Daily Bet Research Report — Thursday, September 10, 2026

Report generated at 2:40 PM ET on Thursday, September 10, 2026

Bankroll: $20.92 | Stake (12.5%): $2.62 Record: 43-41-3 Main (all bets 46-63-3) | Hit rate: 51.2% Main (42.2% all) Slate: 7 games across MLB, NFL, NCAAF · 5 shortlisted for deep analysis (15 market candidates priced)

Pending bets: None. The Log shows no Open rows — bets #99 and #100 (9/8) both settled Loss.

Standing flags active this run: Yankees ML tilt-fade (Strategy tab) · deep drawdown (bankroll 93.2% below the $309.11 ATH), so preference shifts toward higher-hit-rate qualifiers, not a higher edge bar.

---

Section 1: Slate Inventory (Pass 1)

In-season check: MLB and NFL are running. NCAA Football has exactly one FBS game. NBA, NHL, NCAA Basketball are out of season (all resume late Oct / Nov) — verified against the ESPN season calendars, not assumed.

7 games scheduled, 2 already underway, 5 playable. Every playable game was deep-dived — on a slate this small, triage collapses into full coverage.

---

Section 2: Shortlist Analysis

Coverage note: 2 of 7 games were skipped without deep analysis (Rays @ Braves, Astros @ Phillies — both already in progress at report time). The other 5 were priced across every market bet365 posts.

EV is expected return per $1 staked (p × decimal − 1). Star bands are read off that figure per the scale at the foot of this report.

---

Section 3: My Recommendations

Staked total: $1.79 + $0.83 = $2.62 = 12.5% of $20.92. ✓

Only three plays cleared a positive edge — the slate offered five playable games, and two of them (Pirates–White Sox, Florida A&M–Miami) were unpriceable rather than unattractive. Picks 2 and 3 share one thesis on one team; on a three-market qualifying pool that concentration was unavoidable, not a sign the screen stopped early.

Straight single-leg plays only. Build any SGP yourself from the watchlist below.

---

Section 3b: Prop Watchlist

Researched, at bet365 prices, not recommendations:

1. Jacob deGrom Over 6.5 K — bet365 1.87 (−115). bet365 is the outright best price in the market (BetMGM −120, DraftKings −129, Caesars −124, Sleeper −141, Rivers −139). The reason it's cheap is the matchup: deGrom has faced Seattle three times in 2026 for 6, 3 and 4 strikeouts across an average of 4.7 innings — 8.4 K/9 against them versus 10.9 K/9 overall. Good number, wrong side. The Under is the play if you can find a fair price for it; bet365's Under is not it.
2. Max Fried Over 5.5 K — bet365 2.00 (+100). The single most generous prop price on the board (every other book −103 to −137). Also the most misleading: Fried's last three starts produced 4, 4 and 4 strikeouts, and since returning from a ten-week IL stint on July 22 he has averaged 4.86 innings and has not been past 6.1 in any start. A managed workload against a bottom-tier Rockies lineup projects ~4 K, not 6.
3. Logan Gilbert Over 6.5 K — bet365 1.71 (−140). Tied-best with BetMGM. Gilbert's last five K totals: 7, 5, 5, 4, 10 — he has cleared 6.5 twice in five. At −140 you need 58.3%; he isn't there, and his innings are shrinking (6.0, 5.0, 5.0, 5.0, 4.2). If you want a Gilbert leg as an SGP amplifier, drop to a lower alternate line rather than take this one.

---

Section 4: Reasoning

Pick 1 — 49ers @ Rams, Under 48.5 (bet365 1.91) — STAKED $1.79

The math. bet365 posts u48.5 at −110 → implied 52.38%. Multi-book board: DraftKings u48.5 −115, FanDuel −115, BetMGM −110, Fanatics −110, Caesars u48 −107, HardRock u48 −110, Rivers u48 −108; consensus u48.5 −115 / o48.5 −105. De-vigging the consensus pair: 51.22% + 53.49% = 104.71% → fair Under = 51.08%, fair Over = 48.92%. My estimate is 54.0%. Edge = 54.0 − 52.38 = +1.62%. EV = 0.540 × 1.91 − 1 = +3.1% per $1.

Putting consensus next to bet365 is what makes this play readable. The total opened −110/−110 at DraftKings and the Under has since been bid to −115 there and at FanDuel, with Caesars, HardRock and Rivers dropping the number outright to 48. bet365 has not moved either way. So the edge here is not me disagreeing with the market — it is bet365 holding a stale −110 on the side the market is actively buying.

The evidence (NFL framework). Injury/QB status first, per the framework: - San Francisco: Brock Purdy starting. Only ruled out is DL Alfred Collins (torn patellar tendon, out for the season). James Thompson Jr. questionable (hamstring). Nick Bosa (knee) and George Kittle (Achilles) carry no game designation — Bosa said he is "100 percent ready to go," but Shanahan stated explicitly that the team will manage both players' workloads. - Los Angeles: Aaron Donald is the only player ruled out and did not travel to Australia. Myles Garrett and Jaylen Watson were full participants Thursday after missing Tuesday.

That leaves both elite edge rushers active (Garrett and Bosa) with an interior DT missing on each side — a pass-rush-heavy, interior-light matchup, which suppresses efficient passing more than it opens running lanes.

The situational factor doing the real work is rest and body clock, which is the framework's rest-differential input in its most extreme form. ESPN flags this competition as neutralSite: true — the Rams are the nominal home team but there is no home field at the Melbourne Cricket Ground. Kickoff is 10:35 a.m. Friday local after both teams crossed roughly sixteen time zones and practiced in Melbourne all week. Both offenses are timing-and-motion systems (Shanahan and McVay); timing degrades under circadian disruption faster than defensive assignment football does. Add an unfamiliar drop-in surface on a converted cricket oval, Week 1 starters with minimal preseason snaps, and a divisional matchup — the framework's own note that divisional games trend lower-scoring.

Risk factors. McVay and Shanahan are two of the league's most efficient play-callers and can score on script regardless of conditions; a 48.5 total is already low enough to embed a defensive expectation, so I am buying a number the market has partly priced. Week 1 totals also carry wide variance, and one special-teams or defensive score blows through the thesis. The edge is modest — this is a small, price-driven play, not a conviction bet.

Source: 49ers final Week 1 injury report (49ers.com, Sep 10)

---

Pick 2 — Rangers ML, first 5 innings (bet365 1.95) — STAKED $0.83

The math. bet365 posts Texas F5 at −105 → implied 51.22%. F5 board: BetMGM −105, DraftKings −105, FanDuel −104, Rivers −120 on Texas; Seattle F5 at bet365 −125, BetMGM −118, DraftKings −125, FanDuel −120, Rivers −109. Taking the tightest cross-book pair (FanDuel −104 / Rivers −109): 50.98% + 52.15% = 103.13% → fair Texas F5 ≈ 49.44%. Full-game consensus de-vigs to Texas 47.52% / Seattle 52.48%. My estimate is 52.0%. Edge = 52.0 − 51.22 = +0.78%. EV = 0.520 × 1.95 − 1 = +1.5% per $1. Decimal 1.95 clears the 1.50 floor.

The evidence (MLB framework — starting pitcher is the dominant variable). This is a starter-form divergence play, which is why I am taking it in the F5 market where the starters are the whole bet rather than the full game where two bullpens dilute it.

Logan Gilbert (SEA), season 3.71 ERA. Last five starts: 6.0 IP/0 ER, 5.0/3, 5.0/2, 5.0/3, 4.2/7 — 25.2 IP, 15 ER, 5.26 ERA. That is a +1.55 divergence from his season line, which clears the framework's ~1.5 flag. The innings trend is the corroborating signal: 6.0, 5.0, 5.0, 5.0, 4.2. He is not finishing five cleanly, which is exactly the failure mode an F5 line punishes.

Jacob deGrom (TEX), season 3.96 ERA. Last five: 2.0 IP/2 ER, 6.0/0, 3.2/8, 7.0/0, 6.0/2 — 24.2 IP, 12 ER, 4.38 ERA. A +0.42 divergence, well inside the flag, and the shape is bimodal rather than declining: two scoreless outings (7 IP vs the Athletics, 6 IP vs Washington) against one blowup. Most recent start, 9/5 vs Tampa Bay: 6.0 IP, 2 ER, 7 K.

So the market has Seattle as a −120/−125 home favorite while the home starter is 1.55 runs of ERA worse than his season line over five starts and the road starter is essentially at his. Seattle is also 67-78 and eliminated; Texas is 72-74. I move Texas from a fair ~49.4% to 52.0% on the Gilbert divergence alone — a 2.6-point adjustment, deliberately smaller than the raw form gap because five starts is a small sample and Gilbert has handled Texas adequately this year (3.86 ERA across three starts).

Risk factors. Five starts is noisy in both directions — deGrom's own last five contain a 3.2-inning, 8-earned-run disaster, and his floor is as low as Gilbert's. Gilbert has faced Texas three times in 2026 and gone 6.0, 5.1 and 5.2 innings at a 3.86 ERA, so the matchup-specific record argues against the general form read. F5 markets also carry heavy vig (bet365's own two-way here holds 6.8%), which is most of why the edge is under one point. T-Mobile Park suppresses offense, which helps Seattle's pitcher more than Texas's bats.

Sources: Jacob deGrom 2026 game log (MLB Stats API) · Logan Gilbert 2026 game log (MLB Stats API)

---

Pick 3 — Rangers ML, full game (bet365 2.00) — GRADED ONLY

The math. bet365 posts Texas at even money → implied 50.00%, decimal 2.00. Board: BetMGM even, Fanatics even, Rivers even, DraftKings +101, Caesars +105, FanDuel +108, HardRock +110; consensus +101 / Seattle −122. De-vigging consensus: 49.75% + 54.95% = 104.70% → fair Texas = 47.52%, Seattle 52.48%. My estimate is 50.5%. Edge = 50.5 − 50.00 = +0.50%. EV = 0.505 × 2.00 − 1 = +1.0% per $1.

The evidence. Same Gilbert-versus-deGrom form divergence as Pick 2 (5.26 last-five ERA against a 3.71 season line, versus deGrom's 4.38 against 3.96), diluted by nine innings of bullpen instead of five. Note that bet365 is at the bottom of the market on this price — HardRock +110 and FanDuel +108 are materially better numbers — which is the documented pattern of bet365 being among the worst moneyline books. At even money the play still clears zero, but only barely, and it clears by less than the F5 version of the identical thesis. That is why it ranks third and gets graded rather than staked.

Risk factors. All of Pick 2's risks plus full-game bullpen variance, which is the noisiest input in a baseball moneyline. The edge is half a point — inside my own estimation error, and the known ~5-point optimism in these Hit Rate estimates would erase it entirely.

Source: Rangers @ Mariners multi-book odds board (VegasInsider)

---

Why the other three playable games produced nothing

Pirates @ White Sox — voided, not rejected. MLB.com's official probable-pitchers page lists the White Sox starter as TBD. ScoresAndOdds shows "Smith (L)". That is one source, and the Strategy tab requires two sources dated within 24 hours or the league's own page. It fails both. bet365 is offering a genuine pick'em here (Pirates −110 / White Sox −110) plus the best Under 7.5 price in the market at even money and a tied-best Pirates −1.5 at +160 — attractive numbers I am declining to price, because pricing a baseball game without a confirmed starter is guessing.

Rockies @ Yankees — no side. Max Fried's last five starts read 1.38 ERA against a 2.73 season line, but the number that matters is his innings: since returning from a ten-week IL absence on July 22 he has thrown 5.0, 3.0, 5.1, 6.1, 5.0, 3.2 and 5.2 — average 4.86, never past 6.1, with a second 16-day gap in late August. A starter on that cap hands four-plus innings to the bullpen, which is precisely the mechanism that stops a −300 favorite from covering −1.5. The market prices NYY −1.5 at a fair 54.83%; bet365's −130 asks 56.52%, and my read on the innings limit pushes my estimate below fair, not above. The moneyline is dead twice over: 1.33 is beneath the 1.50 odds floor, and the Strategy tab's standing Yankees ML tilt-fade applies. Neither total side beat the vig.

Florida A&M @ Miami — no play. The only FBS game on the board is an FCS visitor at a Power Four host. bet365's number never surfaced in the sources I opened, and I will not substitute another book's price to manufacture a candidate.

---

Sources

• MLB Stats API — schedule and probable pitchers, 2026-09-10
• MLB.com Probable Pitchers, September 10, 2026
• Jacob deGrom 2026 pitching game log (MLB Stats API)
• Logan Gilbert 2026 pitching game log (MLB Stats API)
• Max Fried 2026 pitching game log (MLB Stats API)
• ScoresAndOdds — MLB scores and live odds, Sep 10
• VegasInsider — Rangers vs Mariners multi-book board
• VegasInsider — Rockies vs Yankees multi-book board
• VegasInsider — Pirates vs White Sox multi-book board
• VegasInsider — 49ers vs Rams multi-book board
• VegasInsider — NFL Las Vegas odds
• VegasInsider — MLB strikeout player props
• ESPN NFL scoreboard API — Sep 10 (open vs current lines, neutral-site flag)
• ESPN NCAAF scoreboard API — Sep 10
• NFL.com Week 1 schedule, 2026
• 49ers final injury report ahead of #SFvsLAR (49ers.com)
• Rams Week 1 injury report vs 49ers (therams.com)

---

Research limitations

State plainly what did not work this run:

1. https://www.espn.com/{sport}/odds and /scoreboard returned empty bodies for MLB, NFL and NCAAF — the HTML pages are client-rendered. I substituted the ESPN JSON APIs (site.api.espn.com, sports.core.api.espn.com), which carried the open-versus-current drift data the odds page would have given, but only from DraftKings — ESPN exposes no bet365 line.
2. vegasinsider.com/mlb/odds/las-vegas/ served a cached September 9 board (rotation numbers 961+, all games Final). The ?date=2026-09-10 parameter returned an empty body, and /mlb/odds/money-line/ and /nfl/odds/totals/ both 404'd. I recovered today's bet365 MLB prices from the per-game matchup pages, which loaded current data correctly. The NFL Las Vegas board loaded live without issue.
3. No bet365 price was captured for the NCAAF game. One failed avenue is not proof a market is unpriceable, so Florida A&M @ Miami is marked no-play on the mismatch, not on the price — but I did not price it, and I am not claiming otherwise.
4. Prop coverage is Over-side only. VegasInsider's strikeout board displays the Over line and price per book and does not expose the Under. My read on deGrom, Fried and Gilbert all points to the Under, and I could not price a single one of them at bet365. That is the largest gap in this report: the sport-and-market combination most likely to qualify at bet365 (props) is the one I could only see half of.
5. Batter props (HR / total bases / RBI) were not pulled at all, nor were any NFL player props for the Melbourne game. On a five-game slate that is a real omission rather than a triage decision.
6. The White Sox starter could not be confirmed to a second source, voiding three markets in a game where bet365 held two of the best prices on the board.
7. The MLB runline orientation conflicts between sources. VegasInsider's Rangers–Mariners runline rows show Texas at −1.5 while Seattle is the moneyline favorite; ScoresAndOdds shows the logically consistent Seattle −1.5 / Texas +1.5. I did not resolve which display is correct, so I recommended no MLB runline in that game.
8. 2 of 7 games went unscreened (Rays @ Braves, Astros @ Phillies), both already in progress at 2:40 PM ET.
9. Team records differ trivially between sources (Mariners 67-78 on MLB.com vs 68-78 on the Stats API). Immaterial to any pick.

---

Star Rating Scale

---

Not logged or graded by this run. Logging and grading happen in the main chat.