BOS vs ATL prediction for May 15, 2026: Our Monte Carlo simulation ran 10,000 game iterations and projects ATL 3.5 - BOS 2.9. ATL is favored with a 59.4% win probability. The run line is -1.5 and the total is 7.5. Model projects 6.4 total runs.
ATL
3.5
Projected Score
VS
O/U 7.5
BOS
2.9
Projected Score
Win Probability
ATLBOS
-1.5
Run Line (ATL)
7.5
Total Line
10,000
Simulations
Calibrated accuracy at this confidence: 57.6% (2,085 games)
Projected Runs Range 10th – 90th percentile
BOS
135
ATL
245
Projected
ATL 3.5 — BOS 2.9
Actual
ATL 3 — BOS 2
Starting Pitcher Matchup
Connelly Early L
BOS
FF33%94 mph19% whiff
CH21%83 mph24% whiff
SI19%92 mph5% whiff
Spencer Strider R
ATL
FF50%96 mph14% whiff
SL34%84 mph47% whiff
CU10%78 mph56% whiff
Weather Impact
Truist Park
79°F6 mph wind
HR: 1.021 Total: 1.009
thin air
Bullpen Comparison
BOS
3.82ERA
4.54FIP
8.85K/9
3.79BB/9
1.29WHIP
ATL
2.79ERA
3.13FIP
9.85K/9
2.77BB/9
1.04WHIP
Betting Edges
RUN_LINE AWAY +1.5
-46.1% EV
-164
TOTAL OVER 7.5
-31.3% EV
-115
TOTAL UNDER 7.5
+23.3% EV
-105
F5 UNDER 4.5
+15.2% EV
-145
F5_ML HOME
-8.2% EV
-152
RUN_LINE HOME -1.5
-8.1% EV
+136
First 5 Innings & NRFI
BOS F5
1.5 runs
35.9% win
ATL F5
1.8 runs
43.0% win
F5 Total
3.3
NRFI
64.7%
YRFI
35.3%
Avg 1st Inn Runs
0.65
HR Spotlight
Avg HRs
1.9
Over 0.5 HR
85%
Over 1.5 HR
56%
No HR
15%
Wilyer Abreu BOS30.0%
ISO: 0.193 | Barrel: 8.3% | vs Spencer Strider | Platoon: 1.12x
Matt Olson ATL30.0%
ISO: 0.281 | Barrel: 19.3% | vs Connelly Early
Drake Baldwin ATL28.0%
ISO: 0.258 | Barrel: 12.7% | vs Connelly Early
Pitcher Strikeout Projections
Connelly Early
0.0 K projected
BOS | K/9: 0.0
Spencer Strider
0.0 K projected
ATL | K/9: 0.0
Injury Report
BOS8 injured
Tanner Houck SP60-DAY-IL
Garrett Crochet SP15-DAY-IL
Connor Wong CDAY-TO-DAY
Roman Anthony LF10-DAY-IL
Hobie Harris RPDAY-TO-DAY
Triston Casas 1B60-DAY-IL
+2 more
ATL8 injured
Eli White CF7-DAY IL
Joe Jimenez RP60-DAY-IL
Ronald Acuna Jr. RF10-DAY-IL
Spencer Schwellenbach SP60-DAY-IL
Sean Murphy C10-DAY-IL
Danny Young RP60-DAY-IL
+2 more
AI Intelligence Analysis
LEAN +1YELLOW ZONE50.4% WR (n=258)
UNDER 7.5 has 23.3% edge (63.3% model) in YELLOW zone for totals (50.4% WR). Both pitchers are quality (Strider 3.12 ERA, B-grade; Early 3.41 ERA, C+ grade) and ATL has elite bullpen (2.79 ERA, 0.87 closer). Weather warm (79°F) slightly inflationary but wind suppresses (4.2mph in). Edge is real but YELLOW zone means coin flip long-term. ATL is home favorite and bullpen is strong — market under at 7.5 is reasonable but model sees run suppression. Light lean on under supported by SP quality.
Key Factors
- Bullpen strength dominates: ATL 2.79 ERA (1.613 quality, best in slate) + 0.87 closer ERA is elite. BOS 3.82 ERA is good but not elite. Road team going against ATL's elite relief = run suppression likely.
- SP quality supports under: Strider (3.12 ERA, B-) is solid home starter, Early (3.41 ERA, C+) is decent visitor. Both below-average stuff but acceptable command = low-scoring openers.
- Weather mixed: 79°F temperature inflates runs (+0.5 expected) but 4.2mph wind blowing in suppresses (-0.3 expected) = net neutral. NRFI (6.3% edge, 62.5% model) suggests slow starts.
- YELLOW zone total: 50.4% WR means this is a coin flip. 23.3% edge is high but falls in YELLOW zone not RED, so not as severe a trap as BAL/TB under.
- F5 UNDER 4.5 at 15.2% edge (68.1% model) offers cleaner early-game value aligned with bullpen dominance thesis.
Risk Factors
- YELLOW zone totals are 50.4% WR (coin flips). Edge paradox still applies: bigger disagreement = worse outcomes. Market might be right here.
- 23.3% edge is high; calibration caps totals at 20%. This edge exceeds cap, suggesting model overconfidence.
- Home team (ATL) typically gets some public support. Market at 7.5 might be already incorporating ATL's bullpen strength; our 'discovery' might be priced in.
YELLOW ZONEELITE BULLPEN FACTORSP QUALITYINJURY SUPPORT
Edge Analysis
Moneyline
ATL 59.4%
-8.1 pts
Run Line
-1.5
-8.1 pts
Total
7.5
+23.3 pts
How this prediction was generated: This page shows output from the Olympus Bets MLB Baseball Monte Carlo engine. Each game is simulated 10,000 times using real-time team data, injury reports, and current odds. Probabilities are calibrated using Bayesian methods and sized via the Kelly Criterion. Full methodology →