Any Interest in this kind of Game 1 Media Summary?

[Statto]

Oh man, I really touched a nerve here. I originally created for my own edification, but the output was so ... comprehensive, that I thought others might be interested.

My goal was to compile a summary of the themes mentioned across all media, identify them, and state the variations on those themes. I wanted the sources to be bulleted for each theme in the report (you can see that in the template I provided) so I could click through and read them directly, but it didn't come out that way.

I omitted the game summary from the post.

I counted a grand total of 1 yes, 1 do it and I'll skip it if I don't want to read it, and the rest no. So ... I'll refrain.

Happy to be transparent about the background:
The tool I used is Google Deep Research with the Gemini 2.5 Pro Experimental model.
The prompt:

# Edmonton Oilers vs. Los Angeles Kings on April 21, 2025

You are an expert research assistant. Your task is to compile a comprehensive report on the online media and opinion coverage of the specified hockey game.



1. **Gather data** from a variety of online sources, including:

* News articles (local news sources, national news sources, nytimes.com/athletic , mayorsmanor.com, hockeyroyalty.com, nhl.com, www.dailynews.com/author/andrew-knoll/ , www.naturalstattrick.com/games.php , espn.com/nhl , prohockeyrumors.com, laxsportsnation.com/category/kings, hockeyfeed.com, sportsnet.ca/hockey, tsn.ca/nhl, thehockeywriters.com, thehockeynews.com, thefourthperiod.com)

* Blog posts

* Social media discussions (X, Facebook, YouTube, Reddit, forums.hfboards.com

* YouTube channels (@NHL, @TSN_Sports, @ESPNNHL, @lakings, @lakingsinsider4112, @laroyalty, @HockeyRoyalty, @TheHockeyGuy, @thefourthperiod, @thehockeypdocast, @LockedOnKingsNHL, @sdpn, @TopShelfHockey, @daily_faceoff, @legorocks99, @nextmanupNMU, @WhatChaosShow, @donnieanddhali)





2. **Identify and analyze** the different points of view expressed in the coverage.



3. **Aggregate** these viewpoints into distinct themes.



4. **Ensure** that the themes cover the following aspects of the game:

* Analysis of game strategies

* Player performances (positive impact, negative impact, lack of significance)

* Controversial calls by referees (if consensus indicates significance)

* Injuries sustained during the game and their impact (present and future)



5. **Present** the output as a structured report, adhering to the template provided above.



6. **Organize** the report into the following structure:



* **Game Overview:** \[Final score, game summary by period, and key statistics]

* **Game Strategies:** \[Analysis of strategies employed by both teams, including any tactical adjustments made during the game.]

* **Player Performances:**[Identify players with significant positive impacts (e.g., goals, key plays), negative impacts (e.g., errors, penalties), or lack of significance.]

* **Controversial Calls:** Discuss any significant controversial referee decisions, focusing on those widely considered impactful.

* **Injuries:** Detail any injuries sustained during the game, the players involved, and their potential impact.

* **Other Key Themes:**

* **Theme 1:** \[Name of Theme]

* Description: \[Detailed description of the theme]

* Sources: \[List of sources supporting this theme, ranked by credibility, then prominence, then uniqueness of viewpoint]

* Variations: \[Subsections detailing variations within the theme, with supporting sources]

* **Theme 2:** \[Name of Theme]

* Description: \[Detailed description of the theme]

* Sources: \[List of sources supporting this theme, ranked by credibility, then prominence, then uniqueness of viewpoint]

* Variations: \[Subsections detailing variations within the theme, with supporting sources]

* \[Continue with additional themes as needed]

* **Unique Points of View:**

* **Viewpoint 1:** \[Description of the unique perspective]

* Source: \[Source expressing this unique viewpoint]

* Significance: \[Explanation of why this viewpoint is noteworthy]

* **Viewpoint 2:** \[Description of the unique perspective]

* Source: \[Source expressing this unique viewpoint]

* Significance: \[Explanation of why this viewpoint is noteworthy]

* \[Continue with additional unique viewpoints as needed]

Hey for your own interest use, skillset development etc crack on; don’t let grumpy arses like me put you off. My points are much wider ones about AI and its direction.

If you share though, I personally think crediting sources would be good.

Also, are you real or an AI not? If so I take it all back

 

[bland]

Now I'm thinking I should have picked the name AI-Powered Kings Fan. Yes, I'll see if I can go change that now.

EDIT:
Nope, not for over a month. I guess... I'm treated like a bot for now?

How many pictures of stop signs are in this post?

 

[Ziggy Stardust]

I’d like a statistical breakdown of how many Oilers fans whine about the officiating in every game.

 

[AI Kings Fan]

I’d like a statistical breakdown of how many Oilers fans whine about the officiating in every game.

Whoa... hold on now. The power use might break the grid!

 

[chris kontos]

Oh man, I really touched a nerve here. I originally created for my own edification, but the output was so ... comprehensive, that I thought others might be interested.

My goal was to compile a summary of the themes mentioned across all media, identify them, and state the variations on those themes. I wanted the sources to be bulleted for each theme in the report (you can see that in the template I provided) so I could click through and read them directly, but it didn't come out that way.

I omitted the game summary from the post.

I counted a grand total of 1 yes, 1 do it and I'll skip it if I don't want to read it [Edit: Opps, that was the same person], and the rest no. So ... I'll refrain.

Happy to be transparent about the background:
The tool I used is Google Deep Research with the Gemini 2.5 Pro Experimental model.
The prompt:

# Edmonton Oilers vs. Los Angeles Kings on April 21, 2025

You are an expert research assistant. Your task is to compile a comprehensive report on the online media and opinion coverage of the specified hockey game.



1. **Gather data** from a variety of online sources, including:

* News articles (local news sources, national news sources, nytimes.com/athletic , mayorsmanor.com, hockeyroyalty.com, nhl.com, www.dailynews.com/author/andrew-knoll/ , www.naturalstattrick.com/games.php , espn.com/nhl , prohockeyrumors.com, laxsportsnation.com/category/kings, hockeyfeed.com, sportsnet.ca/hockey, tsn.ca/nhl, thehockeywriters.com, thehockeynews.com, thefourthperiod.com)

* Blog posts

* Social media discussions (X, Facebook, YouTube, Reddit, forums.hfboards.com

* YouTube channels (@NHL, @TSN_Sports, @ESPNNHL, @lakings, @lakingsinsider4112, @laroyalty, @HockeyRoyalty, @TheHockeyGuy, @thefourthperiod, @thehockeypdocast, @LockedOnKingsNHL, @sdpn, @TopShelfHockey, @daily_faceoff, @legorocks99, @nextmanupNMU, @WhatChaosShow, @donnieanddhali)





2. **Identify and analyze** the different points of view expressed in the coverage.



3. **Aggregate** these viewpoints into distinct themes.



4. **Ensure** that the themes cover the following aspects of the game:

* Analysis of game strategies

* Player performances (positive impact, negative impact, lack of significance)

* Controversial calls by referees (if consensus indicates significance)

* Injuries sustained during the game and their impact (present and future)



5. **Present** the output as a structured report, adhering to the template provided above.



6. **Organize** the report into the following structure:



* **Game Overview:** \[Final score, game summary by period, and key statistics]

* **Game Strategies:** \[Analysis of strategies employed by both teams, including any tactical adjustments made during the game.]

* **Player Performances:**[Identify players with significant positive impacts (e.g., goals, key plays), negative impacts (e.g., errors, penalties), or lack of significance.]

* **Controversial Calls:** Discuss any significant controversial referee decisions, focusing on those widely considered impactful.

* **Injuries:** Detail any injuries sustained during the game, the players involved, and their potential impact.

* **Other Key Themes:**

* **Theme 1:** \[Name of Theme]

* Description: \[Detailed description of the theme]

* Sources: \[List of sources supporting this theme, ranked by credibility, then prominence, then uniqueness of viewpoint]

* Variations: \[Subsections detailing variations within the theme, with supporting sources]

* **Theme 2:** \[Name of Theme]

* Description: \[Detailed description of the theme]

* Sources: \[List of sources supporting this theme, ranked by credibility, then prominence, then uniqueness of viewpoint]

* Variations: \[Subsections detailing variations within the theme, with supporting sources]

* \[Continue with additional themes as needed]

* **Unique Points of View:**

* **Viewpoint 1:** \[Description of the unique perspective]

* Source: \[Source expressing this unique viewpoint]

* Significance: \[Explanation of why this viewpoint is noteworthy]

* **Viewpoint 2:** \[Description of the unique perspective]

* Source: \[Source expressing this unique viewpoint]

* Significance: \[Explanation of why this viewpoint is noteworthy]

* \[Continue with additional unique viewpoints as needed]

Sure didnt mean to insult or piss anyone off.
Just having big issues with sillicon valley and cyber generally. Ubu

 

[AI Kings Fan]

If you share though, I personally think crediting sources would be good.

Yeah, I put the sources in the first reply, but it's in moderation. Probably because of the number of links (48).
EDIT: Just added them to the end of the original post.

Also, are you real or an AI not? If so I take it all back

*facepalm*

 

[King'sPawn]

I posted the tooling and prompt earlier, so hopefully that question was answered. No special training, just a prompt.



The Google Deep Research tool was pointed to what data to seek out.
Yes to articles other writers wrote.
Yes to based on Reddit and HFBoard. Well, HFBoard was a suggested source, but it doesn't look like it was used.



True. That's what I instructed it to do.



Fair enough. That was exactly what I asked it to do. Maybe it would have been helpful for me to include the goal, tooling, and prompt in the initial post.



Other than the plagiarizing, that's exactly what it's doing. It's aggregating the writing, categorizing, summarizing, and reporting. Huh. I just noticed that my reply with the list of sources used is stuck in moderation. One of my main goals was to have a list of article links I could click-through to for a given opinion, right in the report. That's not what it did.

The original report generated a game summary, which I didn't include. I figured there's great versions of that done by others.



I agree with that as a general statement. Hopefully the context I've added later (my goal, the tools, the prompt) helps.

I appreciate your transparency in your process. I truly do.

I work in IT and have some basic certifications in AI. I only bring this up for transparency in my perspective. I'm not anti-AI. But I believe it's important to have standards and ethics in its application, and while I know it's well-intentioned, there are concerns.

We're still in the infancy of AI. There are numerous valuable uses for AI as a tool. However, it's best when trained on specific data models, of which I don't think social media sentiment is a valuable or effective method. At least not in a way where it appeals to a broad group (of course, if you find it helpful for yourself, go for it).

I also write for MayorsManor. Jesse (All the Kings Men) provides content outside of HF. We also have people who write for Hockey Royalty. So, it stings for all of us when we put in hours of work for articles just for AI to take some of it and repurpose it without getting cited. Even if it's none of us, we still won't quite know the quality of the source.

But that's just my perspective. There are certainly ways AI could be trained to analyze a game. But I, personally, don't benefit from summarizing social media sentiments. But I do appreciate your passion for trying to apply AI to learn more about the game.

 

[Brownie to Pancakes]

AI is dogshit for losers and I never want to see it in any setting. Hope this helps

 

[Faterson]

That’s the issue though because most people tend to trust the internet as a source and will certainly trust AI, history tells us that.

Exactly. So this alleged threat by "AI" (I always call them "the robots", a word originating from my native country, not "AI") is nothing new. The threat from the robots is exactly the same as from (a)social networks or the Internet itself. Dumb folks will blindly trust the Internet, (a)social networks, and the robots – no difference there. Immense damage has already been wreaked worldwide by folks blindly trusting their (a)social networks – even folks who have never heard of or used "AI". It can't really get any worse than it already is, in terms of folks blindly trusting the Internet – "AI" is just a new variation of the same old thing. The countermeasure is to improve the users' education worldwide: instill healthy skepticism and a critical stance as the default attitude towards IT tools in everyone, not to depict any of these particular tools as scarecrows.

And yes, @King'sPawn, I know that the summaries may include my, your, or anyone else's writing. That's not a bug or a flaw, though – it's the standout feature and exactly what it was asked to do. It's not "plagiarism", either: if you look more carefully at the opening post, there are footnotes interspersed within the summary (@AI Kings Fan just didn't copy the footnotes over to the opening post), so the automatic summary very much cites the sources of the statements it includes in the summary, which is standard academic practice.

 

[AI Kings Fan]

I appreciate your transparency in your process. I truly do.

[...]

I also write for MayorsManor. Jesse (All the Kings Men) provides content outside of HF. We also have people who write for Hockey Royalty. So, it stings for all of us when we put in hours of work for articles just for AI to take some of it and repurpose it without getting cited. Even if it's none of us, we still won't quite know the quality of the source.

I appreciate that feedback. I had put all the sources into the first reply, which is in moderation. I just edited the original post to add them all in at the end. That seems to bury it a bit, but it's better than not being there at all.

 

[King'sPawn]

I appreciate that feedback. I had put all the sources into the first reply, which is in moderation. I just edited the original post to add them all in at the end. That seems to bury it a bit, but it's better than not being there at all.

I saw that after I posted. I appreciate that.

 

[AI Kings Fan]

I saw that after I posted. I appreciate that.

No, you pointed that out, and I edited the post afterwards.

I'd realized the moderation thing and was a bit concerned that editing the original post would trip the same safety measures. But it seemed to be the right thing to do.

 

[AI Kings Fan]

Just curious, everyone...

What about a similar research report in a trade thread when someone brings up a trade target? For example, at the trade deadline, I read a main board thread about Tarasenko being available. At the time, I just idly wondered how his career was going recently. Just now I built a prompt to give me a player's draft profile assessment, how he progressed through D+5 and in the current year.

Or during the draft season, having an amalgamation of the scouting reports, with consensus and varying opinions.

I'm trying to get the sources in-line so I can just click through and read the originals. Again, might just be interesting just for me. Or maybe others are as well.

 

[Maynard]

f*** off skynet!

 

[King'sPawn]

Just curious, everyone...

What about a similar research report in a trade thread when someone brings up a trade target? For example, at the trade deadline, I read a main board thread about Tarasenko being available. At the time, I just idly wondered how his career was going recently. Just now I built a prompt to give me a player's draft profile assessment, how he progressed through D+5 and in the current year.

Or during the draft season, having an amalgamation of the scouting reports, with consensus and varying opinions.

I'm trying to get the sources in-line so I can just click through and read the originals. Again, might just be interesting just for me. Or maybe others are as well.

If it's social media sentiment, I'm not interested. If it's compiling public data, it would be interesting to see how it's compiled.

Not sure about anyone else though.

 

[Statto]

Just curious, everyone...

What about a similar research report in a trade thread when someone brings up a trade target? For example, at the trade deadline, I read a main board thread about Tarasenko being available. At the time, I just idly wondered how his career was going recently. Just now I built a prompt to give me a player's draft profile assessment, how he progressed through D+5 and in the current year.

Or during the draft season, having an amalgamation of the scouting reports, with consensus and varying opinions.

I'm trying to get the sources in-line so I can just click through and read the originals. Again, might just be interesting just for me. Or maybe others are as well.

Scouting reports might be interesting (to me). Perhaps broken into 2 parts, pro-scouts and internet bloggers. It would certainly reduce my Google searches!!

 

[AI Kings Fan]

If it's social media sentiment, I'm not interested. If it's compiling public data, it would be interesting to see how it's compiled.

Oh, I think I understand what you meant a bit earlier, now. I don't think there's much value in being fed back social media sentiment. My main targets were pro media, podcasts, YouTube channels, and bloggers. How about blocking X/Twitter, Reddit, and discussion boards?

 

[Bandit]

Is anyone interested? No.

Is anyone interested in complaining? HELL YES.

 

[King'sPawn]

Oh, I think I understand what you meant a bit earlier, now. I don't think there's much value in being fed back social media sentiment. My main targets were pro media, podcasts, YouTube channels, and bloggers. How about blocking X/Twitter, Reddit, and discussion boards?

I'm wary to say yes because I'm sure there are other sources in which I have no idea of the quality of the content.

I'm with Statto on scouting reports.

As I focus more on prospects, drafting, development, etc, I would love to try to train AI on watching tbe actual games. A combination of raw data (like what the NHL's EDGE provides) but also a qualitative analysis (e.g. how does he handle a 2-on-1? How is his shot placement? Etc) would be very cool.

Obviously I know that's not available with Google's tools, but if a team can train AI to watch players tbrough the eyes of their head scouts, they could cover a lot more hockey and looking at players to zero in on.

I'm just not sure how much I would enjoy general sentiment for a trade, even with certain sites blocked. Sorry, I promise I'm not trying to be a pain. I understand you're trying to find a compromise that you can share with us to enjoy. I just may not be the best one to offer suggestions for what you're trying to achieve.

 

[Statto]

If AI could be trained to collect stats that’d be huge as it would create consistent definitions of hits etc. instead of a subjective opinion.

 

[bland]

Oh, I think I understand what you meant a bit earlier, now. I don't think there's much value in being fed back social media sentiment. My main targets were pro media, podcasts, YouTube channels, and bloggers. How about blocking X/Twitter, Reddit, and discussion boards?

I would like to ask of your intentions.

Are you trying to formulate a singular opinion based on collected data? If so, why? Why does this need to be presented?

If it's just because you find it interesting, by all means carry on. But if it's an effort to try and find a "truth" with weight behind it, I don't think this method supports that kind of finding.

 

[AI Kings Fan]

If AI could be trained to collect stats that’d be huge as it would create consistent definitions of hits etc. instead of a subjective opinion.

You might need accelerometers or impact sensors built into the uniforms.

 

[AI Kings Fan]

I would like to ask of your intentions.

Are you trying to formulate a singular opinion based on collected data? If so, why? Why does this need to be presented?

If it's just because you find it interesting, by all means carry on. But if it's an effort to try and find a "truth" with weight behind it, I don't think this method supports that kind of finding.

Well, I don't think it's possible to have a single opinion. There's too many people. I mean, even a single person can have more than one, conflicting opinion at a time.

My goal was to create a summary of the themes of the game from the various media sources out there as a landing page for me to click through and read the articles. I wanted the links to the various sources at the bottom of each section so I could click-through, but that didn't happen. I'll have to tweak the prompt to ensure that it does. But even without that, I got a ton out of the overview, especially the bit about perception of momentum. I still need to comb through the sources to see where that came from.

What I got was a full-blown period-by-period breakdown of the action (which I didn't include), and what I thought was a really thorough analysis of the various factors of the game.

As an aside, my LA Kings bookmark list includes over 20 different sources, not including podcasts. I assume when we play another team, that list would at least double, and I don't want to compile that sources list for each round of the playoffs we get to, much less during the regular season. We're getting more coverage from the national media than we normally do during the regular season, so even the standard coverage links have way more information than normal. I don't think I mentioned the large number of duplicate wire-service reports from different newspapers that show up with Google Search or even Google News.

Does that answer the question?

 

[AI Kings Fan]

I'm just not sure how much I would enjoy general sentiment for a trade, even with certain sites blocked.

I hadn't even thought about trade analysis, I was thinking about trade targets that I didn't follow the careers of.

Last trade deadline, I think the Hockey Royalty podcast brought up the idea of trading for Martin Nechas. I don't watch enough Carolina games or read their hockey media enough to know why, other than what they described. At the time, I would look up pre-draft reports to get a feel for the projections as a prospect, but I didn't have the time to look for D+1, D+2, etc writeups. Contemporary articles kind of assume one has most of the knowledge of the player and sprinkle in some info, unless it's specifically about the person as a trade target. I wanted to shorten the time it would take for me to read up on a player.

I think the only reason I read as much PLD trade reaction content as I did was that I didn't know much about him.

Anyway, analyzing general trade sentiment wouldn't be that interesting. Understanding a player a bit more by reading the analysis that's out there about their skills, then coming to my own conclusion is what I think would be more interesting. Anyway, I DM'd you and Statto the Tarasenko and Matthew Shaefer output to read and respond to.

 

[All The Kings Men]

Y'all never watched the twilight zone or black mirror or read any ray bradbury?

Don't make the robots stronger.