Skip to main content
MLB Deserve-to-Win
← All pitchers·All hitters

Brad Keller

2026
Phillies· RHP · throws R· 31.1 IP· 4.02 ERA· 1.34 WHIP· 135 batters faced
Compare

Grade

His grade shows how good he has been, and at what.
Bbetter than 60% of qualified pitchers at limiting bases
RelieverCommand Artist

Rarely walks anyone and lets the defense work.

Keller sits right around league average

The simulator expects him to allow 0.447 bases per batter faced, better than 60% of qualified pitchers. His best skill is limiting hard contact (81st percentile). Short relief outings push strikeout rates up a little, so read his K% spoke with that in mind.

His grade rests on 135 batters faced, with data through 2026-09-27.

How to read this chart
Each spoke is a skill estimate adjusted for sample size, not a raw season stat, and farther out is better. A ↓ marks a stat where a lower number is better, and its spoke is flipped so farther out still means better. The middle ring marks the league median and the outer ring the league best. Percentiles compare him with 517 qualified pitchers, those with 100+ batters faced, starters and relievers together.

Plays like…

Each of these pitchers shares his archetype and sits closest to him in xEB/PA. Pick one to see him side by side with Brad Keller.

Hunter GaddisGuardiansTroy MeltonTigersTrevor McDonaldGiantsJosé UrquidyPhillies

Pitcher type

This map shows what kind of pitcher he is, next to the league.

Pitcher type

He lands in the “Misses bats, limits hard contact” corner of the map.

League best:K% − BB% Jacob Misiorowski (26.8%)Hard-hit% allowed Blake Snell (29.0%)

How to read this chart
Two skills separate pitching styles: strikeout rate minus walk rate, where higher is better, and hard-hit rate allowed, where lower is better. The hard-hit axis is flipped, so the best corner is the top right. Each dot is a pitcher with 80+ batters faced this season, sized by batters faced. The dashed lines are the league median on each skill, and the league best on each is named above the chart. Click any dot to open that pitcher's page.
 Brad KellerLeague medianPercentile
Bases per ball in play(lower = better).436.527—
Walk rate(lower = better)10%9%41st
Strikeout rate(higher = better)22%22%32nd

A higher percentile is better, and the 99th is the top of the league.

Arsenal

These cards show what he throws and what each pitch does.

Brad Keller mixes his four-seam fastball (32%) and slider (23%). His best pitch has been the slider, saving 6.3 bases per 100 thrown against league average.

Pitch mix & swing-and-miss

+19 bases better than a replacement arm this season

He allowed about 51 estimated bases, against a replacement arm's 70 over the same batters faced.

Replacement level is per-batter and can't be split by pitch type, so the bars below use league average.

All pitches

He saved 8.1 bases against league average across 533 pitches, or 1.5 per 100 pitches.

Four-Seam Fastball
32%
97.0 mph · RHP median 95.1
21% whiffs · Lg median 21%
cost 1.7 per 1003.0 total
Slider
23%
87.1 mph · RHP median 86.9
25% whiffs · Lg median 33%
saved 6.3 per 1007.6 total
Sinker
20%
96.7 mph · RHP median 94.7
15% whiffs · Lg median 13%
saved 3.8 per 1004.0 total
Changeup
13%
92.8 mph · RHP median 87.1
32% whiffs · Lg median 30%
saved 4.7 per 1003.3 total
Sweeper
12%
85.7 mph · RHP median 83.5
24% whiffs · Lg median 31%
cost 5.7 per 1003.8 total
How to read this chart
Each row is a pitch he throws at least 3% of the time, from pitch tracking in the regular season. The bar is how often he throws it, then its average velocity and its whiff rate, the share of swings that miss. Beside the velocity and the whiff rate is the median for that pitch, among pitchers of his hand for velocity and across the league for whiffs. More whiffs are better for a pitcher, while usage and velocity describe how he pitches. The Bases vs average bars score each pitch against a league-average pitch in the same count, and a bar to the left (saved) is better.

Mix by batter side

Compared with the league, his four-seam fastball tilts 25 points more toward lefties.

Four-Seam Fastball
47% vs LHH · RHP median 40%
16% vs RHH · RHP median 33%
Slider
22% vs LHH · RHP median 20%
23% vs RHH · RHP median 30%
How to read this chart
Each row shows how often he throws a pitch to left-handed hitters (LHH) and right-handed hitters (RHH), next to the median for pitchers of his hand (RHP median). Medians come from right-handed pitchers who throw that pitch at least 3% of the time. Between 217 and 384 of them qualify, depending on the pitch. A mix is a plan, so a higher or lower share is neither better nor worse. Overall results by batter side are under Platoon splits.

Brad Keller's pitch movement, 2026

His Changeup and Sweeper have moved 2+ inches from their early-season shapes.

The verdict above compares each pitch with his own early-season outings in 2026, not with other pitchers, after taking out the league-wide drift in break.

Brad Keller's pitch movement, 2026-12012-12012in← glove sidearm side →more risemore drop
Four-Seam FastballFF
Rise 13.3 in
−2.4 vs Lg median
Break −0.3 in
−8.0 vs Lg median
Vs. early season steady
SliderSL
Rise −5.7 in
−7.5 vs Lg median
Break −5.9 in
−2.2 vs Lg median
Vs. early season steady
SinkerSI
Rise 11.0 in
+2.7 vs Lg median
Break 13.6 in
−1.6 vs Lg median
Vs. early season steady
ChangeupCH
Rise 6.6 in
+2.5 vs Lg median
Break 10.9 in
−3.3 vs Lg median
Vs. early season changed
SweeperST
Rise 1.4 in
+0.3 vs Lg median
Break −16.7 in
−2.9 vs Lg median
Vs. early season changed
Early season → nowLg medianNormal wobble around the dot
How to read this chart
Each dot is a pitch's current break in inches, averaged over his latest outings. Higher is more rise and further right is more arm-side break, and neither is better or worse. The faint trail runs from his early-season outings to the dot, and the soft halo shows how far normal noise can move it. League median is the small hollow ring, and a pitch marked * has none because too few pitchers throw it. The “changed” verdict removes a league-wide drift that the trail still shows, so a long trail can read as steady. The by-appearance chart below has the exact dates. More on the About page

By appearance

These charts follow his velocity, movement and mix from outing to outing.

These charts follow his velocity, movement and pitch mix one outing at a time, and each pitch keeps the same color in every chart. The velocity and movement charts also shade the middle of the league for scale.

Platoon splits

This section splits his results against lefties and righties.

Platoon splits (EB/PA allowed)

He handles righty bats better than lefty bats, by 0.041 EB/PA allowed.

50% probable range: +0.033 to +0.049

vs LHB66 PA0.428

vs LHB: shrunk estimate 0.428 EB/PA, 50% probable range 0.407 to 0.449, based on 66 plate appearances. Raw (unshrunk) rate: 0.347 EB/PA. League median vs LHB: 0.451 EB/PA.

His raw rate vs LHB (0.347) is off this scale. Small samples can push a raw rate this far, which is why the dot, the shrunk estimate, is the number to trust.

vs RHB69 PA0.387

vs RHB: shrunk estimate 0.387 EB/PA, 50% probable range 0.366 to 0.408, based on 69 plate appearances. Raw (unshrunk) rate: 0.436 EB/PA. League median vs RHB: 0.435 EB/PA.

0.3560.3820.4090.4350.461
← betterworse →
Below league medianAbove50% probable rangeHis raw rateMiddle half of the league

His split is about typical for a righty arm (median 0.042). He allows less than the league median against both sides, so even his weaker side is a strength.

How to read this chart
  • Each dot is his estimate against that side, with its 50% probable range as the bar. These are bases allowed, so lower is better.
  • The diamond is his raw rate, which small samples make noisy, so the simulator pulls the dot toward a more trustworthy value.
  • The dashed tick is the league median, and the shaded strip is the middle half of pitchers with 100 or more batters faced.
  • The two splits average back to his modeled overall EB/PA allowed, not the xEB/PA headline.
  • Tap or hover a mark for its number. LHB and RHB are left- and right-handed batters. More on the About page

Contact allowed

These cards show what hitters do when they make contact.

Batters go up the middle off him most often, and that's also where they do the most damage.

Pull side and opposite field depend on which side each batter hits from, so this pools every hitter he faced.

Batted balls allowed

Every ball hit off Brad Keller this season is drawn where it landed.
hit · out
EBlowhigh
vs Sal StewartHome run
101 mph · 27° · worth 3.7 bases
05/18 vs Reds
vs Ramón LaureanoHome run
103 mph · 28° · worth 3.3 bases
05/26 vs Padres
vs Troy JohnstonDouble
104 mph · 14° · worth 2.0 bases
05/08 vs Rockies
vs Seiya SuzukiHome run
104 mph · 39° · worth 1.8 bases
04/23 vs Cubs
vs Otto LopezSingle
106 mph · 12° · worth 1.3 bases
05/01 vs Marlins

Luck check

How to read this chart
Dot color is estimated bases (EB), what the contact was worth no matter what the defense did with it. Darker means more bases, so a dark hollow ring in the outfield is a crushed ball his defense ran down. Tap or hover a dot or a row to link the two, and use the ▶ links to watch a play on Baseball Savant.

Deserved vs. actual

Batters have been credited with about 5 more bases than their contact off Brad Keller deserved, so his results have been unlucky so far.

Deserved by the contact he allowed38
Actually credited to batters43

Both bars count 87 balls in play, with walks and strikeouts left out.

How to read this chart
Every ball in play is worth the bases its exit velocity, launch angle and direction usually produce. If batters were credited with more than that, his results have been unlucky, and fewer means lucky. BABIP against is the batting average on balls in play he allows, next to the league median.

ERA 4.02 · WHIP 1.34 · BABIP against 0.301 (Lg median 0.286)

Times through the order

These charts show what happens as hitters see him a second and third time.

Best games

These are his biggest outings this season.

Best games

These are his best outings this season, ranked by bases saved against a replacement arm.

Multi-season trends · Brad Keller

Compared with last season, his HR allowed% is up, his BB% is up, his K% is down, his Barrel% against is down, his GB% against is down, his FB% against is up and his Pull% against is down.

Actual rates show what happened on the field, season by season.