WTDS: TAMPA BAY RAYS @ NEW YORK YANKEES — FULL ANALYSIS

Part 1 RAYS @ YANKEES — FULL ANALYSIS [Part 1/5]
Sunday May 24, 2026 — 1:35 PM ET, Yankee Stadium
D. Rasmussen (R) vs R. Weathers (L)

CURRENT MARKET
Moneyline: Rays +120 / Yankees -132
Run Line: Rays +1.5 (-185) / Yankees -1.5 (+161)
Total: O7 (-103) / U7 (-117)
F5 ML: Rays +115 / Yankees -135

Pinnacle reference: Rays +120 / Yankees -130
Implied: Rays 45.5%, Yankees 56.5%, vig ~2%

==========================================
TEAM QUALITY — THE FUNDAMENTAL LAYER
==========================================

Rays Yankees
Record: 34-16 (.680) 30-22 (.577)
Runs scored: 236 252
Runs allowed: 194 187
Run differential: +42 +65
PYTHAGOREAN WIN%: 58.9% 63.3%
Luck (W% - Pythag): +9.1 (lucky) -5.6 (unlucky)

This is the core of the read. The Rays' record looks dominant, but their
+42 run differential implies a 59% team that has OVERPERFORMED. The
Yankees' middling record is masking a +65 differential — a 63% team
that has UNDERPERFORMED.

By underlying quality, the Yankees are the better team by ~4.4
percentage points. The market price of Yankees -132 (56.5% implied) is
actually slightly below their true probability per Pythag.

Rays are +9.1 luck-points above their expected record. That's the kind
of overperformance that regresses toward Pythag as the season goes on.
Yankees are -5.6 below. Both teams are due for regression in the
direction that closes this gap.

If you only read one section of this analysis, it's this one. Continue
to Part 2 for the SOS / head-to-head / recent form context.

RAYS @ YANKEES — FULL ANALYSIS [Part 2/5]

==========================================
STRENGTH OF SCHEDULE — WHERE THE WINS CAME FROM
==========================================

RAYS' 5-game winning streak coming in:
5/22 W vs Yankees (only quality opp)
5/20 W vs Baltimore (BAD tier, 22-29)
5/19 W vs Baltimore
5/18 W vs Baltimore (swept slumping O's)
5/17 W vs Marlins (BAD tier, 23-29)
5/16 L vs Marlins

YANKEES' 3-game losing streak coming in:
5/22 L vs Tampa Bay (GOOD tier)
5/21 L vs Toronto
5/20 L vs Toronto

Yankees lost three games to two playoff-caliber opponents. Rays won
four against the bottom of the AL.

SOS-adjusted reads:
- Rays' "hot streak" is mostly schedule-driven
- Yankees' "cold streak" was against tough competition
- The momentum gap is significantly smaller than the raw streaks suggest

==========================================
HEAD-TO-HEAD — YANKEES vs RAYS 2026
==========================================

Yankees: 0W-4L vs Tampa Bay this season

Date Location Result Yankees Streak Coming In
4/10 Away L L2
4/11 Away L L3
4/12 Away L L4 (Tampa swept this series)
5/22 Home L L2

The H2H is striking but the sample is small. Three of the four losses
came in a single April sweep. Tampa Bay has owned this matchup — fact
— but four games is not a basis to project a 100% Tampa rate going
forward. Regression suggests Yankees win one of the next two or three
meetings just on talent normalization.

Tonight is meeting #5. Two more remain in early July (four-game series).

==========================================
RECENT FORM (QUALIFIED)
==========================================

Rays L6: 5-1. Built mostly against BAD-tier opposition. Adjusted for
SOS, true performance level over the stretch is closer to Pythag
(58.9%) than the raw 83% record.

Yankees L5: 2-3. All three losses against playoff teams. Adjusted for
SOS, closer to Pythag (63.3%) than the 40% recent record.

NOTE: Saturday 5/23 game was POSTPONED. Yankees did not extend the
streak. They come in on L3, NOT L4.

Continue to Part 3 for the pitcher matchup breakdown.

RAYS @ YANKEES — FULL ANALYSIS [Part 2/5]

==========================================
STRENGTH OF SCHEDULE — WHERE THE WINS CAME FROM
==========================================

RAYS' 5-game winning streak coming in:
5/22 W vs Yankees (only quality opp)
5/20 W vs Baltimore (BAD tier, 22-29)
5/19 W vs Baltimore
5/18 W vs Baltimore (swept slumping O's)
5/17 W vs Marlins (BAD tier, 23-29)
5/16 L vs Marlins

YANKEES' 3-game losing streak coming in:
5/22 L vs Tampa Bay (GOOD tier)
5/21 L vs Toronto
5/20 L vs Toronto

Yankees lost three games to two playoff-caliber opponents. Rays won
four against the bottom of the AL.

SOS-adjusted reads:
- Rays' "hot streak" is mostly schedule-driven
- Yankees' "cold streak" was against tough competition
- The momentum gap is significantly smaller than the raw streaks suggest

==========================================
HEAD-TO-HEAD — YANKEES vs RAYS 2026
==========================================

Yankees: 0W-4L vs Tampa Bay this season

Date Location Result Yankees Streak Coming In
4/10 Away L L2
4/11 Away L L3
4/12 Away L L4 (Tampa swept this series)
5/22 Home L L2

The H2H is striking but the sample is small. Three of the four losses
came in a single April sweep. Tampa Bay has owned this matchup — fact
— but four games is not a basis to project a 100% Tampa rate going
forward. Regression suggests Yankees win one of the next two or three
meetings just on talent normalization.

Tonight is meeting #5. Two more remain in early July (four-game series).

==========================================
RECENT FORM (QUALIFIED)
==========================================

Rays L6: 5-1. Built mostly against BAD-tier opposition. Adjusted for
SOS, true performance level over the stretch is closer to Pythag
(58.9%) than the raw 83% record.

Yankees L5: 2-3. All three losses against playoff teams. Adjusted for
SOS, closer to Pythag (63.3%) than the 40% recent record.

NOTE: Saturday 5/23 game was POSTPONED. Yankees did not extend the
streak. They come in on L3, NOT L4.

Continue to Part 3 for the pitcher matchup breakdown.

RAYS @ YANKEES — FULL ANALYSIS [Part 4/5]

==========================================
DNA FRAMEWORK SIGNALS
==========================================

YANKEES COLD + GOOD OPPONENT (historical 2026):
L2 streak vs GOOD: 0-3 (fade rate 100%)
L3 streak vs GOOD: 0-1 (fade rate 100%)
L4 streak vs GOOD: 0-1 (fade rate 100%)

Combined: Yankees on 2+ game L streak vs GOOD opp: 0-5 in 2026.

HONEST QUALIFIER: 4 of these 5 losses came specifically against Tampa
Bay. The pattern is essentially "Yankees cold + Tampa Bay = loss." It
hasn't been tested at scale against other GOOD opponents.

==========================================
ML MODEL PICKS
==========================================

Model Pick Confidence
ML-LogReg Rays +118 2/5
ML-XGBoost Rays +118 3/5
ML-Calibrated Rays +118 1/5
ML-RandomForest Yankees -130 2/5

3 of 4 models on Rays, but ALL at low confidence (1-3 out of 5).
ML-RandomForest dissents — picks the Yankees.

Tampa Bay team ML accuracy: 75.3% (highest in MLB across 81 picks).
Yankees team ML accuracy: 64.8% (89% when models fade them, n=18).

==========================================
THE COMPETING NARRATIVES
==========================================

BULL CASE FOR RAYS +120:
1. 0-4 H2H — Rays have genuinely owned this matchup
2. Yankees on 3-game cold streak
3. ML models lean Rays 3-of-4
4. Rays SP has slightly better season ERA/WHIP
5. Yankees compound-DNA fade has hit 5-of-5 historically
6. Rays on hot recent run

BEAR CASE FOR RAYS +120:
1. Yankees are the better team by Pythag (63.3% vs 58.9%)
2. Yankees are -5.6 unlucky (regression positive)
3. Rays are +9.1 lucky (regression negative)
4. Rays' "hot streak" was padded against bad teams (SOS-inflated)
5. Yankees' "cold streak" was against tough opp (SOS-deflated)
6. Weathers has elite K-rate (10.91) — competitive with Rasmussen
7. 0-4 H2H is small sample, prone to variance
8. Line moved toward Yankees (sharp money agrees with Pythag)
9. Most "signals" supporting Rays are correlated measures of the
same thing — recent form

THE KEY INSIGHT:

The nine bullish signals are not nine independent observations. They
all measure essentially the same underlying variable — short-term
form. Pythag is the independent measurement, and it disagrees with
the recent-form stack.

Continue to Part 5 for the fair value calc and verdict.

RAYS @ YANKEES — FULL ANALYSIS [Part 5/5]

==========================================
FAIR VALUE CALCULATION
==========================================

PYTHAG-ADJUSTED MATCHUP PROBABILITY (log5 method):

Yankees Pythag: 63.3%
Rays Pythag: 58.9%
League avg: 50.0%

P(Yankees) = (0.633 × (1-0.589)) / (0.633 × (1-0.589) + (1-0.633) × 0.589)
P(Yankees) = 0.547

Yankees true prob at neutral site ˜ 54.7%
Add ~3.5% home field advantage:
Yankees true prob at home ˜ 58.2%

Fair Yankees moneyline: -139
Fair Rays moneyline: +139

Current market: Yankees -132, Rays +120
Rays priced ~19 cents TOO SHORT.

EV AT RAYS +120 (true prob 41.8%):
(0.418 × $1.20) - (0.582 × $1) = -$0.08
NEGATIVE EV of about 8% per dollar

EV AT YANKEES -132 (true prob 58.2%):
(0.582 × $0.758) - (0.418 × $1) = +$0.023
Positive EV of about 2.3% per dollar

==========================================
HONEST ADJUSTMENTS TO PYTHAG
==========================================

Pythag isn't perfect. Some real adjustments:

REDUCE YANKEES EDGE BY:
- Recent cold form has SOME signal beyond Pythag (~2 pts)
- 0-4 H2H, small sample but not zero info (~1 pt)
- Tampa's SP slightly better tonight (~1 pt)

That brings Yankees true probability to ~54-55%, fair price near -120.

ADJUSTED EV:
- Rays +120 (45.5% implied) vs ~45% true ? roughly fair value
- Yankees -132 (56.5% implied) vs ~54-55% true ? slightly negative EV

With reasonable adjustments, this game is close to a coin flip with
the market priced fairly.

==========================================
VERDICT
==========================================

SKIP THE GAME.

- Rays +120 looks like negative EV after Pythag adjustment
- Yankees -132 needs 56.9% to break even, true probability ~54-55%
- Neither side clears the threshold for a confident play

This is essentially a coin flip with a slight Yankees lean, and the
market knows it. The Rays-bull narrative looked compelling because
nine signals all pointed the same way — but they were nine flavors
of the same underlying observation (short-term form), not nine
independent confirmations.

If forced to bet: Yankees -132 has the better fundamentals but EV
is thin. A $5 token on Yankees would be defensible. Bigger sizing
isn't supported by the math.

==========================================
WHAT TO WATCH IF NOT BETTING
==========================================

1. Rasmussen's first inning — early traffic gets punished
2. Weathers' fastball command — wild = walks for patient Tampa bats
3. Tampa's approach vs lefties — they've been mediocre vs LHP
4. Yankees pressing at -132 — slumping favorites often press

If Yankees win 5-2 with Weathers going 6 strong: Pythag was right.
If Rays win 4-1 with Rasmussen dominating: H2H may have real causal
weight worth investigating.

Either way, we learn something.

LESSON: Pythagorean and SOS adjustments should come BEFORE
recent-form analysis, not after. Signal stacks of correlated
short-term indicators can mask the underlying fundamentals.

[End of analysis. Generated Sunday May 24, 2026.]

All that being said, i wanted to take the Rays today but I am going to skip the game. Good luck today!

Nice work, now do the Dodgers game hahaha

Done.

Great work, but you forgot this...Yankees have one win vs teams (1-10) with a .500 record or better in 2026.

The run differential lucky/unlucky is hard to swallow. Baseball games are won/lost by 1 run. It is less impressive to beat bad teams 10-1 than good teams 1-0. That is where run differential falls apart. The Yankees SHOULD have a much higher differential because they have played THE WORST SCHEDULE IN MLB SO FAR THIS YEAR. Yankees have played only 10 games vs teams at .500 or better, AND HAVE LOST 9 of those games. In the same category the Rays are 14-11. 

Question: Can the Yankees go 13-1 vs +500 teams to level things off? With a .100 probability that looks unlikely.

@River_fish

 Good call. I thought I was maybe the only capper on this site to notice. Let's not talk about it though, we need all that Yankee-backer-money to keep the odds high.