MLB Over/Under Picks: Data-Backed Totals Strategy for Today’s Games

Editorial Team «Baseball Bets of the day» mayo 20, 2026 Tiempo de lectura: 17 min

Totals Are the Most Data-Friendly Market in Baseball

I will never forget the night I bet the over on a game at Wrigley Field and watched the wind shift 90 degrees between my bet placement and first pitch. The pregame forecast showed 15 mph blowing out to center field – a dream for over bettors. By game time, the wind had swung around and was blowing straight in from Lake Michigan. The game finished 2-1. That loss did not just cost me a unit – it reshaped how I approach totals betting entirely. Now, I check wind direction within 30 minutes of first pitch, not the day before.

Totals betting – the over/under on combined runs scored in a game – is the market where environmental data gives bettors the most measurable, repeatable edge. Unlike moneylines and runlines, which depend on which team wins or by how much, totals are driven by conditions that are quantifiable before the first pitch: wind speed and direction, park dimensions, pitcher profiles, and bullpen fatigue. You can crunch these numbers in a way that is not possible with team-versus-team outcomes, where clubhouse chemistry, hot streaks, and managerial decisions inject noise into the equation.

Over nine years of tracking totals bets, my highest-ROI season has always been the one where I leaned hardest into weather and park factor data. The US sports betting industry generated $16.96 billion in revenue during 2025, and a meaningful slice of that came from bettors treating totals as 50/50 propositions. They are not. The data says certain conditions systematically push games over or under the posted total, and the market does not fully price those conditions in.

What makes totals my favorite market is the layering effect. Each data point – wind, park factor, pitcher FIP, bullpen workload, temperature – adds a small increment of edge. Individually, none of them is a magic bullet. Stacked together, they create high-conviction spots that hit at rates well above the 52.4% break-even line. The 2,430 games in an MLB regular season mean I can be selective and still find two or three quality totals plays per day during the heart of the schedule. That selectivity is the foundation of everything in this article.

Wind Direction and Speed: The ROI-Proven Edge

Three years into my betting career, I had a conversation with a fellow analyst who told me he refused to bet any totals game without checking the weather first. I thought that was excessive until I saw his annual ROI numbers. Since then, wind has been the first variable I check for every totals play I consider.

The data is unambiguous. When wind blows in toward home plate at 5 mph or greater, unders have hit at a 55.1% rate across a sample of over 1,800 games, producing an ROI of 6.4%. That is an extraordinary edge in a market where 52.4% is the break-even threshold at standard -110 juice. The mechanism is intuitive: wind blowing in suppresses fly ball distance, turning potential home runs into warning-track outs and cutting the scoring expectation by a run or more.

On the other side, when wind blows out from the plate at 8 mph or greater, overs have connected at a 52.9% rate across more than 2,200 games, with an ROI of 3.6%. The threshold difference matters – you need a stronger outbound wind (8 mph vs. 5 mph for inbound) to get a statistically meaningful effect on overs. Moderate outbound wind at 5-7 mph does boost scoring slightly, but not enough to overcome the vig consistently.

How do I apply this in practice? I run a three-step filter. First, I check the wind report for the specific stadium – not the city-wide forecast, because stadium orientation varies. Wrigley Field, for example, faces northeast, so a north wind blows in while a south wind blows out. Second, I cross-reference the wind data with the posted total. A game with a total of 7.5 in a park where wind is blowing in at 12 mph is a higher-conviction under than the same wind conditions in a game totaled at 9.5, because the 9.5 already accounts for some offensive environment. Third, I confirm the starting pitcher matchup supports the weather thesis. Wind blowing in at a park where two contact-heavy pitchers are starting is an A-grade under spot. Wind blowing in when two fly-ball-heavy lineups are batting is even better.

One nuance that separates good totals bettors from great ones: dome and retractable-roof stadiums eliminate wind as a variable entirely. That sounds obvious, but I have seen bettors apply weather data to games at Tropicana Field or Minute Maid Park with the roof closed. If the roof is closed, weather does not exist for your totals model. Check roof status before building your thesis.

Park Factor Rankings and Their Impact on Totals

My first trip to Coors Field in Denver was an education. I watched batting practice balls sail into the upper deck like they were hit off a tee. At 5,280 feet above sea level, the thin air reduces drag on batted balls and causes breaking pitches to flatten. The effect on scoring is massive – Coors has consistently ranked as the most hitter-friendly park in baseball, with a park factor that inflates run scoring by 20-30% compared to a neutral venue.

Park factors measure how much a specific stadium inflates or deflates run scoring relative to league average. A park factor of 1.15 means 15% more runs are scored there than average. A factor of 0.88 means 12% fewer. For totals bettors, park factors are essential because they tell you whether the posted total already accounts for the venue – and how much.

The hitter-friendly parks that matter most for totals include the usual suspects: Coors Field in Colorado, Great American Ball Park in Cincinnati, Globe Life Field in Texas, and Fenway Park in Boston (especially for right-handed power). On the pitcher-friendly side, parks like Oracle Park in San Francisco, Petco Park in San Diego, and Dodger Stadium trend under. But the totals market is relatively efficient at pricing the most famous park effects. Everyone knows Coors inflates scoring. The edge is not in knowing that Coors is hitter-friendly – it is in knowing how park factors interact with the specific pitching matchup and the day’s weather conditions.

Here is where park factors create value. A mid-tier hitter-friendly park – say, one with a park factor of 1.08 – hosting a game between two fly-ball-prone pitchers on a warm, humid afternoon with wind blowing out. None of those factors alone would scream «bet the over.» But stacked together, they push the expected scoring environment well above the posted total. The market prices each factor partially but rarely prices the combination fully. I use park factors as a multiplier on top of my wind and pitcher analysis, not as a standalone signal.

Conversely, a pitcher-friendly park with a factor of 0.90 hosting two ground-ball pitchers on a cool evening with wind blowing in creates an under environment that the total often does not fully reflect. The posted total might be 7.5 when the expected scoring environment supports a 6.5. That half-run gap is where profit lives.

Matching Pitcher Profiles to Over/Under Lines

A few seasons back, I noticed something in my betting log that changed my approach. My under bets on games featuring two ground-ball pitchers were hitting at a significantly higher rate than my under bets on games featuring two strikeout pitchers. Both types suppress scoring, but ground-ball pitchers do it in a way that is more predictable for totals purposes. Strikeout pitchers can get burned by one mistake pitch – a solo home run that adds a run to the total without any base traffic. Ground-ball pitchers generate outs through contact, which tends to produce more consistent inning-to-inning run prevention.

Organizations that integrate sabermetric analysis into their decision-making have shown roughly 12% improvement in performance metrics. For totals betting, the relevant sabermetric indicators are ground-ball rate, fly-ball rate, hard-hit rate allowed, and FIP. A pitcher with a 50%+ ground-ball rate and a FIP under 3.50 is a strong under indicator. A pitcher with a fly-ball rate above 40% and a hard-hit rate allowed above 38% is an over indicator, regardless of what his ERA says.

The pitcher matchup filter I use for totals is different from the moneyline filter. For moneylines, I care about who is the better pitcher. For totals, I care about the combined run-prevention profile of both starters. Two mediocre pitchers with high walk rates and elevated fly-ball rates create an over environment even if neither team is particularly good offensively. Two quality starters with elite ground-ball rates create an under environment even in a hitter-friendly park.

I also track how deep each starter typically pitches into games. A starter who averages 5.2 innings per start turns the game over to the bullpen a full inning earlier than a starter who averages 6.2 innings. That extra inning of bullpen exposure matters for totals – middle relievers are generally less effective than starters, and the transition from starter to bullpen is the highest-scoring phase of most games. When both starters are short-outing pitchers, the total should be higher than the surface-level ERA matchup suggests.

How Bullpen Workload Tilts Late-Game Totals

I had a stretch in August 2023 where I went 2-7 on under bets over ten days. Every loss followed the same pattern: the starters held up their end, both teams were scoreless or low-scoring through five innings, and then the bullpens caved. Three-run seventh innings. Two-run eighth innings. Blown saves. I went back through the data and found the common thread – every one of those games featured at least one bullpen that had thrown 15+ innings over the prior four days. Fatigue was the variable I had missed.

Conor Yunits of the Sports Betting Alliance has argued that keeping prop bets on the regulated market is the only path to protect integrity and consumers. That philosophy extends to how bettors should treat bullpen data – transparency and access to workload information is a public good. Bullpen usage logs are freely available through MLB’s official game logs and sites like FanGraphs. Checking them takes five minutes and can save you from the exact trap I fell into.

For totals betting specifically, the bullpen workload filter works like this. If both teams have rested bullpens – their top three relievers have thrown fewer than 20 pitches in the last 48 hours – the under gets a boost, because fresh arms perform closer to their true talent. If either team has a taxed bullpen – multiple relievers used in high-leverage situations over consecutive days – the over gets a boost, because the manager will be forced to use lower-tier arms in pressure spots.

The timing of this check is critical. Bullpen usage from yesterday’s game may not be reflected in the morning total. If a team used four relievers for a combined 4.1 innings last night, and the total for tonight’s game was set before that information was fully digested, there is a window to bet the over before the market adjusts. I have found that totals set before 10 AM Eastern on days following extra-inning games or bullpen-heavy contests are the most likely to be mispriced.

There is another bullpen angle specific to totals that deserves its own mention: the series opener effect. When a team travels to a new city for the first game of a series, their bullpen has usually had a travel day to rest. That means the closer and setup men are fresh and available. First games of a road series have historically trended slightly under compared to the second and third games, when bullpen fatigue starts accumulating. I do not bet this pattern in isolation, but it reinforces my under thesis when the other factors – wind, park, pitching matchup – already point that direction.

Temperature and Humidity: Secondary Weather Factors

Wind gets the headlines, but temperature and humidity are the supporting actors that can tip a marginal totals play into a strong one. I did not start tracking temperature systematically until 2020, and the results were not as dramatic as wind data – but they were consistent enough to matter.

The physics are straightforward. Warmer air is less dense than cooler air, which means batted balls travel farther in heat. The effect is modest but measurable: games played above 85 degrees Fahrenheit average roughly 0.3-0.5 more runs than games played below 65 degrees, all else being equal. That half-run differential does not sound like much, but in a totals market where the vig requires you to win 52.4% of your bets, every fractional edge compounds over 162 games.

Humidity is the more misunderstood variable. The common assumption is that humid air is «heavier» and suppresses offense. The physics say the opposite – water vapor is lighter than the nitrogen and oxygen it displaces in the air, so humid air is actually less dense. In theory, high humidity should boost ball flight. In practice, the effect is tiny compared to wind and temperature, and I treat humidity as a tiebreaker rather than a primary factor.

Where temperature and humidity data become most useful is in combination with park factors. A 90-degree afternoon game at Great American Ball Park in Cincinnati – already one of the most hitter-friendly parks in baseball – with moderate outbound wind creates a scoring environment that can push the expected total a full run above what is posted. My highest-confidence over plays tend to be these triple-stacked spots: hitter-friendly park, warm temperature, and wind blowing out. The reverse combination – pitcher-friendly park, cool temperature, wind blowing in – produces my highest-confidence unders.

One practical note: afternoon games in open-air stadiums are more affected by temperature than night games. The temperature drops 5-10 degrees after sunset at most outdoor parks, and that decline coincides with the late innings when bullpens take over. An afternoon game that starts at 88 degrees stays warm through the entire contest. A 7 PM game that starts at 82 degrees might finish at 72 degrees. The late-inning temperature drop is a subtle under factor that I weight in my model but do not hear discussed often in totals analysis.

Totals Betting FAQ

Creado por la redacción de «Baseball Bets of the day».