diff --git a/cmd/lotus-chainwatch/processor/reward.go b/cmd/lotus-chainwatch/processor/reward.go
index 3d5eaa7b4..5411cb68a 100644
--- a/cmd/lotus-chainwatch/processor/reward.go
+++ b/cmd/lotus-chainwatch/processor/reward.go
@@ -3,6 +3,7 @@ package processor
 import (
 	"bytes"
 	"context"
+	"github.com/filecoin-project/specs-actors/actors/util/smoothing"
 	"time"
 
 	"golang.org/x/sync/errgroup"
@@ -24,6 +25,8 @@ type rewardActorInfo struct {
 
 	// base reward in attofil for each block found during this epoch
 	baseBlockReward big.Int
+
+	epochSmoothingEstimate *smoothing.FilterEstimate
 }
 
 func (p *Processor) setupRewards() error {
@@ -53,6 +56,15 @@ create table if not exists chain_power
 			primary key,
 	baseline_power text not null
 );
+
+create table if not exists reward_smoothing_estimates
+(
+    state_root text not null
+        constraint reward_smoothing_estimates_pk
+        	primary key,
+	position_estimate text not null,
+	velocity_estimate text not null
+);
 `); err != nil {
 		return err
 	}
@@ -103,6 +115,7 @@ func (p *Processor) processRewardActors(ctx context.Context, rewardTips ActorTip
 
 			rw.baseBlockReward = rewardActorState.ThisEpochReward
 			rw.baselinePower = rewardActorState.ThisEpochBaselinePower
+			rw.epochSmoothingEstimate = rewardActorState.ThisEpochRewardSmoothed
 			out = append(out, rw)
 		}
 	}
@@ -162,6 +175,13 @@ func (p *Processor) persistRewardActors(ctx context.Context, rewards []rewardAct
 		return nil
 	})
 
+	grp.Go(func() error {
+		if err := p.storeRewardSmoothingEstimates(rewards); err != nil {
+			return err
+		}
+		return nil
+	})
+
 	return grp.Wait()
 }
 
@@ -243,3 +263,43 @@ func (p *Processor) storeBaseBlockReward(rewards []rewardActorInfo) error {
 
 	return nil
 }
+
+func (p *Processor) storeRewardSmoothingEstimates(rewards []rewardActorInfo) error {
+	tx, err := p.db.Begin()
+	if err != nil {
+		return xerrors.Errorf("begin reward_smoothing_estimates tx: %w", err)
+	}
+
+	if _, err := tx.Exec(`create temp table rse (like reward_smoothing_estimates) on commit drop`); err != nil {
+		return xerrors.Errorf("prep reward_smoothing_estimates: %w", err)
+	}
+
+	stmt, err := tx.Prepare(`copy rse (state_root, position_estimate, velocity_estimate) from stdin;`)
+	if err != nil {
+		return xerrors.Errorf("prepare tmp reward_smoothing_estimates: %w", err)
+	}
+
+	for _, rewardState := range rewards {
+		if _, err := stmt.Exec(
+			rewardState.common.stateroot.String(),
+			rewardState.epochSmoothingEstimate.PositionEstimate.String(),
+			rewardState.epochSmoothingEstimate.VelocityEstimate.String(),
+		); err != nil {
+			return xerrors.Errorf("failed to store smoothing estimate: %w", err)
+		}
+	}
+
+	if err := stmt.Close(); err != nil {
+		return xerrors.Errorf("close prepared reward_smoothing_estimates: %w", err)
+	}
+
+	if _, err := tx.Exec(`insert into reward_smoothing_estimates select * from rse on conflict do nothing`); err != nil {
+		return xerrors.Errorf("insert reward_smoothing_estimates from tmp: %w", err)
+	}
+
+	if err := tx.Commit(); err != nil {
+		return xerrors.Errorf("commit reward_smoothing_estimates tx: %w", err)
+	}
+
+	return nil
+}