Thursday, September 17, 2026

Daily bet research report

Daily Bet Research Report — Thursday, September 17, 2026

Generated 11:42 AM ET

Bankroll: $61.03 | Stake (12.5%, one play): $7.63 Record: 2-0-0 (Main bets, main leg, since 9/14 reset) | Hit rate: 100.0% Slate: 10 games (9 MLB · 1 NFL) · 8 analysed · Playbooks: MLB v1 (NFL/NHL/NBA stubs, not used) · Verified at 11:12 AM ET

Playbooks read: mlb.md (v1, 2026-09-16) in full. nfl.md, nhl.md, nba.md opened and confirmed to be NOT YET BUILT stubs. Per the stub rule, the single NFL game on today's card was left unanalysed because MLB — which has a real playbook — offers nine.

Step 1 freshness check: Verified against statsapi.mlb.com/api/v1/schedule at 11:12 AM ET — 9 MLB games. Cross-checked every participant and probable starter against the VegasInsider Las Vegas board at 11:22 AM ET: all 18 teams matched, and the board supplied the four starters statsapi still had as TBD (Bachar–PIT, Wrobleski–LAD, Pecko–HOU, Bradford–TEX). H3 (two independent sources on both starters) is satisfied for every game analysed. ESPN core-API events endpoint returned the same 9 game IDs. NFL slate confirmed at 1 game via ESPN core API.

Section 2: Game Analysis

A note on method that applies throughout. The playbook (§2) requires xFIP or SIERA rather than ERA. FanGraphs leaderboards were not reachable inside the research window, so I computed FIP directly from MLB StatsAPI counting stats (13·HR + 3·(BB+HBP) − 2·K)/IP + 3.15, and an xFIP-proxy that replaces each pitcher's actual home runs with an expected total at a 1.20 HR/9 league rate. That is a rougher instrument than true xFIP — it normalises home-run luck but does not adjust for a pitcher's genuine fly-ball tendency, so it flatters extreme fly-ball arms. Every xFIP figure below is that proxy, and I have flagged where the distinction matters.

Milwaukee Brewers @ Pittsburgh Pirates — 12:35 PM ET, PNC Park Key inputs: Kyle Harrison (L) — 3.98 ERA / 4.08 FIP / 3.47 xFIP-proxy over 110.2 IP, 10.90 K/9, 2.68 BB/9. His 1.63 HR/9 is the whole gap; strip the home-run luck and he has been a mid-3s pitcher. That is the playbook's §2 buy direction — ERA above skill, regression runs his way. Against him: Lake Bachar, a career relief arm with 11 starts in 112 career appearances (161.2 career IP, 3.73 ERA, 1.39 HR/9). H4 clears — he is not a sub-3-start rookie — but this is functionally a bullpen game in the third game of a three-game series. Line: No opening number recorded on the board. bet365 MIL −130, Consensus −130, Caesars/FanDuel −132. DraftKings is not posting this game at all, moneyline or total. Read: The starter comparison is real but modest; the under-priced piece is §3 — Pittsburgh needs roughly six or seven innings from a pen already two games into the series, and Bachar's career pitch-count profile (16.4 P/IP, 23 games finished, only 11 starts) means a short leash is close to certain. Play: Milwaukee moneyline at bet365. I could not confirm actual Pirates bullpen innings from the prior two games — that gap is in Limitations, and it is the load-bearing assumption here.

Los Angeles Dodgers @ Cincinnati Reds — 12:40 PM ET, Great American Ball Park Key inputs: Justin Wrobleski (L) — 3.78 ERA, 4.34 FIP, 4.03 xFIP-proxy, 133.1 IP, 2.09 BB/9, 0.81 GO/AO. He has been modestly lucky (ERA half a run under FIP), which is a mild fade signal. Brady Singer (R) — 5.38 ERA, 5.45 FIP, 4.60 xFIP-proxy, 150.2 IP, 7.47 K/9, 3.35 BB/9, 1.79 HR/9, .283 opponent average, 1.51 WHIP. Even after normalising his home runs he is the clearly worse pitcher, by about 0.57 runs of projected skill — and GABP is exactly the park that will not forgive his fly-ball rate. Line: Opened LAD −172. Current: bet365 −165, DraftKings −172, Caesars −170, FanDuel −168, Consensus −172. Totals: DK o9.5 −118 / u9.5 −102; bet365 o9.5 −115 / u9.5 −105. Read: No movement off the open — the market's number is −172 and it has stayed there. The tradeable fact is that bet365 is seven cents soft of the consensus, which is the entire mechanical edge: 62.3% implied instead of 63.2%. The handicapping case (starter skill gap by the playbook's preferred metric, in a park that punishes the worse arm) supports the side. Play: Dodgers moneyline at bet365 −165. The total is a pass — GABP plus a weak Reds starter argues over, but 9.5 already prices that.

Athletics @ Tampa Bay Rays — 1:10 PM ET, Tropicana Field (dome) Key inputs: Drew Rasmussen (R) has been the best pitcher on the slate — 2.81 ERA, 2.93 FIP, 0.92 WHIP, 6.74 H/9, 1.52 BB/9, .204 opponent average over 160.1 IP. His xFIP-proxy of 3.52 is worse than his FIP, meaning some home-run suppression is luck, but nothing here says fade. Jeffrey Springs (L) is the opposite pole: 5.93 ERA, 6.04 FIP, 135 IP, 2.20 HR/9 — 33 home runs allowed. His xFIP-proxy is 4.59, a 1.34-run gap that is the largest on the card, but with 171 air outs against 119 ground outs he is a genuine fly-ball pitcher, so the proxy overstates the correction. This is the case where the missing true xFIP matters most. Line: Rays −290 / A's +232 — both sides outside the 1.50–2.10 band, no moneyline available. Total 7.5 is uniform across all nine books: o7.5 −115 / u7.5 −105 at both DK and bet365, and it opened there. Zero movement. Read: A dome removes the weather layer entirely (§4), so this is a pure pitching read — and the market has already made it. A −290 favourite is expected to score roughly five and a half runs on its own; asking for under 7.5 requires Tampa's offence to go quiet against a pitcher giving up two homers a start. No play.

San Diego Padres @ Colorado Rockies — 3:10 PM ET, Coors Field No play — Hard Rule H1. Coors Field is an automatic pass regardless of the numbers. For the record the market has this at PDR −167 / COL +138 with a total that ranges from 10.5 to 11.5 across books, which is precisely the instability H1 exists to avoid.

Philadelphia Phillies @ New York Mets — 7:15 PM ET, Citi Field Key inputs: Aaron Nola (R) — 4.69 ERA, 4.90 FIP, 3.90 xFIP-proxy, 9.20 K/9, 161.1 IP. He has allowed 34 home runs, 1.90 HR/9, which is the single most extreme home-run rate among today's established starters. Playbook §2 is explicit that this is a buy on his team, not a fade. Nolan McLean (R) is the mirror image: 3.24 ERA but 3.73 FIP and 4.13 xFIP-proxy, with 66 walks and 15 hit batsmen in 166.2 IP. His 0.92 HR/9 and .214 opponent average have carried a shaky command profile, and the playbook says that gap points at regression the wrong way. Line: Mets −144 (open) → bet365 −140, DK −144, Rivers −134. Phillies +119 open, +120 bet365, +119 DK — out of band on the plus side, so the side I actually like is unplayable. The total is the interesting artefact: DraftKings and bet365 are at 8, while Caesars, FanDuel and Fanatics are at 8.5. Three books have the line a half-run higher. Read: The Phillies moneyline is the clean expression of the Nola signal and the odds band rules it out. What remains is the line disagreement — buying Over 8 at a book sitting a half-run below three competitors, with a starter matchup where the underpriced arm is Nola and the overpriced one is McLean. Play: Over 8 at bet365 −110. Thin, and I am not pretending otherwise.

Kansas City Royals @ Houston Astros — 7:15 PM ET, Daikin Park (retractable roof) Key inputs: Seth Lugo (R) — 4.99 ERA but 4.22 FIP and 4.45 xFIP-proxy over 164 IP, with 44 doubles allowed and a 1.45 WHIP. A three-quarter-run ERA-over-FIP gap is the playbook's buy signal, driven by sequencing rather than skill collapse. Houston counters with Pecko, for whom I could not retrieve season statistics inside the window. Line: Astros −150 at bet365, −145 Caesars/Fanatics, −148 FanDuel, Consensus −150. DraftKings is not posting this game. Total 9 at bet365 (o −105 / u −115), 8.5 at Caesars and FanDuel. Read: The playbook signal points at Kansas City — and KC at +125 (2.25) is outside the band, so the signal is unplayable. Taking Houston at −150 would mean betting against the one thing my method says is mispriced here. No play. With the Houston starter unquantified I would be guessing at half the matchup anyway.

Detroit Tigers @ Chicago White Sox — 7:40 PM ET, Rate Field Key inputs: This is the closest thing on the card to two identical pitchers. Framber Valdez (L): 4.01 ERA, 4.22 FIP, 4.76 xFIP-proxy, 163.2 IP, 1.61 GO/AO, 3.30 BB/9, 6.65 K/9. Erick Fedde (R): 3.97 ERA, 5.17 FIP, 4.67 xFIP-proxy, 133.2 IP, 6.13 K/9, 3.23 BB/9, 1.55 HR/9. Both have outperformed their peripherals, Fedde dramatically so — a 1.20-run ERA-under-FIP gap is the clearest fade signal on the slate under §2. Note Fedde has 14 starts in 31 appearances, so he is a swingman and his pitch-count ceiling is a live question. Line: Opened CWS −136. bet365 −130, BetMGM −135, DK −136, FanDuel −126, Consensus −136. Detroit: open +113, DK +113 (2.13, out of band), bet365 +110 = 2.10, exactly at the band edge. No meaningful movement off the open. Read: Projected skill is within a tenth of a run, yet the market makes Chicago a −136 home favourite — more than plain home-field would justify between two even arms, and it is the side with the larger luck cushion to give back. Play: Detroit moneyline at bet365 +110. The risk is that Rate Field's park effect plus a right-handed Chicago lineup against a ground-ball lefty is a real edge I am treating as noise.

Minnesota Twins @ Los Angeles Angels — 9:38 PM ET, Angel Stadium Key inputs: Taj Bradley (R) — 3.81 ERA, 4.04 FIP, 3.97 xFIP-proxy, 10.34 K/9 but 3.92 BB/9 and 0.65 GO/AO over 165.1 IP. Skill and results are aligned; no mispricing signal in either direction. Opposite him, Walbert Ureña — I could not retrieve his season line before the cutoff, and with a 2007-born player ID he is plausibly inside H4's three-career-start window. Line: Twins −136 open → bet365 −135, DK −136, Fanatics −130. Angels +112 DK / +115 bet365 (2.15, out of band). Total 8, o −115 / u −105 at both books. Read: No play. Half the matchup is unquantified, and under H4 I cannot rule out that the Angels starter is a near-debut arm — in which case the only permitted play would be against Los Angeles built on that inexperience, which is exactly the case I lack the data to make.

Boston Red Sox @ Texas Rangers — 8:05 PM ET, Globe Life Field. Sonny Gray (R) has been excellent — 2.76 ERA, 1.17 WHIP, 156.2 IP, 3.33 K/BB — against Bradford (L) for Texas. The entire VegasInsider row is N/A across all nine books, moneyline and total: this game is off the board as of 11:22 AM ET. Unpriceable, therefore unanalysed beyond this. Not a truncation issue — the cells are genuinely empty.

Section 3: Today's Plays

# Play Sport Market Book Odds Confidence Calibrated Edge %
1 Los Angeles Dodgers ML (at CIN) MLB Moneyline bet365 1.61 64% 61% −1.27%
2 Milwaukee Brewers ML (at PIT) MLB Moneyline bet365 1.77 61% 58% +1.47%
3 Over 8 — PHI @ NYM MLB Over/Under bet365 1.91 55% 51% −1.38%
4 Detroit Tigers ML (at CWS) MLB Moneyline bet365 2.10 50% 46% −1.62%

Note: the whole slate prices tight. Three of four plays carry negative edge at the quoted price. That is shown rather than hidden, per the playbook's instruction that edge is a label and not a gate — but it is the honest headline of this card.

Section 4: Reasoning

#1 — Dodgers ML, bet365 1.61. The argument rests on MLB playbook §2, [STRONG] — xFIP and SIERA are the best available predictors of forward ERA, and ERA itself is noise. By that standard Singer's 4.60 xFIP-proxy against Wrobleski's 4.03 is a genuine 0.57-run skill gap, and Singer's 1.79 HR/9 and .283 opponent average will not improve in Great American Ball Park (§4, park effects are stable and multi-year). The line has not moved off its −172 open at DraftKings or Consensus, which tells me sharp money has no quarrel with the number — I am not fighting it, I am buying the same number seven cents cheaper at bet365. What beats me: Wrobleski's own ERA sits half a run under his FIP, so he is the one due to regress worse, and the Dodgers' September lineup is the kind a contender rests (§6) — I could not confirm today's actual lineup card.

#2 — Brewers ML, bet365 1.77. This is a §3 play, the playbook's designated under-priced factor, supported by §2. Lake Bachar has made 11 starts in 112 career appearances; Pittsburgh is asking its bullpen to cover six-plus innings in the finale of a three-game series, and the §3 [MODERATE] finding is that relief effectiveness holds through back-to-backs and breaks on the third consecutive day — which is where a series-closing bullpen game puts the high-leverage arms. Harrison adds the §2 [STRONG] buy signal: 3.98 ERA against a 3.47 xFIP-proxy, with the entire gap sitting in home-run luck. What beats me: I could not verify the Pirates' actual bullpen innings from Tuesday and Wednesday, so the core §3 claim is inference from Bachar's role rather than measured usage — if Pittsburgh's pen is rested, this play is mostly just Harrison against a bad team, which is already in the price. DraftKings declining to post the game is itself a small warning that the market considers it hard to price.

#3 — Over 8, PHI @ NYM, bet365 1.91. Rests on §2 [STRONG] applied in both directions: Nola's 1.90 HR/9 has inflated a 4.69 ERA over what looks like ~3.90 true skill, while McLean's 3.24 ERA sits well under a 3.73 FIP built on 66 walks and 15 hit batsmen. The market has McLean −144 and Nola +119, which prices the surface numbers rather than the peripherals. The mechanical support is the half-run line disagreement — three books at 8.5 against DK and bet365 at 8 — so I am buying the cheaper number on the side the skill read supports. Line movement: DK opened and remains at 8, no drift. What beats me: Citi Field is a run-suppressing park and 8 is already a low total; if Nola's home-run problem is real rather than luck — and a 34-homer season is a lot of evidence — the correction I am counting on is imaginary. This is the weakest-supported play on the card, and its evidence base is a proxy metric rather than true xFIP.

#4 — Tigers ML, bet365 2.10. §2 [STRONG] on the fade side: Fedde's 3.97 ERA against a 5.17 FIP is a 1.20-run gap, the largest overperformance on the slate, and the playbook is unambiguous that ERA well below skill is a fade signal regardless of how the recent line looks. Valdez has overperformed too (4.01 vs 4.22 FIP), but by a quarter of the margin. With xFIP-proxies of 4.76 and 4.67 these are the same pitcher, and a −136 home price is more than neutral home-field between equals. No movement off the +113 open. What beats me: Fedde has started only 14 of 31 games, so Chicago may be managing his innings deliberately and handing the game to a fresher pen — which turns a starter-skill argument into a bullpen argument I have not researched. At 2.10 this sits exactly on the band edge and is ranked last on purpose.

Sources

Research limitations

← Back to all reports
Betting Tracker · Generated 2026-09-18 12:21 · Personal use only