David Hamilton cost the Brewers a run in the fourth inning of World Series Game 4 and got it back in the fifth. His throwing error let Junior Caminero score and put Tampa Bay ahead 2-1. An inning later, Milwaukee’s No. 9 hitter drove an 0-2 fastball into the right-field seats to tie it.
BBMI had the Rays through five innings, so that home run turned a winning pick into a push. The model also had Tampa Bay plus 1.5 runs, and that pick won when the Rays scored twice in the bottom of the 12th for a 4-3 win.
The next night, Jacob Misiorowski threw 5 2/3 scoreless innings and the Brewers beat the Rays 4-1 for their first World Series title. BBMI had the same two Rays picks in Game 5, and both lost.
Those four picks closed BBMI’s MLB season at 1-2-1. The rest of the window was football. NFL spreads and totals went 10-3-1, and college football went 12-11-1. Overall, BBMI went 23-16-3 from Oct. 27 through Nov. 8, up 4.7 units.
That is what this column is for. BBMI forecasts games in the NFL, college football, college basketball, Major League Baseball and college baseball, and runs a paper-tracked WNBA model. Every pick is logged before kickoff, tipoff or first pitch and graded from the final score, and nothing is removed from the record. Each edition goes through those results, the close finishes and the clear misses, whichever way they broke. This one covers games dated Oct. 27 through Nov. 8, following the previous installment.
The results, by product
The window finished up 4.7 units, +11.1 Units/100 on 42 picks, pushes included.
For the season, the overall premium record stood at 2,563-1,674-61 at +12.0 Units/100. Every product except NFL spreads was positive, and NFL spreads moved from −13.5 to −5.8 Units/100 over the window.
Snapshot: window results cover Oct. 27 through Nov. 8. Season figures are frozen at the Nov. 8 close. Units/100 is net units per 100 picks at one unit each; zero is break-even, and at -110 that takes a 52.4% win rate on decided picks. MLB rows are final for 2026. College basketball and college baseball rows show their most recently completed seasons; no 2026-27 college basketball picks were logged in this window. Records read wins-losses-pushes where pushes are reported. F5 means the first five innings; F9 means the full game.
Gap is the distance in points between BBMI’s projected margin or total and the final. For reference, the average gap across the window’s 31 spread picks was 11.6 points, and across its 7 totals picks 8.9 points.
NFL: 10-3-1 across spreads and totals
The spread picks went 2-0 in Week 8 and 3-1-1 in Week 9. Totals went 2-1 and 3-1. Together the two NFL products were 10-3-1.
Denver was in three of the wins. BBMI has taken the Broncos on the spread four weeks in a row and won all four. In Week 8, Denver plus 1 against Kansas City needed a stop on the last play. The Broncos trailed 24-13 early in the fourth quarter and went ahead 28-24, and Malcolm Roach sacked Patrick Mahomes on fourth-and-goal on the final play. A Chiefs touchdown there would have made the pick a loss. The over 42 in the same game won with 52 points.
In Week 9, Denver minus 3 won 26-21 at Carolina. The Rams covered twice, minus 6.5 against the Chargers (24-9) and minus 4.5 at Washington (20-13), and Detroit minus 6.5 won 19-9 at Miami.
The push came in Indianapolis, where BBMI had the Colts plus 4 against Dallas. Spencer Shrader’s 50-yard field goal put the Colts ahead 23-20 with 1:37 left, which had the pick winning. Javonte Williams then scored on a 1-yard run with 1:08 left, and Dallas won 27-23, exactly four points. The only spread loss was the Chargers plus 5, covered above.
The Week 8 totals wins were the overs in Chiefs-Broncos (52 against 42) and Patriots-Dolphins (43 against 41). Eagles-Commanders stopped at 37 against 40.5 as Washington won 27-10. In Week 9, the over 40 in Jaguars-Ravens won with 43, and the over 44.5 in Packers-Patriots won with 72: Jordan Love accounted for four touchdowns and Green Bay forced four turnovers in a 45-27 win. The under 52.5 in Bengals-Falcons, played in Madrid, won by a point and a half when Cincinnati won 34-17.
Broncos +1
vs. Chiefs · 28-24
W
Broncos −3
at Panthers · 26-21
W
Colts +4
vs. Cowboys · 27-23
P
Chargers +5
vs. Texans · 44-22
L
Over 44.5
Packers-Patriots · 72 points
W
Under 52.5
Bengals-Falcons · 51 points
W
For the season, NFL spreads stood at 26-27-5 and −5.8 Units/100. Totals stood at 28-19-1 and +13.5. The full log is on the NFL Model Performance page.
College football: 7-3-1, then 5-8
The premium picks went 12-11-1, down 0.1 units. Week 9 went 7-3-1 and Week 10 went 5-8.
The Week 9 wins included Georgia minus 11.5 against Florida (27-8), Louisville minus 22.5 against Stanford (52-24), Memphis minus 6 against Army (49-22) and Sacramento State plus 7.5, which beat Kent State 31-10.
Besides Troy’s comeback at James Madison, the Week 9 losses were SMU minus 9.5 at Syracuse, where the Orange scored the winning touchdown with 17 seconds left for a 30-27 win, and Indiana minus 20.5, which needed a 60-yard field goal to force overtime before beating Minnesota 23-16.
Kansas State plus 3 at Colorado pushed. K-State tied the game 21-21 with two seconds left on Blake Barnett’s 8-yard pass to Garrett Oakley. In overtime, the Wildcats kicked a field goal, and Colorado answered with Isaac Wilson’s 11-yard touchdown pass to Charlie Williams. A winning touchdown in college overtime ends the game without an extra point, so the final was 27-24, exactly three.
In Week 10, two of the Week 9 losers covered. James Madison minus 13.5 won 54-23 at Southern Miss, and SMU minus 6.5 beat Virginia Tech 51-28. UCLA plus 4 won 34-20 at Minnesota, North Texas minus 14.5 beat Rice 52-23 and UAB minus 8.5 beat Charlotte 23-12.
Three BBMI favorites lost outright. Oregon came back from two touchdowns down to beat Ohio State 41-28 in Columbus; BBMI had Ohio State by 14.6, a 27.6-point miss. Arkansas State, laying 14.5, lost 19-17 to UL Monroe on a 40-yard field goal and a game-sealing interception. Vanderbilt beat Mississippi State 25-23 for its first SEC win of the season. North Carolina beat UConn 38-21 behind three Billy Edwards touchdown passes in Bill Belichick’s return to New England, and BBMI had UConn plus 7.5.
Three more favorites won by less than the spread. Texas Tech, laying 16.5, held off West Virginia 53-48 as Thomas Castellanos threw five touchdown passes. Louisiana, laying 7, beat South Alabama 28-26. Massachusetts, laying 2.5, led Ball State 24-17 with 2:14 left. Ball State tied it with 25 seconds left, scored first in overtime and missed the extra point, and William Watson III’s 13-yard run and the kick won it for UMass 31-30, a point and a half short of the cover.
Kansas State +3
at Colorado · 27-24 OT
P
UMass −2.5
vs. Ball State · 31-30 OT
L
SMU −9.5
at Syracuse · 30-27
L
Louisiana −7
vs. South Alabama · 28-26
L
Texas Tech −16.5
vs. West Virginia · 53-48
L
Sacramento State +7.5
vs. Kent State · 31-10
W
The season record was 79-71-3, a 52.7% win rate and +0.6 Units/100, about one unit ahead. See the college football Model Performance page for every premium pick.
MLB: 1-2-1 to close the season
BBMI’s last four MLB picks of 2026 came in World Series Games 4 and 5. All four were on Tampa Bay: the Rays through five innings and the Rays plus 1.5 runs in each game. They went 1-2-1, down 1.4 units.
Game 4 went 12 innings. Cooper Pratt’s solo home run off Ian Seymour put Milwaukee ahead 3-2 in the top of the 12th, but the run line was safe either way, since Tampa Bay plus 1.5 covers a one-run loss. In the bottom half, Jonny DeLuca’s single tied it after a Joey Ortiz throwing error, and pinch-hitter Ben Williamson, 1 for 15 in the postseason, won it with a bases-loaded single over first base.
In Game 5, the Rays were +125 through five innings and -149 on the run line. Pratt singled in a run in the second, and singles by Pratt, Garrett Mitchell and Hamilton made it 2-0 in the fifth. William Contreras doubled in a run and Pratt added a sacrifice fly in the sixth. Chandler Simpson’s two-out double in the ninth was Tampa Bay’s only run. Pratt set a Series rookie record with six straight hits, and Jackson Chourio set a five-game Series record with 11 hits. The Brewers, who led the majors with 103 regular-season wins, went 11-2 in the postseason.
Rays F5
WS Game 4 · 2-2 after five
P
Rays +1.5
WS Game 4 · won 4-3 in 12
W
Rays F5
WS Game 5 · Brewers led 2-0 after five
L
Rays +1.5
WS Game 5 · lost 4-1
L
BBMI’s four MLB products finished the postseason a combined 19-7-3, up 7.0 units, +24.2 Units/100:
Postseason figures run from the Wild Card Series through World Series Game 5. Picks on the same game are listed separately, so this is a pick record rather than a count of independent games. Units/100 includes the three tied first-five games in the denominator.
All four MLB products finished the season positive, from +1.6 Units/100 on the full-game run line to +7.0 on the full-game moneyline. Final records: F5 moneyline 138-92-41, F5 run line 126-100-1, F9 moneyline 125-72 and F9 run line 259-141. Every pick is on the MLB Model Performance page.
Grading and calculation notes
Each pick risks one unit. NFL and college football spreads and totals, college basketball and college baseball picks, and the MLB first-five run line use -110 grading. A win at -110 is credited 0.91 units, as on the Model Performance pages. MLB moneylines and full-game run lines use the price posted with the pick. A losing pick loses one unit; a push scores zero.
Units/100 follows the Model Performance pages: every push counts as a pick at zero units. The window therefore divides by 42 picks: 39 decided picks plus three pushes, two in football and one in MLB.
The season total combines the products in the main table and matches the Overall premium record on the Model Performance hub in the Nov. 8 snapshot. It includes completed college basketball and college baseball seasons. Its third number counts every product’s pushes.
At -110, the break-even win rate is 52.4%. For products priced at varying moneylines, the return depends on the mix of prices, so a single fixed break-even win rate is less useful. F5 run lines are plus or minus 0.5 runs; full-game run lines are plus or minus 1.5.
Projection misses are the gap between BBMI’s number and the final margin or total. Game details are linked in the text.
What BBMI does
For every game, BBMI builds its own number, a point spread, a total or a run margin, from team and player ratings and game simulations, then compares it with the betting market. When the gap clears a threshold set in advance, the model logs a pick before the game starts. The thresholds, every logged pick and every result are public on the Model Performance pages.
01
BBMI builds its own number
Spread, total or run margin
02
Compares it with the market
Threshold set in advance
03
Logs the pick before the game
Only when the gap clears it
04
Grades it from the final score
Public on Model Performance
Each sport has its own Game Picks, Rankings and Playoff Pulse pages; start with this week’s NFL, college football and MLB picks. Longer analysis is in Research.
Next up
The next review picks up with Monday night’s Bills-Vikings game and covers NFL Week 10 and college football’s Week 11.