.model_fit()`
+#'
+#'  * is a tibble
+#'  * has as many rows as there are rows in `new_data`
+#'  * has standardized column names, see below:
+#'
+#' For `type = "numeric"`, the tibble has a `.pred` column for a single
+#' outcome and `.pred_Yname` columns for a multivariate outcome.
 #'
-#' For numeric results with a single outcome, the tibble will have
-#'  a `.pred` column and `.pred_Yname` for multivariate results.
+#' For `type = "class"`, the tibble has a `.pred_class` column.
 #'
-#' For hard class predictions, the column is named `.pred_class`
-#'  and, when `type = "prob"`, the columns are `.pred_classlevel`.
+#' For `type = "prob"`, the tibble has `.pred_classlevel` columns.
 #'
-#' `type = "conf_int"` and `type = "pred_int"` return tibbles with
-#'  columns `.pred_lower` and `.pred_upper` with an attribute for
-#'  the confidence level. In the case where intervals can be
-#'  produces for class probabilities (or other non-scalar outputs),
-#'  the columns will be named `.pred_lower_classlevel` and so on.
+#' For `type = "conf_int"` and `type = "pred_int"`, the tibble has
+#' `.pred_lower` and `.pred_upper` columns with an attribute for
+#' the confidence level. In the case where intervals can be
+#' produces for class probabilities (or other non-scalar outputs),
+#' the columns are named `.pred_lower_classlevel` and so on.
 #'
-#' Quantile predictions return a tibble with a column `.pred`, which is
+#' For `type = "quantile"`, the tibble has a `.pred` column, which is
 #'  a list-column. Each list element contains a tibble with columns
 #'  `.pred` and `.quantile` (and perhaps other columns).
 #'
-#' Using `type = "raw"` with `predict.model_fit()` will return
-#'  the unadulterated results of the prediction function.
+#' For `type = "time"`, the tibble has a `.pred_time` column.
 #'
-#' For censored regression:
+#' For `type = "survival"`, the tibble has a `.pred` column, which is
+#'  a list-column. Each list element contains a tibble with columns
+#'  `.time` and `.pred_survival` (and perhaps other columns).
+#'
+#' For `type = "hazard"`, the tibble has a `.pred` column, which is
+#'  a list-column. Each list element contains a tibble with columns
+#'  `.time` and `.pred_hazard` (and perhaps other columns).
 #'
-#'  * `type = "time"` produces a column `.pred_time`.
-#'  * `type = "hazard"` results in a list column `.pred` containing tibbles
-#'     with a column `.pred_hazard`.
-#'  * `type = "survival"` results in a list column `.pred` containing tibbles
-#'     with a `.pred_survival` column.
+#' Using `type = "raw"` with `predict.model_fit()` will return
+#'  the unadulterated results of the prediction function.
 #'
 #' In the case of Spark-based models, since table columns cannot
 #'  contain dots, the same convention is used except 1) no dots
diff --git a/man/bart-internal.Rd b/man/bart-internal.Rd
index 690f715f7..2665bd380 100644
--- a/man/bart-internal.Rd
+++ b/man/bart-internal.Rd
@@ -20,9 +20,9 @@ dbart_predict_calc(obj, new_data, type, level = 0.95, std_err = FALSE)
 \item{level}{Confidence level.}
 
 \item{type}{A single character value or \code{NULL}. Possible values
-are "numeric", "class", "prob", "conf_int", "pred_int", "quantile", "time",
-"hazard", "survival", or "raw". When \code{NULL}, \code{predict()} will choose an
-appropriate value based on the model's mode.}
+are \code{"numeric"}, \code{"class"}, \code{"prob"}, \code{"conf_int"}, \code{"pred_int"},
+\code{"quantile"}, \code{"time"}, \code{"hazard"}, \code{"survival"}, or \code{"raw"}. When \code{NULL},
+\code{predict()} will choose an appropriate value based on the model's mode.}
 
 \item{std_err}{Attach column for standard error of prediction or not.}
 }
diff --git a/man/other_predict.Rd b/man/other_predict.Rd
index ee49f70d1..04442d2da 100644
--- a/man/other_predict.Rd
+++ b/man/other_predict.Rd
@@ -53,28 +53,29 @@ predict_survival(object, ...)
 predict_time(object, ...)
 }
 \arguments{
-\item{object}{An object of class \code{model_fit}}
+\item{object}{An object of class \code{model_fit}.}
 
 \item{new_data}{A rectangular data object, such as a data frame.}
 
-\item{...}{Arguments to the underlying model's prediction
-function cannot be passed here (see \code{opts}). There are some
-\code{parsnip} related options that can be passed, depending on the
-value of \code{type}. Possible arguments are:
+\item{...}{Additional \code{parsnip}-related options, depending on the
+value of \code{type}. Arguments to the underlying model's prediction
+function cannot be passed here (use the \code{opts} argument instead).
+Possible arguments are:
 \itemize{
-\item \code{interval}: for \code{type}s of "survival" and "quantile", should
+\item \code{interval}: for \code{type} equal to \code{"survival"} or \code{"quantile"}, should
 interval estimates be added, if available? Options are \code{"none"}
 and \code{"confidence"}.
-\item \code{level}: for \code{type}s of "conf_int", "pred_int", and "survival"
+\item \code{level}: for \code{type} equal to \code{"conf_int"}, \code{"pred_int"}, or \code{"survival"},
 this is the parameter for the tail area of the intervals
 (e.g. confidence level for confidence intervals).
-Default value is 0.95.
-\item \code{std_error}: add the standard error of fit or prediction (on
-the scale of the linear predictors) for \code{type}s of "conf_int"
-and "pred_int". Default value is \code{FALSE}.
-\item \code{quantile}: the quantile(s) for quantile regression
-(not implemented yet)
-\item \code{time}: the time(s) for hazard and survival probability estimates.
+Default value is \code{0.95}.
+\item \code{std_error}: for \code{type} equal to \code{"conf_int"} or \code{"pred_int"}, add
+the standard error of fit or prediction (on the scale of the
+linear predictors). Default value is \code{FALSE}.
+\item \code{quantile}: for \code{type} equal to \code{quantile}, the quantiles of the
+distribution. Default is \code{(1:9)/10}.
+\item \code{time}: for \code{type} equal to \code{"survival"} or \code{"hazard"}, the
+time points at which the survival probability or hazard is estimated.
 }}
 
 \item{level}{A single numeric value between zero and one for the
diff --git a/man/predict.model_fit.Rd b/man/predict.model_fit.Rd
index 4632bb205..df30a6f12 100644
--- a/man/predict.model_fit.Rd
+++ b/man/predict.model_fit.Rd
@@ -13,74 +13,80 @@
 predict_raw(object, ...)
 }
 \arguments{
-\item{object}{An object of class \code{model_fit}}
+\item{object}{An object of class \code{model_fit}.}
 
 \item{new_data}{A rectangular data object, such as a data frame.}
 
 \item{type}{A single character value or \code{NULL}. Possible values
-are "numeric", "class", "prob", "conf_int", "pred_int", "quantile", "time",
-"hazard", "survival", or "raw". When \code{NULL}, \code{predict()} will choose an
-appropriate value based on the model's mode.}
+are \code{"numeric"}, \code{"class"}, \code{"prob"}, \code{"conf_int"}, \code{"pred_int"},
+\code{"quantile"}, \code{"time"}, \code{"hazard"}, \code{"survival"}, or \code{"raw"}. When \code{NULL},
+\code{predict()} will choose an appropriate value based on the model's mode.}
 
 \item{opts}{A list of optional arguments to the underlying
 predict function that will be used when \code{type = "raw"}. The
 list should not include options for the model object or the
 new data being predicted.}
 
-\item{...}{Arguments to the underlying model's prediction
-function cannot be passed here (see \code{opts}). There are some
-\code{parsnip} related options that can be passed, depending on the
-value of \code{type}. Possible arguments are:
+\item{...}{Additional \code{parsnip}-related options, depending on the
+value of \code{type}. Arguments to the underlying model's prediction
+function cannot be passed here (use the \code{opts} argument instead).
+Possible arguments are:
 \itemize{
-\item \code{interval}: for \code{type}s of "survival" and "quantile", should
+\item \code{interval}: for \code{type} equal to \code{"survival"} or \code{"quantile"}, should
 interval estimates be added, if available? Options are \code{"none"}
 and \code{"confidence"}.
-\item \code{level}: for \code{type}s of "conf_int", "pred_int", and "survival"
+\item \code{level}: for \code{type} equal to \code{"conf_int"}, \code{"pred_int"}, or \code{"survival"},
 this is the parameter for the tail area of the intervals
 (e.g. confidence level for confidence intervals).
-Default value is 0.95.
-\item \code{std_error}: add the standard error of fit or prediction (on
-the scale of the linear predictors) for \code{type}s of "conf_int"
-and "pred_int". Default value is \code{FALSE}.
-\item \code{quantile}: the quantile(s) for quantile regression
-(not implemented yet)
-\item \code{time}: the time(s) for hazard and survival probability estimates.
+Default value is \code{0.95}.
+\item \code{std_error}: for \code{type} equal to \code{"conf_int"} or \code{"pred_int"}, add
+the standard error of fit or prediction (on the scale of the
+linear predictors). Default value is \code{FALSE}.
+\item \code{quantile}: for \code{type} equal to \code{quantile}, the quantiles of the
+distribution. Default is \code{(1:9)/10}.
+\item \code{time}: for \code{type} equal to \code{"survival"} or \code{"hazard"}, the
+time points at which the survival probability or hazard is estimated.
 }}
 }
 \value{
-With the exception of \code{type = "raw"}, the results of
-\code{predict.model_fit()} will be a tibble as many rows in the output
-as there are rows in \code{new_data} and the column names will be
-predictable.
+With the exception of \code{type = "raw"}, the result of
+\code{predict.model_fit()}
+\itemize{
+\item is a tibble
+\item has as many rows as there are rows in \code{new_data}
+\item has standardized column names, see below:
+}
+
+For \code{type = "numeric"}, the tibble has a \code{.pred} column for a single
+outcome and \code{.pred_Yname} columns for a multivariate outcome.
 
-For numeric results with a single outcome, the tibble will have
-a \code{.pred} column and \code{.pred_Yname} for multivariate results.
+For \code{type = "class"}, the tibble has a \code{.pred_class} column.
 
-For hard class predictions, the column is named \code{.pred_class}
-and, when \code{type = "prob"}, the columns are \code{.pred_classlevel}.
+For \code{type = "prob"}, the tibble has \code{.pred_classlevel} columns.
 
-\code{type = "conf_int"} and \code{type = "pred_int"} return tibbles with
-columns \code{.pred_lower} and \code{.pred_upper} with an attribute for
+For \code{type = "conf_int"} and \code{type = "pred_int"}, the tibble has
+\code{.pred_lower} and \code{.pred_upper} columns with an attribute for
 the confidence level. In the case where intervals can be
 produces for class probabilities (or other non-scalar outputs),
-the columns will be named \code{.pred_lower_classlevel} and so on.
+the columns are named \code{.pred_lower_classlevel} and so on.
 
-Quantile predictions return a tibble with a column \code{.pred}, which is
+For \code{type = "quantile"}, the tibble has a \code{.pred} column, which is
 a list-column. Each list element contains a tibble with columns
 \code{.pred} and \code{.quantile} (and perhaps other columns).
 
+For \code{type = "time"}, the tibble has a \code{.pred_time} column.
+
+For \code{type = "survival"}, the tibble has a \code{.pred} column, which is
+a list-column. Each list element contains a tibble with columns
+\code{.time} and \code{.pred_survival} (and perhaps other columns).
+
+For \code{type = "hazard"}, the tibble has a \code{.pred} column, which is
+a list-column. Each list element contains a tibble with columns
+\code{.time} and \code{.pred_hazard} (and perhaps other columns).
+
 Using \code{type = "raw"} with \code{predict.model_fit()} will return
 the unadulterated results of the prediction function.
 
-For censored regression:
-\itemize{
-\item \code{type = "time"} produces a column \code{.pred_time}.
-\item \code{type = "hazard"} results in a list column \code{.pred} containing tibbles
-with a column \code{.pred_hazard}.
-\item \code{type = "survival"} results in a list column \code{.pred} containing tibbles
-with a \code{.pred_survival} column.
-}
-
 In the case of Spark-based models, since table columns cannot
 contain dots, the same convention is used except 1) no dots
 appear in names and 2) vectors are never returned but
@@ -97,16 +103,12 @@ Apply a model to create different types of predictions.
 "type" argument for more specificity.
 }
 \details{
-If "type" is not supplied to \code{predict()}, then a choice
-is made:
+For \code{type = NULL}, \code{predict()} uses
 \itemize{
 \item \code{type = "numeric"} for regression models,
 \item \code{type = "class"} for classification, and
 \item \code{type = "time"} for censored regression.
 }
-
-\code{predict()} is designed to provide a tidy result (see "Value"
-section below) in a tibble output format.
 \subsection{Interval predictions}{
 
 When using \code{type = "conf_int"} and \code{type = "pred_int"}, the options