The 79% away read that demands a second look
This Super League Round 29 fixture at Qingdao Guzhenkou University City Sports Center is a mid-table meeting with a one-point gap: Qingdao Youth Island are 7th on 38 points, Yunnan Yukun 6th on 39. That is not usually the stuff of a lopsided forecast. Yet Nutmegly's AI read leans heavily away, giving Yunnan a 79.0% win probability, the draw 16.1% and Qingdao just 5.0%. The model also puts over 2.5 goals at 44.6%, so it is not forecasting a wild shootout. The headline storyline is the clash between that away probability and the home/away splits: Qingdao have lost once in 14 home league games, while Yunnan have lost eight of 14 away.
Qingdao's home floor, and the crack in it
Qingdao's home body of work is the strongest counterargument to the model. In 14 home league games they have five wins, eight draws and one defeat, scoring 17 and conceding 13. That is a low-scoring, hard-to-beat profile, and it explains why the 5.0% home-win probability feels severe. The recent sample, however, is not clean: their last five stored results are a 0-1 away loss to Beijing Guoan, a 0-4 home loss to Dalian Zhixing, a 2-2 away draw at Wuhan Three Towns, a 1-1 away draw at Tianjin Teda and a 2-0 home win over Shandong Luneng. Two defeats and two draws in that run mean the home floor is real but not unbroken. The 0-4 Dalian result is the obvious crack. In the five displayed recent results, both sides have taken five points, which makes the table gap feel even narrower.
Yunnan's away high-wire act
Yunnan's season is defined by variance. They have 11 wins and 11 defeats in 28 league games, with 56 goals scored and 59 conceded. Away from home the split is three wins, three draws and eight defeats, with 24 goals scored and 39 conceded. The recent away sample is even more volatile: a 1-7 defeat at Shanghai Shenhua and a 0-6 defeat at Hangzhou Greentown sit alongside a 3-3 draw at Beijing Guoan. At home they have been a different team, including a 5-0 win over Shenyang Urban and a 1-1 draw with Chongqing Tongliang Long in their latest stored result. The AI's away lean is therefore not built on away solidity; it is built on Yunnan's scoring volume and their extra wins overall. That is a legitimate case, but it is a fragile one.
The case for the model, and the case against it
The case for the AI read starts with output. Yunnan have scored 56 goals to Qingdao's 31, and they have 11 wins to Qingdao's eight. Qingdao's 14 draws are the most striking number in their profile: a side that so often settles for a point can struggle to turn home resilience into three. If the model is weighting Yunnan's goal production and win count, a heavy away lean follows. The countercase is just as concrete. Qingdao have lost once at home all season, and Yunnan have conceded 39 in 14 away games. A 79.0% away win probability is a strong claim against that home record. The model's own caveats matter too: the read was generated on 25 September, so it predates the latest October results; the available data has no odds, no external forecast, no advanced statistics, no player ratings and no confirmed lineups or injury reports. This is a results-and-standings read, not a fully calibrated one.
The watchpoint: whether the game stays small
The match could unfold in two very different ways without either side needing a radical change. If Yunnan's away games continue to produce heavy scorelines, their scoring volume could overwhelm a Qingdao side that has scored only 17 at home in 14 games. If Qingdao keep the game compact and low-scoring, the away probability will look like an overreaction to Yunnan's goal totals and Qingdao's recent defeats. The specific watchpoint is the first half-hour: whether Qingdao's home defence holds and whether Yunnan's away vulnerability appears early. A goalless half-time would not guarantee a Qingdao result, but it would make the 79.0% away read feel far less comfortable than it does on paper.