Chat GPT on stats

ESPN gets an approval from Chat. And given the only other source that seems accessible is Opta, they are indispensable. I often track the stats while the game is on, and the stats that are reported on the screen tend to be the same as the stats ESPN are reporting, with some delay.

ESPN is a $16 billion revenue organization with $3 billion a year in profit. Their financial resources dwarf all others, they have no reason to put out a shoddy product:

Rugby union statistics are gathered through a combination of real-time data collection during matches and post-match analysis. Here’s an overview of how these stats are typically gathered and where you can find the best sources:

1. Data Collection Methods:

• Match Officials and Referees: During the match, referees and touch judges keep track of critical events like tries, conversions, penalties, and infringements. These are the most basic statistics collected in real-time.
• Official Statisticians: Professional statisticians are often employed by leagues, teams, or broadcasters to track a wide range of in-game data, including possession, territory, rucks, mauls, lineouts, scrums, tackles, and passes. They use manual recording methods as well as software designed for live data entry.
• Automated Systems: Advanced camera systems and wearable GPS technology are sometimes used to track players' movements, distances covered, and other physical metrics. These systems can provide detailed data on player performance and match dynamics.
• Post-Match Video Analysis: After the match, video footage is often reviewed to ensure accuracy and to gather more detailed statistics that might have been missed during live tracking. This includes things like line breaks, offloads, and defensive actions.

2. Types of Rugby Union Statistics:

• Team Statistics: Points scored, tries, conversions, penalties, possession percentage, territory percentage, scrum success, lineout success, turnovers won, rucks won, etc.
• Player Statistics: Tries scored, assists, meters gained, tackles made, tackles missed, line breaks, offloads, passes completed, kicks, handling errors, etc.

3. Best Sources for Rugby Union Statistics:

• Official Governing Bodies:
• World Rugby (worldrugby.org): The global governing body for rugby union provides comprehensive statistics for international matches, including the Rugby World Cup and other tournaments.
• Six Nations (sixnationsrugby.com) and The Rugby Championship (sanzaar.com): These bodies offer detailed statistics for their respective competitions.
• Professional Leagues:
• Premiership Rugby (premiershiprugby.com): The top tier of English rugby provides in-depth statistics for all matches in the league.
• Top 14 (lnr.fr): The French rugby league has detailed match statistics for the Top 14.
• Super Rugby (super.rugby): The premier club competition in the Southern Hemisphere provides extensive stats on its website.
• Broadcasters and Media Outlets:
• ESPN Scrum (espn.com/rugby): Offers a wide range of rugby statistics, particularly for international matches and major tournaments.
• BBC Sport (bbc.com/sport/rugby-union): Provides stats and match reports, especially for UK-based competitions.
• Specialized Sports Data Companies:
• Opta: A leader in sports data, Opta provides detailed rugby union stats for media and teams, including in-depth player metrics.
• Stats Perform: Another major provider of rugby statistics, used by teams, broadcasters, and analysts.
• Rugby Clubs and Teams:
• Many professional rugby clubs publish detailed stats about their matches and players on their official websites.

These sources offer reliable and up-to-date rugby union statistics, catering to fans, analysts, and professionals alike.

It’s a complex game to track, but the stats are right most of the time….far more often than arm chair supporters’ intuition.

Stats may be right -but it does not measure te qualty of performances and resulting circumstances.  

Stats may be right -but it does not measure te qualty of performances and resulting circumstances.  

I’ll rephrase….the best way to analyze the game is to watch it and reach your own conclusions. Then check the stats and if they differ from your impression, go back and watch the game with those differences in mind. But for the small number of times the numbers are wrong I’m  happy reach conclusions based on the stats. I’ll be right more times than if I didn’t access the stats at all.

Mozart

CatGP is a reflection what is psted in the Computer by cotributors.  The above is justa reflection what ESPN provided to the platform  to justify the usage of their stats, 

The issues I raised about evaluation of player performances used by World rugby reflects the value of stat info also and most probably all resulted activities  as wella s impact on what folowed on a scientific basis,   

Rad he follwing as to player performances-\

What is Performance Analysis? Performance Analysis is a specialised discipline that provides athletes and coaches with objective information that helps them understand performance. This process is underpinned by systematic observation, which provides valid, reliable and detailed information relating to performance.

"What is Performance Analysis? Performance Analysis is a specialised discipline that provides athletes and coaches with objective information that helps them understand performance. This process is underpinned by systematic observation, which provides valid, reliable and detailed information relating to performance."

Stats in itself is useful to determne deficiencie that needs to be corrected by coaches - but what is related to stats only is not a relable basis for evaluation without conndifeing also the impact follows on related statistical info.  It s in thatc ntext that stats alone is not a sound absis to be used on player performance evaluation.   

Depends on how processed you want the data. You can have defenders beaten, but that stat is more informative if it’s defenders beaten per run. Then if you add number of runs and average distance….you have yet more information. Then dial in the quality of opposition and you have a pretty reliable stat.

But actually that’s what any analyst should do. Sure you can build a few computer models,  automate the input and  quick outputs are obtained. But these relationships aren’t that complex, the main benefit of modeling all this is a consistent methodology. Most of the insights should be apparent in the first stage of data manipulation….defenders beaten per run, balls lost per run, tackles made, tackles missed, metres per run etc.

My problem with ESPN is the fact that they missed an obvious missed tackle by Mostert in that goal line defence in the 2019 WC final

They aren't infallible. They also had Esterhuizen at only 8m run in the Wales test, while just his one run down the touchline alone was probably about 15m. But it's pretty much the only public stats we have access to, and I think they're mostly accurate, or accurate enough so it doesn't make a huge difference to anything.

Just show the highlight real for every score or punishment. 

What you miss in stats if a player is caught out of position, player slips etc or a funny bounce of the ball that causes the player to over run his tackle. There is more to this game than just numbers. 

That is why videos are still the best. 

Really love how the two lions twin props JP Smith lined up in the backline to run a straight lines, draw, run straight, put brother into the gap to break the line and then score a try by the two slowest players on the field. How do you show stats for that?

Okay here’s a test Dave and Corn. How many tackles did Mapimpi   make or miss last weekend.  Give us your best estimate. I’m not asking Clever because for him last weekend was in the 20th century.

While ESPN is a well-established organization with significant financial resources and a long history in sports broadcasting, relying solely on its statistics and data may not always provide the most comprehensive or accurate picture, especially in a complex and dynamic sport like rugby union. Here are several points to consider as a counter-argument:

1. Diversity of Data Sources: Financial resources and size do not necessarily equate to accuracy or depth in data reporting. While ESPN can gather and present data, the depth and specificity of that data can vary. Specialized sports data companies like Opta, Stats Perform, or governing bodies like World Rugby, often have a more focused approach to data collection and analysis, providing detailed insights that might not be prioritized by a broad-spectrum sports network like ESPN. This specialization can lead to more nuanced and accurate statistics, especially for niche sports or specific metrics within rugby union.

2. Potential for Bias or Generalization: As a global sports network, ESPN caters to a broad audience, which can sometimes lead to a focus on more general or popular metrics, potentially overlooking more detailed or specific data points that are crucial for serious fans or analysts. In contrast, organizations like Opta or Stats Perform are specifically designed to serve the needs of teams, analysts, and a more data-centric audience, which can result in more precise and detailed reporting.

3. Real-time Accuracy vs. Post-Match Analysis: While ESPN provides real-time statistics during matches, the immediacy of these updates can sometimes come at the expense of accuracy. In contrast, post-match analysis performed by specialized data companies or governing bodies often reveals discrepancies and corrects errors that might have occurred during the live broadcast. Therefore, relying on a single source like ESPN may not always reflect the most accurate or up-to-date information, especially when immediate reporting is prone to mistakes or oversights.

4. Technological Advances in Data Collection: ESPN may not always employ the latest or most advanced technology for data collection in all sports. Rugby union, with its growing adoption of wearable GPS devices, advanced camera systems, and other cutting-edge technologies, often requires data collection methods that go beyond traditional broadcast statistics. Companies like Opta and Stats Perform are at the forefront of integrating these technologies into their data collection processes, providing insights that are both deeper and more technologically informed than those that might be available through a broadcaster like ESPN.

5. Expertise in Rugby Union: ESPN, being a generalist sports broadcaster, may not have the same level of expertise or focus on rugby union as specialized rugby organizations or data companies. This could result in a less thorough understanding or presentation of the sport's intricacies, especially when compared to sources that focus exclusively on rugby and have deep expertise in the sport's unique dynamics and statistical nuances.

In conclusion, while ESPN is a valuable source of sports data, relying exclusively on it for rugby union statistics might limit the depth and accuracy of the information you receive. Integrating data from specialized sources like Opta, World Rugby, and others can provide a more comprehensive and accurate understanding of the game.

While ESPN provides real-time statistics during matches, the immediacy of these updates can sometimes come at the expense of accuracy. In contrast, post-match analysis performed by specialized data companies or governing bodies often reveals discrepancies and corrects errors that might have occurred during the live broadcast.

Yes, the stats are often adjusted after a few hours.

Exactly what I was going to say…..by the time Opta stats are available ESPN have corrected their ‘real time’ data.

Let’s wait for Corn and I’ll do the reveal.

Well Corn isn’t replying. Mapimpi made 2 tackles and missed 2….a 50% success ratio vs your intuition of a 66.7% success ratio. I rest my case, rugby is too complex for even a good intuition,..to be right you have to look at the numbers.

". . . to be right you have to look at the numbers."

This from the rugby noob who probably never set foot on a rugby field in his life and thinks that a smother tackle in the tight phases has the same value as the kind of bone-jarring destructive tackle that PSdT made on Jordie Barrett in the last RWC final that changed the nature of the match with Jordie not even thinking about anything other than passing quickly after that . . . but you keep counting the simple numbers chump!

Waaaaaaaaaaaaaahahahahahahaha!

Played for my school first team when I was 15, played for Villagers U19, selected  for Eastern Transvaal u12 A team, but was injured before the match . Not particularly distinguished….but I played enough to know the game. 

Do you ever do anything on here except yap at me….do you have any original ideas? 

Something with a 20% chance of happening might come through on the first try, or you might have to keep going for five, six or more attempts before it finally does. That’s just how randomness works. It’s unpredictable in the short term, but over time, it tends to average out—eventually, that 20% will show up around one in every five tries.

The same goes for sports stats. Performance doesn’t always match the numbers, and stats don’t capture every exciting or unpredictable moment in a match. But when you look at statistics over a longer period, they do a pretty good job of reflecting how a player or team is really performing. The short term can be surprising, the long run usually tells the real story.

Dave also told us previously that he doesn't care about try scoring ratio. I mean, is there a more important stat in the game!!!??? 

""I mean, is there a more important stat in the game!!!???"

Hahaha …schplotsky…be kind Plum, (sic) is an old codger.

I'm surprised so many people are so eager to let a robot do their thinking for them...

Not thinking…..collecting data. 

Ja, ja Rooi, weren't you the one punting AI to govern us all just a few months ago?

And no Moz, not collecting data, rather analyzing the data for you...lazy...it's only useful if you ask the right questions...

PS...fuckoff Rooi, you're too authoritarian to understand...you think because you're slightly above the 50%, you have the right to decide for the under 50%, and youre p!$$ed off because some of the other  people above the 50 percententers won't let you...typical leftie, you know what's the best for everyone else...fuck off again!

And it's useless if you can't trust the programmers...

Ps...whoosh!!! Rooidrol.

Plum...it's a tool ....a very useful tool and it should be used as such...it just seems that some make it out to be some sort miracle cure for some of humanity's problems...it's nice to be able to use a calculator,  but one should always be able to do basic arithmetic without it too. ..

There was a meme going around of Bill Gates funding medical research and people calling him the devil while Elon Musk is putting chips in people's brains and they're calling him the saviour.

Well, Musk is building robots now...and has an AI, and his own global internet network. 

I've decided that Musk is a terminator, who is also John Connor, who is a terminator, and also John Connor.

"I've decided that Musk is a terminator, who is also John Connor, who is a terminator, and also John Connor."

You should add Tony Stark into that mix somewhere.