MLB Picks Today: Predictions & Best Bets for September 23, 2026 | SharpSide

Wednesday’s 16-game slate calls for a distinction between the team most likely to win and the price worth taking. Our three divergence matchups are San Diego Padres @ Los Angeles Dodgers, where the model is 1.6 runs more favorable to San Diego Padres than the market—a SIDE edge; Houston Astros @ Seattle Mariners; and Minnesota Twins @ San Francisco Giants. This brief refreshes at 11:30 am Eastern using new analytics, weather factors, pitching changes, and line movement, so the featured matchups may change between morning publication and the refresh. Today’s Best Bet follows immediately: a conviction selection, separate from those market disagreements.

Today's Best Bet: Atlanta Braves ML

Today’s Best Bet is Atlanta Braves ML (-267) in Cincinnati Reds @ Atlanta Braves—the single game and line SharpSide has the highest confidence in today. The model projects a 3.4–4.8 score, away–home, giving Atlanta a clear 1.4-run predicted winning margin. That is roughly in line with the market’s implied margin: model and market agree on the favorite, rather than exposing a substantial pricing disagreement. This is a conviction play, not necessarily our biggest edge, and the expensive moneyline means confidence in the winner should not be confused with a bargain. Best Bet: Atlanta Braves ML (-267)

San Diego Padres @ Los Angeles Dodgers

San Diego starter Robbie Ray carries a 4.73 FIP and .334 xwOBA-against, both unfavorable indicators against Dodgers hitters, while Los Angeles starter Yoshinobu Yamamoto’s 3.28 FIP and .266 xwOBA-against indicate stronger defense-independent pitching and weaker contact allowed. SharpSide projects 4.3–4.2, away–home, for a narrow San Diego win versus consensus at 3.5–5.0 for Los Angeles: the model is 1.6 runs more favorable to San Diego Padres, with San Diego’s relief pitching supplying a concrete counterweight to the starter disadvantage. Yamamoto averages 6.85 innings per start against Ray’s 5.49, extending the exposure to the stronger opposing starter and making the Padres case harder early. Even so, the model still lands on San Diego Padres because San Diego’s 3.49 relief ERA and its +0.06-run offensive regression support the underdog. At +185, San Diego is priced well below an even chance, while the calibrated win read is essentially a coin flip with the faintest edge to Los Angeles—not a resolved probability advantage for the Padres. Heading into Wednesday’s second game of three, San Diego must respond after Los Angeles won the opener 7–0. The park/weather score is 0 with no notable conditions; the model total of 8.5 has a +0.5-run OVER edge against the posted 8.0 total, but exactly 0.0 runs versus consensus. **How the model decided: The biggest factor favoring San Diego Padres is San Diego Padres's 3.49 bullpen ERA, a better-than-average run-prevention mark that also tops Los Angeles Dodgers's 3.82. Supporting it: Los Angeles Dodgers's taxed bullpen (0.32 fatigue index), which the model expects to leak runs in the late innings. At +185 the market prices San Diego Padres near 35% while the model has them around 49%, so a meaningful share of this lean is price value rather than fundamentals. Lean: San Diego Padres ML (+185)

Houston Astros @ Seattle Mariners

Houston starter Ethan Pecko’s 5.51 FIP and 1.57 HR/9 warn of run risk against Seattle in a modest 28.7-inning season sample, while Mariners starter George Kirby’s 3.85 FIP and 1.11 HR/9 favor suppressing Houston. SharpSide projects 4.3–4.1, away–home, for a narrow Houston win against consensus at 3.6–4.5 for Seattle: the model is 1.1 runs more favorable to Houston Astros, supported by Houston’s positive offensive regression. Pecko’s short 4.33 innings per start exposes Houston’s 4.14-ERA bullpen, which adds 0.15 runs to Seattle’s projection. That said, the model still lands on Houston Astros because Houston receives a +0.19-run offensive adjustment and Seattle’s 4.52 relief ERA signals late scoring opportunities. At +108, Houston is priced slightly below an even chance, while the calibrated win read is essentially a coin flip with the faintest edge to Seattle, making this a thin price-based lean. Heading into Wednesday’s two-game series finale, Houston seeks another win after beating Seattle 7–0 in the opener. The park/weather score is 0 with no notable conditions; the 8.4 model total carries a +0.9-run OVER edge against the posted 7.5 total, separately standing 0.3 runs above consensus. **How the model decided: The biggest factor favoring Houston Astros is Seattle Mariners's vulnerable 4.52 bullpen ERA, worse than the roughly 4.10 league average, which improves Houston Astros's scoring outlook late. Supporting it: Houston Astros's unlucky 0.277 BABIP, worth a positive 0.19-run offensive regression. Lean: Houston Astros ML (+108)

Minnesota Twins @ San Francisco Giants

Minnesota starter Connor Prielipp’s 3.76 FIP and .306 xwOBA-against support suppressing Giants hitters, while San Francisco starter Cesar Perdomo’s 4.96 FIP and .338 xwOBA-against suggest harder contact and run risk in a small 19.3-inning season sample. SharpSide projects 4.4–4.5, away–home, for a narrow San Francisco win against consensus at 4.2–3.3 for Minnesota: the model is 1.0 runs more favorable to San Francisco Giants, with Minnesota’s relief weakness providing the clearest baseball support. San Francisco is missing core hitter Jung Hoo Lee, costing approximately 0.09 projected runs, on top of the uncertainty surrounding Perdomo’s small sample. Despite those factors, the model still lands on San Francisco Giants because Minnesota’s bullpen adds 0.38 Giants runs and +134 offers underdog compensation for a near-even matchup. At +134, San Francisco is priced below an even chance, while the calibrated win read is essentially a coin flip with the faintest edge to the Giants; this is low conviction rather than broad fundamental superiority. Heading into Wednesday’s rubber match, San Francisco beat Minnesota 5–2 in the opener before Minnesota answered with a 3–2 victory. The park/weather score is 0 with no notable conditions; the model total of 8.9 has a +0.9-run OVER edge against the posted 8.0 total, separately running 1.4 runs above consensus. **How the model decided: The biggest factor favoring San Francisco Giants is Minnesota Twins's vulnerable 4.79 bullpen ERA, worse than the roughly 4.10 league average, which improves San Francisco Giants's scoring outlook late. At +134 the market prices San Francisco Giants near 43% while the model has them around 52%, so a meaningful share of this lean is price value rather than fundamentals. Lean: San Francisco Giants ML (+134)

First Five Innings: San Diego Padres @ Los Angeles Dodgers (F5 UNDER)

The starter-focused projection puts the first five innings at 2.90 runs versus a market-implied 4.42, a supplied -1.52-run difference supporting the UNDER. Ray’s F5-only ERA is 3.50 across 141.3 innings, paired with his 3.60 ERA over 162.3 innings this season; the supplied blended starter input is 3.60. Yamamoto’s F5-only ERA is 2.27 across 135.0 innings, paired with his 2.51 ERA over 179.0 innings this season; the supplied blended input is 2.50. The Bayesian calculation blends opponent-starter ERA with Statcast xERA at 65/35, applies five innings of exposure, 0.92 early-game damping, and a 0.930 park multiplier. San Diego’s unlucky BABIP adds 0.06 offensive runs, while Los Angeles’s lucky BABIP subtracts 0.11, providing an offensive adjustment alongside the starter inputs. This is a different horizon from the full-game forecast: the early projection emphasizes the starters, while the market-implied 4.42 is an estimate from full-game consensus, not a quoted F5 betting line. Lean: F5 UNDER 4.0 total

Batter Prop: Josh Smith Over 0.5 Hits

Josh Smith’s Hits Over 0.5 in Toronto Blue Jays @ Baltimore Orioles is a market-based lean, not a newly supplied player projection. Cross-book consensus moved from -120 to -153, a 5.9 probability-point move toward the OVER from opening. A consensus adjustment this aggressive historically flags smart money, though price movement does not establish who placed the bets or guarantee the outcome. The current -153 price is materially less favorable than the opener, so the movement is evidence to evaluate—not permission to chase any subsequent number. Lean: Josh Smith Over 0.5 Hits (-153)

How the model decided

San Diego Padres @ Los Angeles Dodgers is primarily a starter-versus-relief disagreement: Yamamoto suppresses Padres scoring, while San Diego’s stronger bullpen and favorable offensive regression support the underdog. Houston Astros @ Seattle Mariners pairs meaningful Pecko risk with upward Houston BABIP regression and a vulnerable Seattle bullpen. Minnesota Twins @ San Francisco Giants has opposing pitching and roster pressures, with Minnesota’s relief weakness supporting San Francisco despite Lee’s absence. Park/weather supplies no meaningful scoring adjustment in these three games, and baserunning contributes no additional runs. The model also considered advanced Sabermetric data and other analytics.

How to read this

These are the model’s views against market prices, not guarantees or financial advice. Recheck the live projection, confirmed lineups, starting pitchers, and available prices before acting, particularly when a narrow lean depends on underdog compensation.

Projections update after publish. This brief is locked in at the 11:30 am ET refresh, but the model keeps recalculating all day as lineups are posted, weather shifts, pitching changes are announced, and lines move — a side or total highlighted here can flip before first pitch. Always confirm the current number on the live projections page (which locks 30 minutes before each game's first pitch) before placing a bet.