diff --git a/s2/encode_all.go b/s2/encode_all.go
index f1d27b6a8d..d5878d1821 100644
--- a/s2/encode_all.go
+++ b/s2/encode_all.go
@@ -12,16 +12,11 @@ import (
 )
 
 func load32(b []byte, i int) uint32 {
-	b = b[i:]
-	b = b[:4]
-	return uint32(b[0]) | uint32(b[1])<<8 | uint32(b[2])<<16 | uint32(b[3])<<24
+	return binary.LittleEndian.Uint32(b[i:])
 }
 
 func load64(b []byte, i int) uint64 {
-	b = b[i:]
-	b = b[:8]
-	return uint64(b[0]) | uint64(b[1])<<8 | uint64(b[2])<<16 | uint64(b[3])<<24 |
-		uint64(b[4])<<32 | uint64(b[5])<<40 | uint64(b[6])<<48 | uint64(b[7])<<56
+	return binary.LittleEndian.Uint64(b[i:])
 }
 
 // hash6 returns the hash of the lowest 6 bytes of u to fit in a hash table with h bits.
diff --git a/s2/encode_best.go b/s2/encode_best.go
index 4b571bf316..9a9587036c 100644
--- a/s2/encode_best.go
+++ b/s2/encode_best.go
@@ -90,16 +90,17 @@ func encodeBlockBest(dst, src []byte) (d int) {
 				if load32(src, offset) != first {
 					return match{offset: offset, s: s}
 				}
-				m := match{offset: offset, s: s, length: 4, rep: rep}
+				m := match{offset: offset, s: s, length: 4 + offset, rep: rep}
 				s += 4
 				for s <= sLimit {
-					if diff := load64(src, s) ^ load64(src, offset+m.length); diff != 0 {
+					if diff := load64(src, s) ^ load64(src, m.length); diff != 0 {
 						m.length += bits.TrailingZeros64(diff) >> 3
 						break
 					}
 					s += 8
 					m.length += 8
 				}
+				m.length -= offset
 				return m
 			}
 
@@ -150,7 +151,7 @@ func encodeBlockBest(dst, src []byte) (d int) {
 					best = bestOf(best, matchAt(s-repeat+1, s+1, uint32(cv>>8), true))
 
 					// s+2
-					if true {
+					if best.length < 100 {
 						nextShort = sTable[hash4(cv>>8, sTableBits)]
 						s++
 						cv = load64(src, s)