Rays vs Yankees Prediction, Odds: Expert MLB Pick for American League Division Series Game 3

If you want more Action Network content, be sure to add us as a preferred source on Google.

The New York Yankees host the Tampa Bay Rays on October 7, 2026. First pitch from Yankee Stadium is scheduled for 8:00 p.m. EDT. The game will be broadcast on HBO Max.

The Yankees are favored by -161 on the moneyline and by -1.5 on the run line. The Rays are +136 on the moneyline and +1.5 on the run line. The total is set at 7.5 runs.

Find my MLB picks and Rays vs Yankees prediction below, as well as probable pitchers, weather report, and more.

Rays vs Yankees Prediction

• Rays vs Yankees Pick: SGP: B. Rice, C. Bellinger, A. Wells 2+ Hits+Runs+RBI (+581, 0.2u)

My Rays vs Yankees best bet is on an HRR parlay. Make sure to find the best odds by checking our live MLB odds page, and make sure to download the Action Network app for more picks and analysis.

• Rays vs Yankees moneyline: Rays +136, Yankees -162
• Rays vs Yankees over/under: 7.5 (+108o / -132u)
• Rays vs Yankees spread: Yankees -1.5 (+140), Rays +1.5 (-170)

Rays vs Yankees MLB Betting Preview

Tampa Bay Rays Betting Preview

The Rays will send Nick Martinez to the mound on Wednesday, attempting to move on to the ALCS in just three games.

When traditional fans see that Martinez won 15 games with a 3.10 ERA, they probably wonder why the Rays held back such a great pitcher until Game 3.

While it’s true that Martinez doesn’t walk anyone (3.6%) and was an elite contact manager this season (32.6 HardHit%), leading all qualified starters in the former category and ranking fourth of 46 starters in the latter, other deficiencies kept his remaining non-pitching modeling estimators all more than 0.85 runs higher than actual results.

Most importantly, Martinez could not miss bats. With a 15.1 K%, 8.2 SwStr%, and 94.1 Z-Contact%, he was as poor at missing bats as he was great at avoiding walks and hard contact. He had the worst qualified pitcher strikeout rate, third-worst swinging strike rate, and third-worst zone contact rate.

Even with such a great contact profile, Martinez ran a 3.96 FIP and 3.98 xERA with 22 of 40 barrels (6.8%) leaving the yard. His more contact-neutral estimators ran as high as a 4.47 xFIP, 4.63 SIERA, and 4.92 dERA. A career-best .265 BABIP and 81.6 LOB% were doing a lot of the heavy lifting and probably aren’t sustainable, especially the latter number.

However, pitch modeling did give him credit for excellent command (62 botCMD was best among qualified pitchers, 107 Location+ was tied for seventh), despite below-average stuff (48 botStf, 98 Stuff+). The latter was a bottom-10 finish, while the former was merely a bottom-half finish among starters.

The caveat being that qualified pitchers are necessarily the best collection in the league; otherwise they wouldn’t be trusted with so many innings. There was only one qualified pitcher in the league with a below-average Bot ERA (Robbie Ray) and 10 with a below-average Pitching+.

If one had to choose, as ugly as it may look at times, you’d always choose the stuff pitcher first because it can’t be taught and can get away with more mistakes. In other words, Martinez can be effective, but needs to be near perfect to do so.

Martinez throws six different pitches, none more than one-third of the time, but only four of them more than 10% of the time, and his curveball and slider only 4% of the time each.

None of those six pitches reached 50 botStf grades, though only the two least used offerings fell below a 55 botCmd.

Stuff+ actually liked his changeup (109) more than Location+ (102), but Pitching+ only gave him credit for one other above-average offering (115 sinker). Martinez had an overall 3.55 Bot ERA and 104 Pitching+.

As far as who saw what, LHBs got changeups, sinkers, and cutters from 29% to 22% of the time in that order. RHBs saw 39% sinkers, 22% cutters, and 21% changeups. Batters from either side saw fewer than 5% curves and sliders and 11-14% four-seamers.

Here’s the crazy thing. While most of Martinez’s pitches moved about how you’d expect those pitches to move, his sinker had much more rise than average at 10.2 inches of iVB, allowing him to throw them higher in the zone, therefore tunneling off his changeup more effectively against LHBs.

The resulting reverse split should not be surprising:

LHBs: .286 wOBA, .297 xwOBA
RHBs: .307 wOBA, .321 xwOBA

Last season (2025) was Martinez’s first since returning to the league in 2022 with a reverse split, but only by 11 points.

He’s faced the Yankees three times this season, all with great surface results, but completely different peripherals:

• 4/11: 4.2 IP – 22 BF – 5 H – 1 R – 1 HR – 3 BB – 4 K – 6.7 GB% – 2 BR – 53.3 Hard%
• 5/22: 6 IP – 26 BF – 9 H – 1 R – 1 HR – 1 BB – 1 K – 37.5 GB% – 1 BR – 45.3 Hard%
• 9/22: 6 IF – 24 BF – 5 H – 2 R – 1 HR – 1 BB – 7 K – 26.7 GB% – 1 BR – 25 Hard%

The only patterns to be drawn are that home runs were the only real damage and there was a lot of elevated contact.

What I was most interested in were the lineups. Early in the season, the Yankees were healthier, and the reverse split may not have been as explicit. They went with six LHBs the first two times, but then later in the season only went with four (Ben Rice, Cody Bellinger, Luis Garcia Jr., and Spencer Jones).

However, the September game was part of a doubleheader, which really doesn’t tell us much.

Amed Rosario actually hit RHP better than LHP (131 wRC+ in 84 PA), and Giancarlo Stanton was at a 111 wRC+ in 66 PA, but no other right-handed Yankee with at least 60 PA against RHP reached a 90 wRC+ against them.

I don’t think the Yankees want either of those two players playing defense, which would mean pushing Ben Rice into the field. He’s not catching an elimination game, and I don’t think they want to pull Garcia (133 wRC+, .267 ISO vs RHP) from the lineup to accommodate Rice at 1B.

I could see Ryan McMahon sitting (86 wRC+ v RHP), but while neither Rotowire nor BARTOLO are considering the reverse split, both are projecting the eight-LHB lineup.

All but George Lombard Jr. (85 wRC+), the lone RHB, and Austin Wells (97) exceeded a 105 wRC+ at home. Only Rice (165), Bellinger (118), Garcia, and Chisholm (107) were above average against RHP. The projected lineup averaged a 98 wRC+ over the last 30 days of the season, ranging from Rice’s 170 wRC+ to McMahon’s 57 and Grisham’s small-sample -6.

None of the projected Yankees have hit what Martinez throws all that well either, with the exception of Spencer Jones, who smashed cutters (19.2 wFC/C) in a small sample size. Google AI tells me a total of 56 of the 1035 pitches he saw at the major league level were cutters.

Martinez was neutral at holding runners (0 Net Bases Prevented), with runners taking off on 1.5% of 810 opportunities. Nick Fortes, who generally catches against LHP, was slightly above average with 1 CS Above Average. He only threw out 20% of attempted stealers, but that was against an expectation of 19%. We’ve recently learned that stolen bases are more on pitchers than catchers.

At the team level, the Rays were an above-average defense with 10 Runs Prevented and 6 OAA, ranking 13th and 15th in the league, respectively. The downside of platooning as much as the Rays do is that they’ll often put a more marginal defense on the field against LHP.

They’re projected for -2 FRV (Fielding Run Value) on Wednesday, and that holds with Yandy Diaz’s 0 FRV at DH. It gets worse when you consider Jonny DeLuca is at 10 FRV on his own. In fact, the non-outfield portion of the projected Tampa Bay defense has accumulated -12 FRV.

BARTOLO ranks Tampa Bay the fifth-best bullpen in the league, a unit they’ve only had to use for three innings of Seymour, two innings of Baker, and two batters of Booser so far.

New York Yankees Betting Preview

Max Fried’s 2026 season was full of starts and stops. He missed two months from mid-May through the All-Star break and then had another short hiatus in August.

As a result, he only made 20 starts, his 112.2 innings being his second-lowest non-COVID season since becoming a consistent part of the Atlanta rotation in 2019. Fried faced more than 22 batters in only two of his last 11 starts, and they were his last two starts of the season.

Fried’s 16.4 K-BB% was above his career 15.1% rate, but since sitting between 17.5% and 19.9% each season from 2021 to 2023, he hasn’t exceeded 2025’s 17.2% since. His 49.1 GB% was also his first time ever below 51%.

The good news is that Fried had his best contact profile ever, allowing just 2.4% Barrels/BBE with a 29.8 HardHit%.

This combination produced results anyone who understands ERA estimators could see coming: a 2.65 FIP and 2.51 xERA below his 2.80 ERA, but a 3.21 dERA, 3.59 xFIP, and 3.78 SIERA well above it.

What’s most difficult to understand is the discrepancy between his 4.20 Bot ERA and 109 Pitching+ until you realize Fried threw six different pitches between 12% and 23% of the time. There’s just so much room for discrepancy with so many pitches.

However, LHBs saw 50% sinkers (RHBs 12%) with 19% sweepers and 16% four-seamers, with a smattering of cutters, curves, and more traditional sliders. RHBs saw cutters 28% of the time with less than 1% traditional sliders, but each of his other five pitches was between 10% and 18% of the time.

Both systems saw Fried’s stuff (54 Bot, 107 Stuff+) better than his command (51 Bot, 102 Location+), but while PitchingBot gave every offering overall grades between 47 and 53, Pitching+ loved the curveball (119), sweeper (112) and sinker (110).

Fried threw six innings of one-run ball with six strikeouts, allowing just three hits and a walk in the Wild Card round against the Red Sox. He allowed only three hard-hit batted balls, despite just 20% of his contact on the ground.

Fried threw the Red Sox only six sinkers, but no pitch more than 25 times. His velocity was up 0.7 mph (95.5) on his four-seamer, but did not reach a 0.5 mph difference on any other pitch.

Fried’s two starts against the Rays, including his last of the season in that same double-header, were his two longest of the season:

4/11: 8 IP – 30 BF – 6 H – 3 R – 0 HR – 0 BB – 6 K – 47.8 GB% – 0 BR – 37.5 HardHit%
9/22: 6.2 IP – 27 BF – 6 H – 4 R – 0 HR – 2 BB – 4 K – 70 GB% – 1 BR – 33.3 HardHit%

The takeaway here is that half the runners allowed scored, though Fried wasn’t much better than Martinez from a bat-missing perspective when he faced the Rays, with swinging strike rates of 8.2% and 8.5%, around two points below his 10.3% season rate.

When it comes to individual pitch matchups, Fried throws enough pitches that he can just avoid throwing what hitters hit well. For the Rays, that means don’t throw Junior Caminero any fastballs (3.63 wFA/C) and, surprisingly, don’t throw Ryan Vilade changeups (7.26 wCH/C).

Hopefully, the Tampa Bay lineup is easier to project against Fried, who doesn’t really have much of a split at all, at least by Statcast metrics:

• LHBs: .219 wOBA, .251 xwOBA
• RHBs: .253 wOBA, .245 xwOBA

Expecting the standard Tampa Bay lineup against LHP puts them in their worst split, averaging a 99 wRC+ on the road, 93 wRC+ (.124 ISO) vs LHP, and 83 wRC+ over the last month of the season.

Caminero exceeded a 120 wRC+ in all three categories and is the only projected Ray to have been above average in all three as well. After his .205 ISO vs LHP and Vilade’s .204, Diaz (.174) is the only other projected Ray above a .115 ISO against LHP.

The Yankee defense was just 17th in the league, with both -6 Runs Prevented and -5 OAA. Where they stand via FRV on Wednesday again depends on what they decide to do against Martinez. Rotowire and BARTOLO’s projected eight-LHB lineup sits at 10 FRV, and Bellinger (8) and Chisholm (8) will likely play regardless.

A Yankee bullpen that has been used much more extensively than the Rays' this post-season is BARTOLO’s 9th-ranked unit. Six relievers have been used between 1.1 and 4.1 innings, but Paul Blackburn is the only Yankee reliever to throw more than two innings so far this post-season.

Rays vs Yankees Pick, Betting Analysis

The Yankee Stadium run environment is 2% above average according to three-year rolling Statcast park factors. Left-handed home runs are boosted 20%, while right-handed home runs are boosted 15%.

The scheduled home plate umpire is fairly neutral from a run perspective, but boosts strikeouts by about 4%.

The early forecast (MySportsWeather.com & BARTOLO) suggests temperatures in the lower 60s with an 8-9 mph wind blowing out towards the left-field pole. The result would be a 1% to 4% reduction in scoring without much of an effect on home runs.

Both pitchers have top-of-the-league contact profiles, but while Fried walks twice as many batters as Martinez, his strikeout rate is also nine percentage points higher, and his career-low ground ball rate is still 13.6 points higher than Martinez.

I have Fried right around the halfway point between three-quarters of a run and a full run better than Martinez.

The Yankee lineup is the most difficult part of this analysis, but if it’s the eight-LHB version, I give the Yankees about a 10-point wRC+ edge, which would also give them a 12 FRV edge.

Not mentioned, these lineups would give the Rays a 12 BRR (Base Running Runs) edge, which probably almost evens out the Yankees' home-field edge, if that.

The BARTOLO bullpen gap for the Rays is four teams or 0.2 runs of weighted FIP.

I’m completely aligned with the market on the total and make the Yankees just barely above -150 for the game and a bit higher for the first five innings. Unfortunately, that aligns nearly perfectly with the market as well. Not even the prediction markets get below -150 on the Yankees.

As far as strikeout props go, I can see some very borderline value in Nick Martinez’s number with the eight LHB version of the Yankee lineup averaging a 24.5 K% vs RHP, but that’s only if he faces that version of the lineup AND goes at least 22 batters deep AND you can get +135 or better.

From a power perspective, Fried allowed four home runs all season (seven barrels). He’s smart enough not to throw Caminero fastballs if the Yankees even elect to throw him anything hittable at all.

Against Martinez’s reverse split, if Lombard is the only right-handed Yankee bat, he doesn’t exceed a 90 wRC+ at home, against RHP, or over the last month. The three Yankee home runs against Martinez were hit by Austin Wells, Austin Wells, and Cody Bellinger. Hmm…

Wells, 97 wRC+ and .156 ISO v RHP, but 148 wRC+ in September, is +590 (15%) to homer on Novig as of this writing (bet price), probably not enough value for a guy who will bat near the bottom of the order and only get 3-4 attempts (or two against Martinez).

He’s +230 for 2+ Hits+Runs+RBI on FanDuel, though, where Rice is +125, and Bellinger is +145.

To parlay those three, you’d get +581.

I don’t want to leave readers with nothing, which is why I wrote so many words, but I can’t find the value in any part of this game otherwise, at least not without a confirmed Yankee lineup, and I doubt even that would change much.

Just for fun, let’s go with the three-player H+R+R SGP, including every Yankee who has homered against Martinez this year, plus another with a double in their last meeting.

Pick: 

• Ben Rice 2+ Hits+Runs+RBI
• Cody Bellinger 2+ Hits+Runs+RBI
• Austin Wells 2+ Hits+Runs+RBI

Parlay Odds: +581 (FanDuel) | 0.2u