SlatePicksResearched, simulated 10,000 times, and graded for you
We research every game on the slate, factor in what moves the line (matchups, injuries, schemes, home and away splits, weather, rest and travel), and simulate each game 10,000 times with those factors accounted for. Then we grade every bet green, yellow or red against the sportsbook's price, so you can see at a glance where the value is.
These aren't guaranteed locks, because in reality nothing in gambling is. It's just honest math, and we keep it transparent: every green bet is logged on the Record tab, wins and losses alike.
Every sport runs through the same model. A set of factors built for that sport is scored game by game, turned into adjustments, played out 10,000 times, and graded against the sportsbook's price. No pick comes from a hunch, and none comes from repeating what TV analysts are saying.
Start with the market
The betting line is the sharpest public estimate there is, so it's the baseline. The spread, total and moneyline set each team's expected score before any of our research touches it.
Score the factors
Our algorithm works through a checklist of factors for each sport: matchups and head-to-head history, strategy and schemes, injuries and lineups, home and away splits, rest and travel, weather and venue, and recent form. Each one is researched game by game, not pulled from season averages.
Research the player props
Props get their own pass: role and usage, the opponent's weakness at that spot, and how the player has done in this matchup before. Long shots like first touchdown scorer and home runs are researched separately from everything else.
Turn research into adjustments
The factors become specific, measured adjustments to each team's expected score and each player's expected stats. They're kept deliberately modest, because the market is usually close and the edge lives in the details.
Simulate 10,000 games
Every game is played out 10,000 times with a simulation built for how that sport is scored. Football games are simulated by score, with both teams' outcomes linked through game script. Baseball games are played inning by inning, with starters and bullpens pitching their own innings. Every bet is graded from the same set of sims.
Grade against the price
We count how often each bet wins across the 10,000 games and compare it to how often it needs to win to make money at its price. The size of that gap decides the grade: green, yellow or red.
Log every result
Every green bet goes on the Record page before the game starts and gets graded after it ends, wins and losses alike. Nothing is deleted or regraded.
Clears its break-even by a real margin: at least 1 to 1.5 points on game lines, and at least 4.5 points on player props, touchdown and home run bets, which carry more juice and more randomness.
Lean
Right around break-even. Small stakes, or wait for a better number.
Pass
No edge. The market has this one priced about right.
Each chart shows every outcome the sim considers. The football marks the projection, the yellow line is the betting line, and the shaded side is where the bet wins. Touchdown and home run bets show how often the player scores 0, 1, 2 or 3+ times.
NFL
When NFL picks post
Thursday night game: Wednesdays around 4 PM ET, after the first injury report of the week.
Sunday and Monday games: Saturdays around 11 AM ET, after Friday's final injury reports.
Results: graded every morning around 9 AM ET on the Record tab.
Every line and price on this page is the one we pulled at posting time. Lines move after that, so check your sportsbook before you bet.
Green bets
Every green bet on the slate, game lines and player props together, in kickoff order. Each game is played out 10,000 times. The percentage is how often the bet won, color coded by grade, with the raw count in parentheses. The break-even number is how often it needs to win to make money at its price. On each chart, the football is the projection, the yellow line is the betting line, and the shaded side is where the bet wins.
College football: Saturdays around 9 AM ET, before the noon kickoffs. Ranked games plus notable lines.
College basketball: schedule announced when the season starts.
Results: graded every morning around 9 AM ET on the Record tab.
Every line and price on this page is the one we pulled at posting time. Lines move after that, so check your sportsbook before you bet.
College coverage is game lines only: spreads, moneylines, over/unders and team totals. We don't post player props for college games.
Green bets
Every green bet on the slate, spreads, moneylines and totals together, in kickoff order. Each game is played out 10,000 times. The percentage is how often the bet won, color coded by grade, with the raw count in parentheses. The break-even number is how often it needs to win to make money at its price. On each chart, the football is the projection, the yellow line is the betting line, and the shaded side is where the bet wins.
We started keeping this record when SlatePicks launched in beta on October 4, 2026, and every green bet since then is on this page. Only green bets count, game lines and props alike. Each one is logged before kickoff at the line and price it was posted, risks 1 unit, and stays on this page win or lose. Nothing gets deleted or regraded after the fact.
Injuries: if a player is ruled out before kickoff, books void his props, and so do we (no win, no loss). If a key player gets hurt during the game, the bet still counts in the official record, but it's tagged, and we also show the record without tagged bets. The tag follows a fixed rule from the box score and applies to wins and losses alike: a prop's player leaving before halftime (football) or a starting pitcher leaving before the 5th inning (strikeout and outs props), or a starting quarterback or starting pitcher leaving before halftime or the 3rd inning for game lines. Players already listed as out or doubtful when the pick was posted never count.
Awaiting results
Bet
Game
Kickoff
Sim
Sport
Graded picks
Bet
Game
Final
Result
Units
Sim
Sport
MLB
When MLB picks post
Every game day: around 11:30 AM ET, once probable pitchers and lines are out. Games without lines yet show as previews until they open.
Refreshes: if a slate is re-run later in the day for late lines or news, the time below updates.
Results: graded every morning around 9 AM ET on the Record tab.
Every line and price on this page is the one we pulled at posting time. Lines move after that, so check your sportsbook before you bet.
Where things stand
Green bets
Every green bet on today's slate, game lines and props together, in first-pitch order.
Every game, in first-pitch order
LearnBetting explained in plain English
Behind the scenes
How the system works, and why
This is the long version, written for skeptics. We don't sell locks, and nobody honest can. What we do is closer to counting cards in blackjack: find the spots where the odds lean slightly in your favor, bet only those, and let a small edge play out over hundreds of bets. Here's exactly how that works, including the parts that don't.
The idea
The short version
Every game has a line: the sportsbook's best guess at how it will play out. We start with that line, then look for things it may not have fully accounted for, like an injured lineman, a tired bullpen, or a bad weather forecast. Those findings nudge the numbers slightly.
Lines also aren't set in stone. They move all week. The sportsbook adjusts them when news breaks, like an injury or a weather change, and when too much money piles onto one side. Big bets from sharp bettors can move a line on their own, sometimes within minutes. That's why every pick on this site shows the line and price we used, and why it's worth checking your sportsbook before you bet. If the line has moved, the edge may be smaller, gone, or even bigger. This is called line movement, and getting a better number than the market settles on is one of the clearest signs of a real edge.
Once we've adjusted the numbers, a computer plays the game 10,000 times, with all the ups and downs of a real game built in. That's called a simulation. Each pick's percentage is simply how often it won across those 10,000 games.
If a bet won often enough to beat its price, by a real margin, it's green. If it's close, it's yellow. If not, it's red, and we pass. That's it. No gut calls, no "trust me," and no hiding the losses.
Why this is closer to counting cards than picking winners
A card counter at a blackjack table can't tell you what the next card will be. What counting does tell them is when the cards left in the deck slightly favor the player. Even then, the edge is small, often around 1 percent, and they still lose plenty of hands. They come out ahead over time because they bet when the edge is there, sit out when it isn't, and play enough hands for that small edge to add up.
Sports betting works the same way. Sportsbooks build a small fee, called the juice, into every price. That's why a bet at -110, about as close to a coin flip as betting gets, has to win 52.4% of the time just to break even. Most bets are priced about right, so someone who bets on everything slowly loses to the juice. Our job is to find the few bets where the price looks wrong by enough to beat that 52.4%, and skip the rest. That's why most of what we grade every week is red.
What this means for you: a green bet that wins 56% of the time still loses 44% of the time. You will see green bets lose, sometimes several in a row. One bet tells you almost nothing about whether the method works. Hundreds of bets do, which is why every result, win or lose, goes on the Record page.
Underlined words have quick definitions. Hover over them, or tap them on a phone.
The method, step by step
1Start from the market
We don't build a game from scratch and hope it's right. The betting line is the sharpest public estimate available, shaped by bookmakers and by bettors who put real money behind their opinions. So every simulation starts there.
A spread and a total together tell you what the market expects each team to score. A team favored by 3.5 with a 43.5 total is expected to score about 23.5 to 20. In baseball, we take the moneyline, remove the bookmaker's cut to get a fair win chance, and split the posted total into expected runs for each team.
Starting from the market keeps the model honest. If our research found nothing, the simulation would land right on the line, and every bet would grade as a pass. Any edge has to come from something specific.
2Research becomes small, specific adjustments
Each sport has its own checklist: head-to-head history, schemes and strategy, injuries (with extra weight on both lines in football), lineups, starting pitchers and bullpen workload in baseball, home and away splits, rest, travel, weather and venue. That research gets turned into small, measured nudges to the market's numbers:
A top receiver ruled out lowers his team's expected score by about a point.
Both starting centers out trims a running back's projected rushing yards by a few yards.
A team's backup corners facing a top receiver adds a few yards to his projection.
A tired bullpen raises the expected runs it allows in the late innings.
The adjustments are kept deliberately modest. Markets are usually close, so a model that moves lines by touchdowns is almost always wrong. The edge lives in the details the line hasn't caught up to yet.
3Play the game 10,000 times
This is a Monte Carlo simulation, named after the casino. Instead of trying to calculate an exact answer, you let a computer play out the game thousands of times with realistic randomness built in, then count what happened. Flip a fair coin 10,000 times and you'll land very close to 50% heads. Simulate a game 10,000 times and you get a reliable picture of every likely outcome, not just the most likely one.
Football
Each team's score is drawn around its adjusted expected score, with a realistic spread of outcomes (NFL scores typically land within about 9 to 10 points of expectation; college scores swing wider). A shared "game script" factor links the two teams the way real games do: when one team runs away with it, the other usually falls behind. Scores are whole numbers, so pushes on 3 and 7 happen exactly as often as they should.
Baseball
Games are played inning by inning. The starter pitches his expected innings, then the bullpen takes over at its own rate. Runs in each half-inning come from a distribution that matches real baseball: lots of scoreless innings and the occasional crooked number. A shared factor moves both teams together for things like wind and the umpire's zone. Tied after nine? Extra innings get played. First-five-innings and first-inning (NRFI) bets come from the same 10,000 games.
Player props work the same way: each player's stat is simulated around the market's line plus our research adjustment, with a realistic range for that stat, and we count how often it clears the line.
Where we estimate instead of simulate: touchdown scorers, home runs and some moneylines aren't played out play by play. For those, we start from the price, remove the bookmaker's cut, and apply our research adjustment. We label those as estimates on every pick, so you always know which is which.
One more check: before any research is added, the baseball model is calibrated so that, with no adjustments, overs and unders split the way the market prices them. If the model leaned one way on its own, it would create fake edges. Calibration removes that.
4Compare the probability to the price
A bet is only good or bad relative to its price. Every price implies how often the bet has to win for you to break even:
PriceRisk to win $100Must win
-200$20066.7%
-150$15060.0%
-110$11052.4%
+100$10050.0%
+150$100 (wins $150)40.0%
+400$100 (wins $400)20.0%
So a bet that wins 60% of the time is a bad bet at -200 and a great one at -110. That's why every pick on this site shows two numbers: how often it won in our simulations, and how often it needs to win at its price. The gap between them is the edge.
5Grade it, conservatively
Green means the bet clears its break-even by a real margin: about 1 to 1.5 points on game lines, and at least 4.5 points on player props, touchdown and home run bets. Props get a bigger cushion on purpose. They carry more juice, the inputs are noisier, and a model is more likely to be wrong about one player's stat line than about a whole game. Yellow is close to break-even. Red is a pass.
Edges on game lines are thin by design. A model that regularly claims 65% on spreads isn't finding edges, it's broken. Thin, honest edges are what beating a sharp market actually looks like.
Keeping it honest
Why a winning system still has losing weeks
This is the part most pick sellers won't show you. Even a real edge produces long losing stretches, because each bet is close to a coin flip. Try it yourself: this runs 30 bettors who each make the same number of bets at -110 with the win rate you choose, and charts every one of them.
Win rate
Bets per bettor
Notice two things. At 52.4% the bettors scatter evenly around zero, because the juice eats the edge. At 55% most end up ahead over 500 bets, but plenty are underwater after 100. That's what a real edge feels like from the inside, and it's why we track everything in units and why a few bad days don't change the method.
Another visualization
One player prop, played 10,000 times
Here's step 3 in action on a single bet. The example: LeBron James, 24.5 points, over or under at -110. These numbers are made up for illustration, not a current pick. The market's line is the starting point. Our research might nudge his projection up (say the opponent's best perimeter defender is out, or it's a fast-paced matchup) or down (say it's the second night of a back-to-back, or there's talk of a minutes limit). Pick an adjustment, hit simulate, and watch 10,000 games fill in. When the adjustment is negative, the sim evaluates the under instead.
Research adjustment
Each bar counts how many of the simulated games landed in that 2-point range. Green bars are games where the side we'd bet hit (the over for an upward adjustment, the under for a downward one), the yellow line is the betting line, and the basketball marks his projection. With no adjustment, the sim lands right on the line and the bet is a coin flip that loses to the juice. The adjustment is the whole edge, which is why we keep it small and specific.
What this system can't do
It can't see the future. A 56% bet loses 44% of the time, full stop.
It's only as good as its inputs. The research adjustments are judgments, made by a consistent process but judgments all the same. If the research is wrong, the edge is wrong.
Prices move. Our grades apply to the line and price we posted. If the line moves before you bet, the edge may be smaller, gone, or bigger.
Late news happens. A scratch an hour before kickoff can change everything after we've posted.
Past results don't guarantee future ones. A good record over a few weeks can be luck. So can a bad one.
How to check our work
Every green bet is logged before the game, at the posted line and price, and graded after, wins and losses alike. Nothing gets deleted or regraded.
Injuries follow a fixed rule, applied to wins and losses equally, and we show the record both with and without injury-affected bets.
Every pick shows its math: the simulated percentage, the raw count out of 10,000, the break-even for its price, and a chart of the full range of outcomes.
The reds are public too. You can see every bet we graded and passed on, not just the ones we liked.
The bottom line: we're not promising winners. We're showing you exactly where we think the price is wrong, how wrong, and what happened when we said so.
Bet only what you can afford to lose. If gambling stops being fun, call or text 1-800-GAMBLER.
Betting terms
New to betting, or just want to know what a number means? Every term on SlatePicks and most of what you'll see at a sportsbook, explained simply, with examples. Search for a term or browse by topic.