⚾ MLB Bad Bets — The Board Is Back
Second half opens with a 10-game card, two locks, and the heaviest Kelly tag we've posted all season. Also: we're fading the team we called the best value in baseball yesterday.
⚡ The 30-Second Read
Baseball’s back. Day 51. The board goes up like it has every morning since May.
Ten games, and the model came off four days of rest swinging:
🔒 COL +13.2% — and Kelly wants 5.00 units on it. That is the single heaviest tag this column has posted all run, clearing the previous high (CHC 4.80u on Sunday). Coors Field, park factor 112, 91 degrees. The model has the Rockies at 64.4% against a market that’s got them at 51.2% — and it wants max exposure.
🔒 PIT +11.7% — third straight board backing Pittsburgh. Model flips off the -132 Guardians onto the +110 Pirates, who are 6-4 in their last ten at 6.4 RPG on a three-game win streak.
⚖️ AGREE and FLIP are dead even at 5-5. After the first half’s whiplash — flips leaking, flips redeeming, flips cratering — the model comes back into the second half with no lean at all. It’s just taking what’s there.
And the one worth sitting with:
🍺 We’re on Miami at +125 tonight, which means we’re fading Milwaukee. Yesterday we published a whole column arguing the Brewers are the best value on the World Series futures board — 16.5% sim probability against an +1100 price. Today the daily model says their -151 is too short.
Both are true. More on that below, because it’s the most useful thing on this card.
DAY 51 10 games · second half opens
LOCKS 2 COL +13.2% · PIT +11.7%
STRONGS 3 MIN · TB · WSH
AGREE/FLIP 5-5 ← no lean
TOP TAG COL 5.00u ← heaviest of the run
RECORD-TIER none · COL is the ceiling
LEDGER 352-276 56.1% +3.7pp (thru Jul 11)Seventy-four games left. First pitch is at 1:35.
🍺 The Milwaukee Thing
Let’s handle this head-on, because if you read yesterday’s column you’re entitled to ask.
Yesterday: the Brewers are the record-tier play on the futures board. ESPN’s sim gives them a 16.5% title shot; the market has them at +1100, which implies roughly 8% before you even strip the vig out. The market prices Milwaukee and Seattle one notch apart when the projection thinks Milwaukee is six times likelier. That’s a real gap and it’s still there this morning.
Today: the model is taking Miami +125 and laying off Milwaukee at -151.
These are not the same bet. They’re not even the same market.
A World Series future prices 74 games plus a four-round tournament. It’s asking whether this roster, this farm system and this run-prevention profile survive October. Over that horizon the sim says Milwaukee’s quality is badly underpriced, and it’s right.
A moneyline prices one night. Tonight is Sandy Alcantara against Henderson in a dome, projected 4.4 to 4.3 — a 0.15-run gap. That is a coin flip wearing a favorite’s price tag. The model has Milwaukee at 51.7% to win; the market has them at 57.5%. The Brewers are a good team being sold tonight at a great team’s number, and that’s the entire trade.
A team can be underpriced for the season and overpriced for the evening. Those are different questions with different answers, and if that sounds like a dodge, it’s the same logic as buying a stock you think is cheap and still not paying a premium for it on a green day. The season market is slow and narrative-driven. The nightly market is fast and reprices every twelve hours. We live in the fast one.
Kelly tags it at 2.51u. Miami’s 6-4 in their last ten at 5.2 RPG despite dropping three straight — the model reads that as noise, not decline.
🔒 The LOCKs
COL +13.2% vs CIN — AGREE Singer vs Hughes at Coors (PF 112, 91°F). The heaviest exposure this column has ever recommended: Kelly 5.00u, a full unit above anything we’ve posted. Model has Colorado at 64.4% against a market price of 51.2% — a 13.2-point gap and the widest edge on the board. Projection: COL 6.4, CIN 4.8, total 11.14. Live ML COL -115.
Here’s the part that makes this interesting. Our board spent the whole first half buying Colorado on the road — COL +8.1%, +10.8%, +14.8%, +6.4%, all at Oracle Park with the wind out — and mostly got beaten. Same team, different building. This is the Rockies at home, in a launching pad, at 91 degrees, against a Reds team that’s 43-52 and last in the NL Central. The model isn’t confused about Colorado being bad. It’s saying the price doesn’t account for where the game is being played.
The honest counter: this is a 39-59 team with the worst ERA in baseball. A 5-unit tag on the Rockies will make you feel insane. Size it anyway or don’t size it at all — but don’t half-do it.
PIT +11.7% @ CLE — FLIP Jones vs Williams (PF 98, in 5mph, 85°F). The model flips off the -132 Guardians onto the +110 Pirates — PIT 57.3% against the market’s implied CLE 54.4%. Pittsburgh is 6-4 L10 at 6.4 RPG on a three-game win streak (model +8%), and the projection has it PIT 4.8 – CLE 4.1. Over 7.5 also flags at a fat +13.4% (67%, -114).
Third straight board on the Pirates. We flagged the tension yesterday — ESPN’s second-half preview calls Paul Skenes’s season “head-scratching mediocrity” while our model keeps buying Pittsburgh — and note that Skenes isn’t even pitching tonight. The model likes them anyway. Pittsburgh’s tied for the most runs in the NL and owns the league’s fifth-best run differential. Somebody’s early here.
💪 The STRONGs
MIN +9.5% @ CHC — FLIP Ober vs Rea at Wrigley (PF 102, out 9mph, 93°F). Model flips onto the +119 Twins over Chicago’s -144. The weather’s doing real work: runs +7% vs neutral, total climbs to 11.36. MIN is 7-3 L10 at 5.0 RPG (model +5%). Projection is 5.9 – 5.5, which is close — but the price gap isn’t.
TB +8.1% @ BOS — FLIP Jax vs Bennett at Fenway (PF 107, out 5mph, 84°F). 1:35 first pitch — this is the game that opens our second half. Model flips onto TB +105 against Boston’s -126. Weather adds +4% to runs; total 10.39, over 9.0 flags at +6.6%.
Worth calling out: yesterday’s column flagged Tampa Bay as the one team we couldn’t price — fourth-best championship number in ESPN’s sim, best record in the American League, and nobody talking about them. First game back, the model takes them as a road dog. We asked the question yesterday and the board answered it this morning.
WSH +8.0% @ OAK — AGREE Cavalli vs Jump (PF 91, 62°F). Kelly 4.02u — second-heaviest tag today. Model has Washington at 58.5% vs a market at 50.4%, projected 4.4 – 3.6 in a low-total park (7.98). Live ML WSH -112.
The Nationals are 48-49 and lead the sport with 27 blown saves — second place has 19. That’s the bullpen thesis we’ve been screaming about since May, and it cuts both ways: it’s why Washington loses games it should win, and it’s why the market keeps mispricing them. Tonight the model wants real money on the road favorite anyway.
✅ SOLID (4–6%)
MIA +5.8% @ MIL (AGREE) — Alcantara vs Henderson, dome. Model MIL 51.7% vs market 57.5%. MIA +125. Kelly 2.51u. See above.
📎 LEAN (2–4%)
ATL +2.2% vs TEX (AGREE) — Quantrill vs Sale (PF 100, 91°F). Model ATL 65.9%, projected 5.1 – 3.6, but at -199 there’s not much price left. Kelly 1.53u.
NYY +2.0% vs LAD (FLIP) — Sasaki vs Cole, the marquee game of the night (PF 103, out 3mph, 89°F). Model flips onto NYY -107 over LA’s -112. Flagged: projected gap 0.2 runs — high variance, size small. Over 9.0 at +4.5%. Yesterday’s futures column had both these clubs priced fairly; tonight they’re a coin flip. Consistent.
⚪ PASS (<2%)
SD +1.9% @ KC (AGREE) — King vs Lugo. Book and model aligned, sharps on the same side, Kelly only tags it 0.98u (half-unit). Nothing to buy.
BAL +1.4% @ HOU (FLIP) — Kremer vs Lambert, dome. Baltimore’s 7-3 L10 on a four-game win streak (model +5%) — but the projected gap is 0.1 runs. That’s not an edge, that’s a rounding error. Pass.
🗓️ What The Second Half Is Actually About
Two things are going to distort every board we post between now and Labor Day, and it’s worth naming them on day one.
The deadline is August 3, and it is going to break our model. Every edge we price runs off a roster — pitcher composites, bullpen fatigue, lineup quality. In seventeen days a meaningful chunk of those rosters changes overnight. Tarik Skubal is the loudest domino: Detroit sits eight games under .500 with a fourth-place view of a division nobody’s running away with, and whether they move the best pitcher in the American League decides the AL Central and re-rates half a dozen contenders. The Mets are dismantling a $375 million roster. The Orioles, Angels and Rockies all have pieces moving. When a team’s ace leaves at 4pm, our 9am number was wrong and we won’t know it until we grade it.
And the pens are lying to us right now. The single most reliable input this column has is bullpen burn — relief pitches over the last three days, per-game workload, who’s unavailable. Four days off just zeroed that entire signal league-wide. It’ll take a week of games before those numbers mean anything again. Trust the top of today’s card. Be suspicious of the middle.
⚠️ Caveats
❌ Not betting advice. Model output.
💰 Kelly: 0.25×, 1u = 1% bankroll. COL at 5.00u is the largest tag in the history of this column. WSH 4.02u is second. Eight actionable games, two passes. If the top of this card scares you, that’s the correct response to a number that size — respect it or skip it, don’t split the difference.
🌋 No record-tier (≥15%) today. COL +13.2% is the ceiling.
⚖️ Flip watch: dead even at 5-5. First-half flip line was ugly in stretches (2-6 on the 11th, three flip LOCKs going 0-3), but the big flips carried the good days. PIT +11.7% and TB +8.1% are the ones with real weight today.
🥶 Rust is real and unpriced. Four days off, no live at-bats, bullpens fully rested for the first time since March. Every burned-pen edge we track just got reset to zero. Tonight’s card is the least informative one of the second half and the model can’t see that.
🆘 1-800-GAMBLER.
🔗 Full board + model JSON in Discord.
🏁 Where We Stand
352-276. 56.1%. +3.7 points above the 52.4% breakeven line. Fifty days of graded picks in public, through the worst stretch of the run — and the second half starts with the ledger green and the board loud.
Yesterday we spent 2,000 words explaining that futures are a sucker’s market and the Brewers are the best price on it. Today we’re fading the Brewers. That’s the whole discipline in one card: the horizon changes the answer, and pretending otherwise is how people lose money in two markets at once.
The model’s opening statement for the second half is a 5-unit tag on the Colorado Rockies. Not a sentence I expected to write. But the number is the number, Coors is Coors, and we’ve been grading this thing honestly for fifty days precisely so that it gets to say something uncomfortable on day fifty-one and we take it seriously.
First pitch: 1:35 ET, TB at Boston. Let’s get after it.
See you tonight in Last Night’s Tape.
— Bad Bets