Clutch Puppy Picks
Better Picks. Always Free.
For the community. No fake locks.
Back to dashboard
YesterdayLine moved
MLB

Philadelphia Phillies logoPHI@KCKansas City Royals logo

Philadelphia Philadelphia Phillies at Kansas City Kansas City Royals · 3:00 PM ET
Share X Reddit
Predicted winner
KC
Predicted final score
PHI 5.8 - KC 4.3
Sportsbook line
+117
Implied probability
44%
from market price
Model probability
42%
our estimate
Value gap
-1 pts
market sees more value
Confidence
Final recommendation
Pass0.25u
Model pick · KC +1.5 (-134)

Home no-vig implied moved from 44.5% to 43.6% (toward away). Steam detected across 3+ books in last 30 min.

"The most realistic loss path is Philadelphia’s stronger season record (50-39) and marginally better scoring profile (Avg 4.5 scored · 4.4 allowed) asserting itself, producing the model’s expected outcome (PHI 5.8 - KC 4.3)."

Clutch Puppy Expert Summary

Main prediction

KC +1.5 (-134) at sportsbook line +117. Model win prob 42.3% vs market implied 43.6% (value gap -1.2%), strength score 52% — this is a small negative-edge underdog play where the model still recommends the Royals on the hook because game dynamics and situational inputs reduce variance on the +1.5 line.

Best bet
KC +1.5 (-134) | Sportsbook line: +117
Projected final
PHI 5.8, KC 4.3
Odds & line movement

Opened 116/-136 and is now 117/-148; home no-vig implied moved from 44.5% to 43.6% (toward away). Steam detected across 3+ books in last 30 min, and the move has tightened the favorite's price from opening to current.

Key matchups & handicap
Philadelphia’s run profile vs KC’s run allowance

Philadelphia has the better season scoring baseline with Avg 4.5 scored · 4.4 allowed and a 50-39 record, while Kansas City sits at Avg 4.1 scored · 5.1 allowed with a 35-54 record; the model’s PHI 5.8 - KC 4.3 projection echoes that gap.

Market movement and timing

The book opened 116/-136 and the current line is 117/-148, and the home no-vig implied moved from 44.5% to 43.6%, showing the market has shifted 'toward away' since open — steam detected across 3+ books in last 30 min confirms timing of that move.

Model layer friction

Statistical_edge signal is -0.065 (weight 0.45) with contribution -0.029 and sharp_agreement signal is -0.194 (weight 0.2) with contribution -0.039; both are explicitly 'against pick' in the LAYER CONTRIBUTIONS, which is why this is a low-confidence, small-edge play.

Team trends and sample size

Kansas City’s season win rate is 39% (trend: cold) while Philadelphia’s season win rate is 56% (trend: neutral), which aligns with the model favoring Philadelphia in run expectation but not sufficiently to rule out the +1.5 hook for KC.

Top supporting factors
  • Model win prob 42.3% vs Market 43.6% (value gap -1.2%).
  • Recent form: PHI 50-39 (Avg 4.5 scored · 4.4 allowed), KC 35-54 (Avg 4.1 scored · 5.1 allowed).
  • Line moved from opened 116/-136 to current 117/-148; home no-vig implied moved from 44.5% to 43.6%.
  • Steam detected across 3+ books in last 30 min.
Betting trends
  • KC record 35-54
  • PHI record 50-39
  • KC Avg 4.1 scored · 5.1 allowed (season)
  • PHI Avg 4.5 scored · 4.4 allowed (season)
  • Opened 116/-136
  • Current 117/-148
  • Steam detected across 3+ books in last 30 min
PHI injuries & notes

No major injury notes detailed for this matchup.

KC injuries & notes

No major injury notes detailed for this matchup.

Team & game total picks

Best bet - KC +1.5 (-134) at sportsbook line +117 — buying the half-run against a team with a 50-39 record makes sense when the value gap is only -1.2% and situational inputs produce a zero contribution to the pick.

Total - Model projects PHI 5.8 - KC 4.3, which drives a controlled preference on the hook rather than an all-out total play.

Counterargument

The most realistic loss path is Philadelphia’s stronger season record (50-39) and marginally better scoring profile (Avg 4.5 scored · 4.4 allowed) asserting itself, producing the model’s expected outcome (PHI 5.8 - KC 4.3).

Injury concern

No reported injury impact.

Predicted score

The model’s PHI 5.8 - KC 4.3 prediction reflects season scoring baselines (PHI Avg 4.5 scored · 4.4 allowed; KC Avg 4.1 scored · 5.1 allowed) and a modest edge to Philadelphia on runs scored and prevention.

What this confidence rating means

Strength score 52% reflects a narrow edge: the model sees a -1.2% value gap between Model 42.3 and Market 43.6, so this is a low-confidence, modest-value situation rather than a high-conviction slam.

Final score prediction

Game plays like a close Philadelphia edge with Kansas City clinging to the +1.5; final score projection PHI 5.8 - KC 4.3.

Final recommendation

Play KC +1.5 (-134) at available +117 if you can get it; this is a small, situational value play where you're buying a half-run cover against a superior run-profile team.

How to bet this game

Shop the line — the ticket favors KC +1.5 at available +117 relative to the recommended pick KC +1.5 (-134). Watch the steam noted across 3+ books in last 30 min and the move from opened 116/-136 to current 117/-148; if you can get +117, take it, otherwise avoid paying extra juice.

Top supporting factors

  • Model win prob 42.3% vs Market 43.6% (value gap -1.2%).
  • Recent form: PHI 50-39 (Avg 4.5 scored · 4.4 allowed), KC 35-54 (Avg 4.1 scored · 5.1 allowed).
  • Line moved from opened 116/-136 to current 117/-148; home no-vig implied moved from 44.5% to 43.6%.
  • Steam detected across 3+ books in last 30 min.

Counterargument

The most realistic loss path is Philadelphia’s stronger season record (50-39) and marginally better scoring profile (Avg 4.5 scored · 4.4 allowed) asserting itself, producing the model’s expected outcome (PHI 5.8 - KC 4.3).

Injury impact

No reported impact.

Team status

  • KC · cold
    39% season win rate
  • PHI · neutral
    56% season win rate

Recent form (last 10)

  • KC
    Avg 4.1 scored · 5.1 allowed (season)
    35-54
  • PHI
    Avg 4.5 scored · 4.4 allowed (season)
    50-39

Historical trends

  • [away] Runs-allowed gap -0.70
    Historical comp
    -0.03
  • [home] Home-field baseline
    Historical comp
    +0.01

Line movement

Model vs market

Modelvs
Marketvs

Last updated 19d ago (10/3/2026, 5:16:43 PM)

Strength percentages reflect model-estimated value vs market - not a guarantee of outcome. A 100% rating means the model sees roughly double the value of a toss-up. For research and entertainment.