# Bahia vs Grêmio > Serie A · Kickoff Sun 17 May 2026, 19:00 UTC · [Canonical HTML](https://betsprinter.com/fixtures/36196) **Status:** Finished **Final score:** Bahia 1–1 Grêmio ## Model verdict - **Bahia win:** 28% - **Draw:** 22% - **Grêmio win:** 50% - **Source:** model ## Pre-match deep dive ### Model backs an away shock as home edge looks cosmetic ## The stage Sunday’s Serie A meeting kicks off at 19:00 UTC and lands as a mid-May test with table consequences for both sides[^fact-1]. Bahia are the nominal hosts[^fact-3] and Grêmio the visitors[^fact-3]. ## Form & momentum Recent form offers a contrast: Bahia’s last 10 shows read LDDLW (4-2-4 W-D-L) with 1.40 points per game, 1.50 goals scored and 1.50 conceded per match[^fact-4]. Grêmio’s ten-match run is LDWLD (2-4-4 W-D-L) with 1.00 points per game, 0.60 goals scored and 0.80 conceded per match[^fact-5]. The model still favours the away side markedly — Away 50%, Home 28%, Draw 22% — a gap driven by a clear confidence margin to the runner-up[^fact-2]. That away lean sits alongside an applied Elo advantage for the home team worth +100 points to Bahia on paper, a datum the model has explicitly absorbed when forming probabilities[^fact-3]. ## Personnel Bahia enter with a clear in-form attacking outlet: Luciano Juba has three goals in his last four appearances and an average rating of 7.47 across that run[^fact-8]. Grêmio’s nearest hot-streak contributor in the supplied facts is Gabriel Mec, who has one goal in his last four appearances and an average rating of 6.96[^fact-9]. The list of absentees bites on set-piece and wide options: Santiago Ramos Mingo is suspended for Bahia after 360 minutes in the recent run[^fact-10]; Grêmio will be without Cristian Pavón, also suspended after 347 minutes in the recent run[^fact-11]. ## Where the model sees value Three markets were compared against the model for edges[^fact-12]. The clearest priced discrepancy is the match-winner market: the model assigns a 46% probability to an away win while Unibet offers 4.70, producing an edge of 24.7 percentage points (high confidence flagged by the model)[^fact-6]. A secondary angle is on goals: the model gives Under 2.5 a 54% probability while bet365 prices the line at 1.95, an edge of 3.1 percentage points (low confidence)[^fact-7]. The size and confidence of the two edges differ: the away match-winner number is both larger and labelled high confidence by the model, while the under 2.5 angle is a smaller, lower-confidence nudge[^fact-6][^fact-7]. ## Verdict The model’s lean is for the away side, with a 50% probability assigned to an away win and a 22 percentage-point gap back to the home side’s probability — a clear, model-level tilt despite Bahia’s +100 Elo adjustment[^fact-2][^fact-3]. The strongest model/market mismatch sits in the away match-winner line, flagged with high confidence by the model[^fact-6]. ### Cited facts [^fact-1]: **Kickoff** — Sun 17 May 2026, 19:00 UTC — Serie A [^fact-2]: **Model verdict** — Home 28% / Draw 22% / Away 50% (source: model; confidence high, 22 pp gap to runner-up). [^fact-3]: **Elo edge** — BAH vs GRE — Elo differential +100 points (with home advantage applied). [^fact-4]: **BAH recent form** — LDDLW last 10: 4-2-4 (W-D-L), 1.40 PPG, 1.50 goals scored / 1.50 conceded per match. [^fact-5]: **GRE recent form** — LDWLD last 10: 2-4-4 (W-D-L), 1.00 PPG, 0.60 goals scored / 0.80 conceded per match. [^fact-6]: **Value pick #1** — Away in Match Winner — model 46% vs market price 4.70 at Unibet, edge 24.7 pp (high confidence). [^fact-7]: **Value pick #2** — Under in Goals O/U 2.5 — model 54% vs market price 1.95 at bet365, edge 3.1 pp (low confidence). [^fact-8]: **BAH in-form player** — Luciano Juba — 3 goals, 0 assists in last 4 appearances, avg rating 7.47. [^fact-9]: **GRE in-form player** — Gabriel Mec — 1 goals, 0 assists in last 4 appearances, avg rating 6.96. [^fact-10]: **BAH key absence** — Santiago Ramos Mingo out (suspension), 360 minutes in recent run. [^fact-11]: **GRE key absence** — Cristian Pavón out (suspension), 347 minutes in recent run. [^fact-12]: **Markets analysed** — 3 market(s) compared against the model. --- Methodology: . Canonical HTML: .