The form book says New England; the model says Salt Lake
Real Salt Lake host the New England Revolution at Rio Tinto Stadium on Sept. 27, and the headline is a disagreement between the table and Nutmegly's AI read. New England sit second in the 2026 standings with 46 points from 26 games; Salt Lake are 13th with 29. The recent form is even more one-sided: Salt Lake's last five stored results are four defeats and a draw, while New England have won four of five. Yet the model leans home, giving Salt Lake a 39.9% win probability against New England's 35.1%, with the draw at 25.1%. That gap is the story.
The lean is not a majority. Salt Lake's 39.9% is the single largest outcome, but the draw and away win together account for 60.1% of the model's probability mass. This is a home lean built on venue and matchup, not a forecast that Salt Lake are the better team. The model itself flags that the three-way probabilities are not clearly separated and that the draw must be kept live. That framing matters: a leading outcome is not a guarantee, and the model is telling us the match is closer than the standings suggest.
The home split is Salt Lake's real case
Salt Lake's strongest evidence is their home record. In 13 home games they have seven wins, two draws and four defeats, scoring 24 goals and conceding 20. That is a meaningful home advantage, and it is the kind of split that can keep a team alive even when its overall record is poor. New England's road record is decent rather than dominant: five wins, two draws and six defeats in 13 away games, with 19 goals scored and 20 conceded. The home side has the stronger home split; the away side has the stronger overall record. This is not a mismatch in either direction.
The raw numbers also show why the model might prefer Salt Lake at Rio Tinto. Salt Lake have scored 24 times at home, while New England have conceded 20 on the road. But New England have scored 19 away goals themselves, and Salt Lake have conceded 20 at home. The home edge exists, but it is narrow enough that New England's road attack can erase it. The model's home lean should be read as a slight venue adjustment, not a verdict on quality.
New England's form is the strongest counterargument
The counterargument is obvious and loud. Over the last eight stored matches, Salt Lake have averaged 0.12 points per game; New England have averaged 2.00. In the MLS sample, the gap is even wider: 0.20 points per game for Salt Lake against 2.40 for New England. Salt Lake's last five results were a 0-2 loss at Seattle, a 0-3 home loss to Vancouver, a 0-2 home loss to New York City FC, a 1-2 loss at Houston and a 2-2 home draw with Los Angeles FC. They scored three goals and conceded 11 in that run. New England's last five: a 4-2 home win over Orlando City, 2-1 away wins at Chicago Fire and New York City FC, a 1-2 loss at LA Galaxy, and a 3-1 away win at Columbus Crew. They scored 12 and conceded seven.
The schedule also gives New England a longer gap since their last recorded match: seven days, against three for Salt Lake. That is a factual difference in the stored calendar, but it is not proof of fresher legs or a decisive rest advantage. A gap in records is not extra rest, and missing schedule entries do not mean a team has not played. The form and points-per-game numbers remain the cleanest evidence against the model's home lean.
A lean, not a lock
Nutmegly's AI read is plain: Salt Lake 39.9%, draw 25.1%, New England 35.1%. The model also puts over 2.5 goals at 51.0%, a near coin-flip with a slight lean toward a higher-scoring game. The caveats are as important as the numbers. The model's own counterpoints note that the three-way probabilities are not clearly separated, that historical xG and technical-stat samples are thin, that player-rating samples are thin, and that there is no head-to-head sample to add another check. The available advanced-stat and player-rating samples are also one-sided: they exist for Salt Lake but not for New England, which means any quality adjustment is uneven and should be treated lightly.
Lineups are not fully confirmed, and the available injury feed reports no absentees. That does not mean both squads are fully healthy, and it does not prove no official availability news has happened. It simply means the preview is working without confirmed lineups or injury detail. In a match this close, that missing information is a reason to hold the model's home lean loosely, not a reason to dismiss it.
The watchpoint: can Salt Lake's home split survive New England's road scoring?
The match will likely turn on which version of each team shows up. If Salt Lake's home split is the real signal, their 24 home goals and seven home wins give the model's 39.9% home lean a solid foundation. If New England's recent scoring form travels, their 12 goals in the last five and their 19 away goals this season give the away side a strong case. The goal total is the other live thread: over 2.5 sits at 51.0%, and recent matches support a busy scoreline. Salt Lake's last five produced 14 total goals; New England's last five produced 19.
The specific question to watch: does Rio Tinto remain Salt Lake's refuge, or does New England's road scoring make the model's home lean the most fragile number of the night? If Salt Lake score first, the home split looks justified. If New England's recent attacking output carries over, the 39.9% home probability will look more like a venue adjustment than a form call. That is the tension to track, not a prediction to bank.