Skip to article
The Sporting Page Real Analysis. Not Corporate

MLB · Pitching Data

Gaddis reduced his slider share and found seven misses in Game 5

Hunter Gaddis used more fastballs and changeups than in his season relief sample. Four strikeouts and two softly measured singles complicate the same 24-pitch outing.

Hunter Gaddis during White Sox–Guardians ALDS Game 5 at Progressive Field on October 10, 2026
Hunter Gaddis during White Sox–Guardians ALDS Game 5 at Progressive Field on October 10, 2026. · Credit: MLB.com; photographer not identified in the source
IP1.1
H2
R / ER0 / 0
BB0
K4
Pitches24

Chase Meidroth fouled a bunt attempt against Hunter Gaddis’ 94.3 mph fastball in the seventh inning. Gaddis then threw three changeups: a 78.2 mph swinging strike, an 80.0 mph ball and a 78.8 mph foul tip that completed the strikeout. The right-handed hitter’s plate appearance showed the changeup contributing beyond a conventional left-handed matchup.

Gaddis struck out four White Sox over 1⅓ innings in Cleveland’s 2–0 ALDS Game 5 win at Progressive Field on October 10. He allowed two singles without a walk and threw 24 pitches. He completed the seventh and recorded the first out of the eighth before Cade Smith took over with a runner aboard.

Compared with his regular-season relief work, Gaddis threw fewer sliders and more fastballs and changeups. The changed shares accompanied seven misses, but did not prevent all contact from becoming a hit. Both balls put into play against him became singles despite recorded exit speeds below 77 mph.

Pitch selection

Season repertoire and Game 5

Scale: 0–100%. Gray: regular season. Blue: Game 5.

Four-seam fastball

Season
29.5% (290)
Game 5
41.7% (10)

Change: +12.2 percentage points.

Changeup

Season
19.9% (196)
Game 5
33.3% (8)

Change: +13.4 percentage points.

Slider

Season
49.8% (490)
Game 5
25.0% (6)

Change: -24.8 percentage points.

Splitter

Season
0.7% (7)
Game 5
0.0% (0)

Change: -0.7 percentage points.

How did selection change? Game 5: 24 physical pitches. Season: 983 pitches in 68 relief appearances for Cleveland Guardians, 2026-04-13 through 2026-09-25. Every bar starts at zero and uses the same 0–100% scale. Counts appear in parentheses, and unused season classifications remain at zero. Source: Baseball Savant gamefeed and Statcast Search. One outing and a different hitter sample limit interpretation.

Pitch selection SVG

The changeup supplied a different route to strikes

The slider fell from 49.8% of 983 pitches across 68 regular-season relief appearances to six of 24 pitches, or 25.0%, in Game 5. Four-seamers rose from 29.5% to 41.7%, and changeups from 19.9% to 33.3%. The slider still finished two strikeouts, so its reduced share did not make it incidental to the result.

The changeup produced three misses in four attempts, against two in seven for the fastball and two in three for the slider. The entire outing had seven misses in 14 physical attempts, a 50.0% whiff rate against 29.5% in the baseline. That denominator retains Meidroth’s foul bunt. Savant’s summary lists 13 swings without the bunt, which would produce 53.8%; both methods count seven misses.

His changeup averaged 79.3 mph, 15.3 mph below the outing’s four-seam average. The Meidroth sequence exposed that large speed difference immediately after the foul bunt. Three of Gaddis’ five changeups to right-handed hitters came in this plate appearance. That concentrated sequence is useful evidence, but also shows how a single confrontation can influence a short outing’s pitch mix.

Across the outing, he threw 15 pitches to left-handed hitters and nine to right-handed hitters. His regular-season sample was more evenly divided. Yet the increased changeup share was concentrated against right-handers: five of nine pitches, compared with 49 of 484 in the season comparison. Against left-handers, changeups were three of 15, below their baseline share of 147 of 499. The overall increase therefore conceals different choices within the two hitter groups.

A revealing confrontation

Chase Meidroth: inning 7

2026-10-11T02:09:09.862872 image/svg+xml Matplotlib v3.11.2, https://matplotlib.org/ −2 −1 0 1 2 Horizontal plate location (ft) 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Height at plate (ft) Catcher view 1 2 3 4 2026-10-11T02:09:09.962209 image/svg+xml Matplotlib v3.11.2, https://matplotlib.org/ −2 −1 0 1 2 Horizontal plate location (ft) 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Height at plate (ft) Catcher view 1 2 3 4
How did the plate appearance develop? Outing pitches 8–11. Chase Meidroth strikes out on a foul tip. The 17-inch-wide rectangle uses the first pitch’s recorded vertical boundaries, 1.548–3.068 ft. Markers retain exact coordinates and are not ball size. Overlaps can be resolved with the numbered sequence list and detail buttons. Source: Baseball Savant locations and official MLB final calls.

At-bat diagram SVG

The singles still counted

Tristan Peters singled in the seventh on a slider measured at 76.8 mph off the bat. Sam Antonacci’s eighth-inning single came on a changeup at 66.3 mph. Neither met the 95 mph hard-hit threshold. These measurements describe limited contact speed; they cannot establish that Gaddis intended those outcomes or that similar contact will reliably turn into outs.

The first single was followed by two strikeouts to complete the seventh. In the eighth, Gaddis struck out Randal Grichuk before Antonacci’s single ended his appearance. Smith then struck out Kyle Teel and retired Miguel Vargas, stranding the inherited runner. That resolution belongs in the account of Gaddis’ scoreless line.

Gaddis threw 14 pitches in the seventh and 10 in his partial eighth. His four-seamer averaged 94.9 mph on six pitches in the seventh and 94.1 on four in the eighth. Those small samples do not diagnose fatigue, and his recorded speed does not supply a reason for the pitching change.

The outing supports a narrower conclusion than a permanent repertoire redesign: Gaddis obtained misses from all three pitch types while using the slider less often, and a right-handed changeup sequence influenced the totals. Another appearance would test whether that matchup choice recurs across more hitters. The two singles also remain a reminder that low measured exit speed does not guarantee an out.

Workload and speed

Inning samples in context

2026-10-11T02:09:10.076513 image/svg+xml Matplotlib v3.11.2, https://matplotlib.org/ 7 8* Inning (* partial work) 0 10 20 30 40 Pitches 14 24 Cumulative physical pitches 7 8* Inning (* partial work) 90 92 94 96 98 100 102 mph (90–102 scale) 94.9 n=6 94.0 n=4 Mean four-seam speed 2026-10-11T02:09:10.227355 image/svg+xml Matplotlib v3.11.2, https://matplotlib.org/ 7 8* Inning (* partial work) 0 10 20 30 40 Pitches 14 24 Cumulative physical pitches 7 8* Inning (* partial work) 90 92 94 96 98 100 102 mph (90–102 scale) 94.9 n=6 94.0 n=4 Mean four-seam speed
Workload and measured four-seam samples
InningPitchesCumulativeFour-seam mphFour-seam nOuts
7141494.963
8*102494.041
How much work did each inning require? Cumulative pitches start at zero. Four-seam velocity uses the same explicitly truncated 90–102 mph scale across all seven papers; n counts measured four-seamers. The panels have separate axes and do not imply workload causes a speed change. Partial work: inning 8: entered with 0 out(s), recorded 1. Source: official MLB sequence and Baseball Savant.

Workload SVG

Complete physical pitch record

Every pitch: 24 of 24

Search a hitter, select an inning or pitch type, then inspect a pitch number. Pitch result and completed plate-appearance outcome are separate columns. The table scrolls within its border.

24 of 24 pitches

Hunter Gaddis: ALDS Game 5 physical pitches
PitchInningHitterCount beforeTypemphPitch resultCompleted PA
7Andrew Benintendi0–0CH78.6Called StrikeStrikeout
7Andrew Benintendi0–1CH79.8BallStrikeout
7Andrew Benintendi1–1FF95.0Swinging StrikeStrikeout
7Andrew Benintendi1–2FF95.9Swinging StrikeStrikeout
7Tristan Peters0–0FF95.2BallSingle
7Tristan Peters1–0SL89.4Swinging StrikeSingle
7Tristan Peters1–1SL88.4In play, no outSingle
7Chase Meidroth0–0FF94.3Foul BuntStrikeout
7Chase Meidroth0–1CH78.2Swinging StrikeStrikeout
7Chase Meidroth0–2CH80.0BallStrikeout
7Chase Meidroth1–2CH78.8Foul TipStrikeout
7Colson Montgomery0–0FF95.0Called StrikeStrikeout
7Colson Montgomery0–1FF94.0FoulStrikeout
7Colson Montgomery0–2SL88.4Called StrikeStrikeout
8Randal Grichuk0–0CH78.7Swinging StrikeStrikeout
8Randal Grichuk0–1CH80.3BallStrikeout
8Randal Grichuk1–1FF94.2FoulStrikeout
8Randal Grichuk1–2FF94.0FoulStrikeout
8Randal Grichuk1–2SL88.7Swinging StrikeStrikeout
8Sam Antonacci0–0SL87.6BallSingle
8Sam Antonacci1–0SL87.0BallSingle
8Sam Antonacci2–0FF94.4Called StrikeSingle
8Sam Antonacci2–1FF93.6FoulSingle
8Sam Antonacci2–2CH79.9In play, no outSingle
Select a pitch number to inspect speed, movement, location and contact.
Download all 24 pitches as CSV
Sources, definitions and limitations

The official final MLB record supplies the score, events, calls and pitching line. Baseball Savant’s complete Game 5 feed supplies all 24 tracked physical pitches. They reconcile by unique ID, order, type, final call and pre-pitch count, including count-changing non-pitch events and challenge decisions. The audit also reconciles 6 batters, 4 integer outs and 4 strikeouts. In baseball innings notation, .1 means one out and .2 means two. Final game and data checked October 11, 2026, before publication.

The Statcast Search baseline is 983 pitches in 68 regular-season relief appearances for Cleveland Guardians, 2026-04-13 through 2026-09-25. Each game’s pitch total reconciles to the official appearance log. Other roles, other clubs and postseason work are excluded from this baseline. 1 baseline pitches lack start speed; excluded from that measure. 1 baseline pitches lack spin rate; excluded from that measure. 1 baseline pitches lack extension; excluded from that measure.

Whiff rate means misses divided by physical swing attempts. Swinging strikes, blocked swinging strikes, caught foul tips and missed bunts count as misses. Fouls, balls put into play and bunt attempts enter the attempt denominator. Called strikes and completed strikeout plate appearances are separate. The physical-attempt denominator includes one foul bunt: 14 attempts versus 13 swings in Savant’s summary. Misses reconcile under both definitions. Chase rate uses ball-center coordinates outside a 17-inch-wide rectangle with each pitch’s recorded vertical zone, divided by pitches outside that rectangle with complete coordinates. Savant’s separate zone flag is retained in the detail and CSV; it is not substituted for this geometric definition.

Speed is mph, spin rpm, extension feet and movement inches. Induced vertical break is relative to a gravity-only trajectory. Plate x/z and zone boundaries are feet in catcher view. Charts preserve exact plate coordinates; markers are not ball size and cannot reveal intended targets. See Statcast field definitions. All outing pitches have the displayed tracking measurements. 2 of 2 balls put into play have measured exit speed. Hard-hit means at least 95 mph; the baseline has 174 measured contacts out of 175. Expected batting average in the details is a Savant model estimate, rather than an actual outcome. Missing values remain missing. These descriptive comparisons do not establish intent, fatigue, injury or a lasting trend.

Sources and notes

  1. Official MLB final Game 849831: https://statsapi.mlb.com/api/v1.1/game/849831/feed/live
  2. Baseball Savant complete tracking: https://baseballsavant.mlb.com/gf?game_pk=849831
  3. Regular-season comparison: 2026-04-13 through 2026-09-25, same club and role.
  4. Statcast field definitions: https://baseballsavant.mlb.com/csv-docs