内容来源：https://news.microsoft.com/source/emea/features/fantasy-premier-league-companion-gives-managers-a-new-tool-for-success/

内容总结：

英超官方发布AI助手：奇幻英超助手助力新赛季“经理人”

随着西班牙队捧起今夏欧洲杯冠军，全球球迷的目光已转向即将于8月21日开幕的英超新赛季。与此同时，超过1300万玩家参与的官方奇幻英超（FPL）游戏也迎来新变化——一款名为“奇幻英超助手”的AI工具正式上线，为玩家提供前所未有的“智囊”支持。

据悉，这款由微软Copilot驱动、基于Azure OpenAI和ChatGPT 5.4技术打造的助手，不仅整合了英超官方比赛数据（如进球、助攻、出场时间、射门次数等），还融合了FPL游戏特有的玩家操作数据（如转会次数、队长选择、功能卡使用等）。玩家可通过自然语言对话，随时获取球员推荐、阵容建议和规则答疑。

并非“自动托管”，而是“智能参谋”

英超联盟高级游戏经理安迪·塞尔比在专访中强调，助手不会替玩家做决定或预测积分：“足球充满不确定性，没有唯一正确答案。助手会提供多种选项和支撑数据，比如当你的球员受伤时，它会推荐科尔·帕尔默、萨卡或塞门约等人选，并附上历史表现和赛程分析，但最终决策权仍在玩家手中。”

降低门槛：新手也能轻松上手

对于初次接触FPL的玩家，助手通过对话界面消除“硬核恐惧”。新手可提问“如何选阵容？”“功能卡怎么用？”“怎么和朋友创建联赛？”；随着玩家经验增长，问题将进阶为“哪位前锋射门最多？”“哪位门将未来赛程最有利？”——助手会根据用户水平动态调整回复深度。

增强连接：从“队迷”到“人迷”

塞尔比指出，FPL正推动球迷从“支持单一俱乐部”转向“追随特定球员”。例如，上赛季选了诺丁汉森林中场吉布斯-怀特的玩家，即使并非该队球迷，也会因球员表现而关注球队未来。助手每周更新实时信息，同一问题在周二和周四可能因新赛报获得不同答案。

未来展望：用户反馈驱动迭代

英超联盟表示，新赛季将密切观察用户提问类型、高频使用场景及追问模式，并据此持续优化助手的应答质量和信息呈现。“它会随着赛季自然进化，我们期待看到玩家提出更深入、更聪明的问题。”塞尔比总结道。

中文翻译：

英超幻想助手为经理人提供全新成功工具

今夏精彩赛事以西班牙夺冠圆满收官后，全球球迷正为8月21日开幕的英超新赛季做准备，其中包括参与英超官方幻想游戏"英超幻想联赛"（FPL）的1300多万玩家。在游戏中，经理人拥有1亿英镑预算，需从全联赛挑选15名球员组成阵容，力求在整个赛季中赢得最高积分。

本赛季，经理人将获得另一项得力工具——并非仅仅是选择曼城球星埃尔林·哈兰德（上赛季FPL最佳射手）的机会。由Copilot驱动的英超幻想助手基于微软Foundry构建，利用Azure OpenAI、Chat GPT 5.4及英超官方数据，为经理人在组建和调整阵容时提供建议、解答疑问并给出推荐。

为深入了解这一新工具及其功能，英超高级游戏经理安迪·塞尔比在英超幻想助手上线前接受了专访。

问：FPL助手能直接代管球队吗？

塞尔比：它绝对不是为了实现球队自动托管。它不会预测FPL积分。对我们而言，让FPL经理人自主决策至关重要。你是自己球队的经理人，最终决定权在你手中。FPL助手不会替你决策，而是帮助你做出更明智、更有依据的选择。

因为足球本身充满变数，从来不存在唯一正确答案。比如我队中球员受伤，助手可能会说：你可以考虑（切尔西的）科尔·帕尔默、（阿森纳的）布卡约·萨卡或（曼城的）安托万·塞梅尼奥。它会提供多个选项，附带相关数据支撑，并链接到我们处理过类似问题的相关专题。但最终，是由你来做出选择。

问：FPL助手如何帮助新手经理人站稳脚跟？

塞尔比：我认为任何体育项目的幻想游戏都容易让人望而生畏，常被视为硬核球迷或深度了解计分规则的人群专属。但关键在于，它并没有那么可怕。

你可以通过对话界面进入系统，提出任何问题——没有愚蠢的问题——都会得到答案。这些答案来自可信的数据源，即我们的比赛数据库和FPL数据库。

新用户使用平台的前几周价值巨大。在此之前他们可能会感到困惑或卡壳。现在，随着用户对FPL的了解和技能提升，助手也会同步进化。起初他们可能会问："如何挑选球队？""解释一下这些功能芯片怎么用？""怎么和朋友创建联赛？"我期待看到他们在赛季深入后提出更有深度的问题："哪个前锋射门次数最多？""哪位门将未来赛程最有利？"

问：英超丰富的数据库如何助力助手？

塞尔比：数据分为两类。第一类是比赛数据：进球、助攻、出场、红黄牌、上场时间、抢断次数、射门次数等——即比赛中正在发生或历史发生过的情况。

这些信息本就可以获取，但需要知道去哪里找、如何浏览表格。通过助手直接获取这些信息，堪称快速查找和消化数据的革命性突破。

第二类是游戏数据——即FPL本身产生的数据。这包括某位球员的转会次数、选择某球员担任队长的经理人数、使用特定功能芯片的经理人数等。

这些数据只能通过英超幻想联赛获取。我们将两套数据结合，生成推荐建议。当用户问"我的球员受伤了，该换谁？"时，助手能整合这两组数据，告诉你过去六轮比赛的表现、哪些球员更受欢迎、哪些球员赛程最有利。

问：您认为FPL助手如何帮助球迷在比赛日之外建立更深层联结？

塞尔比：过去球迷通常支持特定俱乐部，但我们越来越多地看到球迷关注特定运动员，而FPL正是实现这种关注的独特方式。比如有人说："我上赛季选了摩根·吉布斯-怀特，我不是诺丁汉森林球迷，但他表现太棒了，所以我现在开始关注森林队，关注吉布斯-怀特，无论他转会到哪里。"

借助助手，球迷可以整周持续对话，获取实时更新的信息。如果我周二提问，周四再问同样的问题，可能会得到略微不同的答案——因为期间有新闻更新，所有这些都会实时输入助手。

问：您认为这个工具未来会如何发展？

塞尔比：我们会倾听用户声音。新赛季前我们会将工具推向市场，很快就能看到用户提问的类型、他们最常使用哪些功能、以及会提出哪些后续问题。

整个赛季中，它在回答问题的方式和信息呈现上都会自然进化。我们将观察用户如何互动、哪些功能获得积极反馈，并在此基础上持续迭代。

英文来源：

Fantasy Premier League Companion gives managers a new tool for success
Now that this summer’s thrilling tournament was capped off by a Spain victory, football fans across the globe are gearing up for the start of the Premier League season, which kicks off August 21. That includes the more than 13 million players who participate in Fantasy Premier League (FPL), the official fantasy contest for the league. In the game, managers have a £100 million budget to pick a squad of 15 players from across the league to try and score the most points throughout the campaign.
This season, managers are going to get another tool at their disposal, and not just the opportunity to choose Manchester City star Erling Haaland, who was the FPL’s top scorer last season. The Fantasy Premier League Companion powered by Copilot is built using Microsoft Foundry. It utilizes Azure OpenAI, Chat GPT 5.4 and official data from the Premier League to offer advice, answer questions and provide suggestions to managers as they build and maintain their squads.
To learn more about the new tool and what managers can expect, Andy Selby, senior gaming manager at the Premier League, sat down for a Q&A about the Fantasy Premier League Companion as it prepares to launch.
Can FPL Companion just run someone’s team then?
Selby: It’s definitely not there to be an autopilot for your team. It will not predict FPL points. It is very important to us that FPL managers themselves are making their own decisions. You’re the manager of your own team; it’s your call at the end of the day. The role of FPL Companion is not to do that instead of you. It’s to help you make better, informed decisions.
Because there’s never one single right answer. Football by its very nature is very unpredictable. So, if I had someone on my team and he’s injured, Companion might say, okay, you could look at [Chelsea’s] Cole Palmer, or [Arsenal’s] Bukayo Saka or [Manchester City’s] Antoine Semenyo. It will always give you a number of options. It will give you the relevant data to back it up. It will link to a relevant editorial where we’ve tackled similar questions, but you then as the manager are going to go and make that decision.
How will FPL Companion help new fantasy managers gain a foothold?
Selby: I think fantasy sports broadly across any sport can be seen as quite daunting. It can be seen as something for people who are really hardcore fans or have a deep understanding of scoring and things like that. And the key point of this is it’s not that scary.
You can come in, you’ve got a conversational interface, you can ask questions – no question’s a dumb question – and you’re going to get an answer to it. You’re going to get an answer from the trusted data source that is our database of match data and FPL data.
A new user’s first weeks on the platform is where I see lots of value. Before they may get stuck or confused. Now, as the user’s knowledge of FPL and skill set grows, Companion moves with it. So at first, they might be asking, “How do I pick my team?”; “Explain how these chips work?”; ”How do I set up a league with friends?” What I’m looking forward to seeing is them getting further into the season and asking more informed questions: “Which striker takes the most shots?” “Which goalkeeper has the best fixtures coming up?”
How did the Premier League’s wealth of data help inform the Companion?
Selby: There are two buckets. There’s match data. That is, goals, assists, appearances, cards, minutes on the pitch, number of tackles, number of shots, which is, what is happening in the matches, what has happened in the matches historically, etc.
That is information which you can access already, but you need to know where to look for it. You need to know how to navigate the tables. That being accessible through Companion is a bit of a game changer to quickly finding and digesting that information.
Then there’s game data, the game being FPL in this instance. That is data which only exists because FPL exists as a game. And so that could be the number of transfers that a player has had, the number of managers that have captained a certain player, the number of managers that have played certain chips.
That stuff you can only get through Fantasy Premier League. You are taking those two data sets and pulling that together to give recommendations. When I’m asking, “My player is injured, who should I bring in as a replacement?” Companion is able to bring together those two data sets and say, here is the performance of the last six game weeks, which players are popular and which players may have the best fixtures.
How do you see FPL Companion helping fans gain a deeper connection beyond match day?
Selby: Historically, we’re used to fans supporting specific clubs, but what we’re seeing more and more is fans following specific athletes and FPL is a unique way to do that. They might say, “I had Morgan Gibbs-White on my team last season. I’m not a Nottingham Forest fan, but he had a great season. So, I’m now interested in Forest. I’m now interested in Gibbs-White, wherever he goes.”
With Companion, fans can have a conversation across the week and there’s new information and that will be updated. So, if I’m asking a question on Tuesday and then I ask the same question on Thursday, I might get a slightly different answer because there’s been news that’s come in and all of that is fed into Companion in real time.
Where do you see this tool going in the future?
Selby: We will listen to users. We’ll be putting this out into the wild ahead of the new season and very quickly we’ll be able to see the sorts of questions that users are asking. Where are they spending the most time with this? What are the follow-up questions that are being asked?
There will be natural evolution across the season in terms of how it’s able to answer questions and the information it surfaces. We’ll see how users interact with it, what they respond positively to, and iterate from there.