2026 NL Rookie Of The Year: 5 Models Fade The Market Favorite

The 2026 NL Rookie of the Year odds tell one story; five models tell another. The National League race is two-thirds run, and Kalshi has split it into 30 separate contracts, 29 named candidates plus a tie market. The market has made up its mind: Cardinals infielder JJ Wetherholt is the favorite near 51 cents, with Reds slugger Sal Stewart the only other name in the twenties. So we asked five models to score the whole class without letting any of them see a price. They came back with a very different board. They fade the market's favorite all the way to a single-digit number, and they put a rookie the market has at a dime right at the top.

The Quick Answer

Kalshi's market has JJ Wetherholt (St. Louis Cardinals) as the favorite near 51 cents, with Sal Stewart (Cincinnati Reds) next around 27 cents and Nolan McLean (New York Mets) a distant third near a dime. Our price-blind, five-model panel flips that order. It blends McLean to 29.5% and Stewart to 28.5% as near-even co-leaders, and it prices the market's favorite, Wetherholt, at just 7.6%, the widest gap on the whole board.

• Market Pick: JJ Wetherholt, about 51¢
• Model Pick: Nolan McLean, 29.5% blend, priced near 10¢ by the market
• Biggest Model-Vs-Market Gap: JJ Wetherholt, 7.6% blend against a 51¢ price
• The One Name Both Sides Like: Sal Stewart, 28.5% blend against a 27¢ price
• The Honest Caveat: the models are reading July stat lines, not a November ballot, and voters weigh pitchers and hitters in ways no model can pin down.

The two rookies the panel splits the crown between, the favorite it fades, the full 30-contract board, and the spots where the five models disagree are all below.

The Market At A Glance

Venue	Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of July 2026)
The Event	"NL Rookie of the Year Winner" for the 2026 MLB season, 29 separate player contracts plus a Tie/Co-Winners contract (Kalshi event KXMLBNLROTY-26)
The Contract	Each market resolves YES only if that player wins the 2026 NL Rookie of the Year, per Kalshi's settlement rules
Award Date	The Baseball Writers' Association of America votes after the regular season; the winner is announced in November 2026, with the market listed to settle into December 2026
Prices Below As Of	July 30, 2026

These are model estimates, not predictions of fact and not financial advice. Kalshi event contracts trade on a CFTC-regulated exchange; you must be 18 or older and in an eligible state to participate.

What Wins A Rookie Of The Year

Two-thirds of the season is already in the books, so this is not a projection contest. It is a running grade on who has actually built a Rookie of the Year resume, and three things shape almost every vote:

• Accumulated Production With A Runway. An everyday bat or a full-time rotation starter banks counting stats every week. A reliever, a bench piece, or a prospect still in the minors cannot catch up in the time left.
• Rate And Role Together. A strong average or ERA only matters if the playing time is there to back it. Voters reward the rookie who pairs good numbers with a real job.
• Narrative And Platform. A contending team, a big market, and national attention all move ballots, which is why the same line can poll differently depending on the uniform.

One more rule shaped the board: rookie eligibility follows MLB's service-time limits, and several listed candidates carry prior-season debut dates. Every name here still qualifies (McLean, for one, threw exactly 48 MLB innings in 2025, two under the 50-inning line). The models weighed each player's actual 2026 major league line, not spring-training hype, and the deep prospect names paid for it.

Here is the full field, sorted by the model blend, with each model's raw number, the market price, and the gap between them. Edge is the blend minus the market price; a positive number means the models see the player as more likely to win than the price implies. Player lines are 2026 major league totals fetched on July 30, 2026.

Candidate (2026 MLB Line)	Market	Opus	Sonnet	GLM	Kimi	DeepSeek	Blend	Edge
Nolan McLean (NYM P): 21 GS, 119.1 IP, 3.32 ERA, 142 K, 7-7	10¢	24.5	32	23	27	41	29.5%	+20
Sal Stewart (CIN 1B): 107 G, .253, 22 HR, 79 RBI, .802 OPS	27¢	25.5	21	28	32	36	28.5%	+2
T.J. Rumfield (COL 1B): 106 G, .293, 13 HR, 53 RBI, .843 OPS	3¢	13.3	13	14	18	10	13.7%	+11
JJ Wetherholt (STL 2B): 103 G, .251, 15 HR, 40 RBI, .736 OPS	51¢	12.2	7	8	9	2	7.6%	-43
Konnor Griffin (PIT SS): 59 G, .276, 5 HR, 25 RBI, .736 OPS	1¢	9.2	4	5	2	1	4.2%	+3
Bryce Eldridge (SF DH): 66 G, .250, 11 HR, 26 RBI, .805 OPS	3¢	3.1	5	3	3	6	4.0%	+1
Carson Benge (NYM RF): 104 G, .262, 11 HR, 41 RBI, .717 OPS	1¢	4.1	6	5	3	2	4.0%	+3
Joe Mack (MIA C): 64 G, .237, 10 HR, 30 RBI, .724 OPS	1¢	1	2	3	1	1	1.6%	+1
Owen Caissie (MIA RF): 80 G, .239, 12 HR, 50 RBI, .756 OPS	1¢	2	2	2	1	0	1.4%	0
Justin Crawford (PHI CF): 99 G, .257, 2 HR, 23 RBI, .645 OPS	1¢	1	2	2	1	0	1.2%	0
Ryan Waldschmidt (ARI CF): 44 G, .276, 1 HR, 13 RBI, .718 OPS	1¢	1	1	1	0	0	0.6%	0
Bubba Chandler (PIT P): 20 GS, 104.2 IP, 4.64 ERA, 93 K, 3-8	1¢	0	1	1	1	0	0.6%	0
Alex Freeland (LAD 2B): 72 G, .231, 3 HR, 19 RBI, .622 OPS	2¢	0	0	1	0	0	0.2%	-2
Thomas White (No 2026 MLB Line Yet, Triple-A)	2¢	0	0	1	0	0	0.2%	-2
Moises Ballesteros (CHC DH): 59 G, .231, 6 HR, 23 RBI, .688 OPS	1¢	0	0	1	0	0	0.2%	-1
Andrew Painter (PHI P): 12 GS, 65.0 IP, 7.06 ERA, 53 K, 1-8	1¢	0	0	1	0	0	0.2%	-1
Rhett Lowder (CIN P): 15 GS, 77.0 IP, 5.61 ERA, 65 K, 3-7	1¢	0	0	0	0	0	0.0%	-1
Chase Petty (CIN P): 2 GS, 30.0 IP, 4.50 ERA, 14 K, 1-2	1¢	0	0	0	0	0	0.0%	-1
Jhostnyxon Garcia (PIT RF): 14 G, .184, 0 HR, 2 RBI, .436 OPS	1¢	0	0	0	0	0	0.0%	-1
JR Ritchie (ATL P): 7 GS, 54.0 IP, 4.50 ERA, 49 K, 1-2	1¢	0	0	0	0	0	0.0%	-1
Hunter Barco (PIT P): 0 GS, 11.2 IP, 7.71 ERA, 8 K, 0-1	1¢	0	0	0	0	0	0.0%	-1
Jonah Tong (NYM P): 0 GS, 10.0 IP, 3.60 ERA, 7 K, 1-1	1¢	0	0	0	0	0	0.0%	-1
Robby Snelling (MIA P): 1 GS, 5.0 IP, 5.40 ERA, 2 K, 0-1	1¢	0	0	0	0	0	0.0%	-1
Jett Williams (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1
Zac Veen (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1
Aidan Miller (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1
Charlie Condon (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1
Jaxon Wiggins (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1
Quinn Mathews (No 2026 MLB Line Yet, Triple-A)	1¢	0	0	0	0	0	0.0%	-1

These are model estimates, not predictions of fact and not financial advice. Market figures are Kalshi YES mid-prices from July 30, 2026; the top of the board is actively traded, while most deep-field contracts sit near the 1-cent floor on light volume. The Blend is the plain average of the five model columns; each model also assigns a small residual to an unlisted field bucket, so the shown blends across the board sum to about 98 percent rather than exactly 100. Rows below the named contenders show each model's small residual mass, so their per-model figures are near-identical by construction rather than independent reads on each deep-field name.

A note on the panel. Our standard Model Verdict lineup runs eight seats. On this run three were unavailable: the ChatGPT (codex) and Gemini (agy) command-line seats were not installed on the machine that produced these verdicts, and the Claude Fable seat hit its usage cap and returned no answer. This board therefore runs on five models: Claude Opus, Claude Sonnet, GLM, Kimi, and DeepSeek. All five scored the field on July 30, 2026, without seeing any prices, working from each candidate's fetched 2026 major league line. Every model's number here also lands in our back-tested Model Verdict ledger, where past price-blind calls are graded against settled results so the method is measured, not just asserted.

The gap that defines this board: the market pays 51 cents for JJ Wetherholt and a dime for Nolan McLean. The models do the opposite. Read the divergence as pedigree versus production, and keep reading to see which rookies the panel favors.

Nolan McLean And Sal Stewart: The Panel's Co-Favorites

The models do not crown one rookie; they crown two, nearly even at the top. McLean carries the highest blend at 29.5%, and Stewart sits right behind at 28.5%. Every seat named one of the two first.

The case for McLean is the arm. He has made 21 starts for the Mets, throwing 119.1 innings with a 3.32 ERA, a 1.11 WHIP, and 142 strikeouts. That is a full starter's workload, the strikeout rate voters love, and a big-market platform on a contending team, all at once. DeepSeek, the model highest on him, put it flatly: "Nolan McLean's 3.32 ERA, 1.11 WHIP, and 142 Ks over 119.1 IP make him the clear frontrunner in a weak NL rookie class." Sonnet framed the ballot case: "a sub-3.35 ERA with a strong strikeout rate and tight WHIP over a near-full starter's workload gives voters the classic best-pitcher rookie case, boosted by the Mets' competitive context." At a dime, the market has the field's most dominant arm priced third.

The case for Stewart is the bat, and it is the one place the market and the models mostly agree. Stewart has played 107 games as Cincinnati's everyday first baseman, hitting .253 with 22 home runs, 79 RBI, and an .802 OPS. That is the loudest power-and-counting-stat package of any rookie bat in the class, and the everyday job means it keeps growing. Opus called it "the counting-stat profile voters reward most in a rookie hitter," and Kimi projected him toward roughly 32 home runs and 115 RBI if the Reds stay in the race. At 27 cents against a 28.5% blend, his price is close to fair, which makes Stewart the panel's floor more than its edge.

JJ Wetherholt, Cardinals: The Favorite The Models Fade

Now the other side of the board. Kalshi has Wetherholt near 51 cents, the market choice. The panel does not see it, blending him to just 7.6%, a gap that is the largest anywhere on this board.

The reason is the line. Wetherholt is hitting about .251 with 15 home runs, 40 RBI, 11 steals, and a .736 OPS across 103 games as the Cardinals' everyday second baseman. That is a useful, full-time big-league season. In the models' reading, it is not a standout rookie campaign. Opus, the model most generous to him, still had him fourth and called him "a steady fourth" behind the two leaders. The market may be pricing reputation and a strong prospect pedigree the models cannot see. The panel is pricing the .736 OPS in front of it. Neither number is a prediction of fact, and the models never saw the price, so read the gap as a difference of method rather than a signal to act.

T.J. Rumfield: The Value Bat With An Asterisk

The panel's other value name is T.J. Rumfield, blended to 13.7% against a 3-cent price, and he is where the hitters split most. Rumfield has hit .293 with a .843 OPS, 13 home runs, and 53 RBI over 106 games as Colorado's everyday first baseman, the shiniest rate line among the qualifying rookie bats. The hesitation is context: four of the five models independently raised the Coors Field question. As GLM put it, he "has the best rate line but plays at Coors Field, which voters discount, and his 13 HR from a 1B is underwhelming relative to Stewart." Sonnet reached the same place, noting "Coors Field inflation and a losing Rockies team likely cost him votes relative to raw numbers." The edge is real at 3 cents, but it rests on voters looking past a park effect they usually punish.

Where The Models Split: The Traps

A blend can hide a fight. On several names the five models disagreed enough that the row is worth trusting least:

• McLean Vs. Stewart At The Top. Sonnet and DeepSeek push McLean clear on the strength of the pitching line (Sonnet 32-21, DeepSeek 41-36). Opus, GLM, and Kimi all keep Stewart even with or ahead of McLean, valuing his 22 home runs and 79 RBI as the loudest counting-stat resume among the hitters. Kimi named the exact fault line: "history shows good-not-historic pitchers lose to full-season hitters, and innings limits plus injury risk loom over his final ten starts." If you think voters reward the bat, you side with Stewart. If you think the arm wins, you side with McLean.
• Rumfield And The Coors Question. His .843 OPS is the best rate line on the board, but the park behind it splits the room, from Kimi's 18% to DeepSeek's more cautious 10%.
• The Prospect Names The Panel Will Not Rank. Andrew Painter, one of the most hyped arms in baseball entering the year, carries a 7.06 ERA and has been optioned to Triple-A. Bubba Chandler (4.64 ERA), Justin Crawford (.645 OPS), and toolsy shortstop Konnor Griffin all drew low-single-digit reads despite the pedigree. Griffin was the lone wildcard Opus flagged, on the strength of 20 steals in 59 games, but even there the rest of the panel stayed near the floor.

None of these are edges to act on. They are honesty about how much a two-thirds-season award still leaves open.

The Case The Models Might Be Missing

Fairness to the market: a thin contract does not make a 51-cent favorite by accident. Traders backing Wetherholt may be weighing scouting reputation, a strong second-half projection, or middle-infield defensive value that a raw 2026 stat line does not capture, none of which the models were given. The panel scored the numbers on the page. The market may be scoring the player. If Wetherholt is holding the trophy in November, this is the paragraph that saw it coming. And McLean's own 29.5% leans on him keeping the ball through a rookie starter's first full-season workload, which is exactly the risk Kimi flagged.

How To Read This Board

The top of this market is actively traded, but most of the deep field barely is. Twenty-plus contracts sit near the 1-cent floor on light volume, which is common for a season-long award that will not resolve until the offseason. That has two practical consequences.

First, the penny prices are not a considered verdict on those players; they are placeholders, and the deep prospects still in the minors sit near zero on the models' board too. With roughly two months left, a candidate who has not logged a 2026 major league inning or at-bat faces a very steep climb, no matter the scouting grade. Second, when a contract is this thin, the bid-ask spread and low liquidity matter a great deal, so a quoted price can move sharply on a single trade, and fees can eat a real share of any small edge before it pays. The award itself is decided by a Baseball Writers' Association of America vote taken after the regular season and announced in November 2026, and this market is listed to settle by the end of 2026 once that result is official.

Read the price like a trader, not a fan. A YES quote is not a clean probability. Across a mutually exclusive field the listed YES prices sum to well over 100 cents, because every contract is padded with the exchange's margin. Here the 29 player contracts add up to about 120 cents of implied probability. To get closer to a fair read, compare each name's price to that sum and re-scale. The 51-cent favorite, JJ Wetherholt, is really closer to 42 percent once the overround comes out, and the models' 7.6 percent still sits far below even that. Nolan McLean's dime de-vigs to roughly 8 percent, against a 29.5 percent model blend. That gap is the whole story of this board.

That same habit is the point of reading a Kalshi market carefully: never take the first number without checking what a fair, vig-free version of it looks like. And remember that Kalshi is a CFTC-regulated event-contract exchange, not a sportsbook: you must be 18 or older and in an eligible state, and availability varies and can change.

Award markets built on a single player's season have close siblings on this board. If you want the other league's version of this exact question, our AL Rookie of the Year model verdict runs the same price-blind panel on Munetaka Murakami's candidacy. For the bigger NL awards, our NL MVP verdict and NL Cy Young verdict score those fields the same way.

Frequently Asked Questions

Who is the favorite for the 2026 NL Rookie of the Year? Kalshi's market makes Cardinals infielder JJ Wetherholt the favorite near 51 cents, with Reds slugger Sal Stewart next around 27 cents. Our five-model AI panel, run price-blind, disagrees: it splits the lead between Mets pitcher Nolan McLean (29.5%) and Stewart (28.5%) and gives Wetherholt only about 7.6%.

Why do the models fade JJ Wetherholt, the market favorite? The models score each rookie's actual 2026 line. Wetherholt is hitting about .251 with a .736 OPS, which the panel reads as a solid everyday player rather than the class of the rookie crop. The models never saw his price, so the gap between his 51-cent market and their 7.6% is a difference of method.

What is the Kalshi NL Rookie of the Year market and how does it settle? Kalshi lists 29 separate yes/no player contracts plus a Tie/Co-Winners contract under event KXMLBNLROTY-26. Each resolves YES only if that player wins the 2026 NL Rookie of the Year, which the Baseball Writers' Association of America votes on after the season and announces in November 2026.

Can I trade the Kalshi NL Rookie of the Year market? Kalshi is a CFTC-regulated event-contract exchange open to users who are 18 or older and located in an eligible state. Availability varies by state and changes over time, and the deep-field contracts are thin and lightly traded, so check Kalshi's current rules and live prices before trading. Nothing here is financial advice.

How were these model estimates produced? Five models scored the entire field without seeing any market prices, working from each candidate's fetched 2026 MLB stat line and standard Rookie of the Year dynamics. Their per-player numbers were blended into one estimate per player. These are model estimates, not predictions of fact and not financial advice.

The Bottom Line

The market and the models agree on one thing: Sal Stewart's power bat belongs near the top of the 2026 NL Rookie of the Year race. Everywhere else they diverge, and openly so. Five models, run price-blind over the whole field, read Nolan McLean's banked pitching line as the equal of Stewart's bat and fade the market's 51-cent favorite, JJ Wetherholt, down to a single-digit number. Those are model estimates, not predictions of fact and not financial advice. On a board where the crowd is paying for a name and the models are paying for a stat line, the more interesting question is not who is right today, but which one November proves out.

Panel: Claude Opus, Claude Sonnet, GLM, Kimi, and DeepSeek, run price-blind on July 30, 2026, from a deterministic candidate ledger fetched live from MLB StatsAPI. Our standard panel also includes Claude Fable, ChatGPT, and Gemini Pro; Fable was over its usage limit and the ChatGPT and Gemini Pro command-line seats were unavailable on this machine, so this article stands on five models and says so. Prices as of July 30, 2026.