By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions. We in turn value your personal details in accordance with our Privacy Policy.
Please log in or register. Registered visitors get fewer ads.
I recall a few posters have used AI on for various reasons and I have a question of them. Is it possible to ask an AI bot to collect all QPR results, and other pertinent variables, from this century e.g. and interpret the possible outcomes against the next team we are playing?
I reckon it would be easier, but I'm not sure how accurate it would be. 40 years ago QPR would have been seen as a bigger team than Chelsea, and Brentford, Brighton and Bournemouth all smaller.
There must be a point where looking at historical results would have little bearing on future outcomes?
"A detailed dataset of QPR's century-long match history isn't readily available in one consolidated format, but season-by-season archives suggest fluctuating performances with notable highs and lows, depending on league tier and managerial changes."
"A detailed dataset of QPR's century-long match history isn't readily available in one consolidated format, but season-by-season archives suggest fluctuating performances with notable highs and lows, depending on league tier and managerial changes."
I think the pertinent variables part is key. Machine Learning is a product of the data you train it on. There's only so many variables with reliable data available, which I'd assume bookmakers etc already use in relatively straightforward regression models. Id guess the fact that football is relatively unpredictable is because the kinds of variables you can measure only capture a fairly small amount of variance in the outcome. Stuff like "players didn't fancy it", "striker had a barney with the manager", "everything clicked in training that week" wouldn't be data available to an AI to learn from.
Just realised Esox said 'this century' not 'last 100 years'.
Either way, there's little value in making predcitions based on results from entirely different squads and managers. AI is better if you give it better data to work with (but it still makes mistakes, as you can see)....
"Based on the recent form of Queens Park Rangers (QPR) and their next five opponents—Bristol City, Preston North End, Swansea City, Norwich City, and Watford—the following predictions consider both teams' performance in their last five matches:
QPR Recent Form
QPR has had mixed results recently:
Win against Oxford United (2-0): A strong performance highlighting their ability to capitalize on opportunities. Win against Norwich (3-0): Dominant play showed their attacking capability. Draw with Stoke (1-1): Resilience but struggled to secure the win. Loss to Middlesbrough (1-4): Defensive frailties exposed. Loss to Leeds (0-2): Inability to break down strong teams Sports Mole Wikipedia .
Predictions
Bristol City vs. QPR Bristol City has been inconsistent, but at home, they tend to be tough to beat. Given QPR’s recent improvement, a 1-1 draw seems plausible.
QPR vs. Preston North End Preston's form has dipped lately, while QPR’s recent home performances have been solid. Prediction: QPR win 2-1.
Swansea City vs. QPR Swansea has been strong at home recently. Considering QPR’s away record, this might be a tough match. Prediction: Swansea win 2-0.
Norwich City vs. QPR Despite QPR’s convincing win over Norwich at home, Norwich will aim for revenge. This could be tight. Prediction: 1-1 draw.
QPR vs. Watford Watford has struggled for consistency, and QPR might exploit this at home. Prediction: QPR win 1-0.
These predictions highlight QPR’s potential to secure points against mid-table opponents but recognize challenges in away fixtures against stronger teams. Keep in mind, football results can be unpredictable!"
I played with a couple of generative AI systems before the Watford game. I fed in stats from three different sources looking at a range of data - this is what it said:
Based on the data, the most likely score for Watford vs. QPR is 2-0 in favour of Watford. Here's why: Watford's Scoring Pattern: Watford averages 1.63 goals per game at home and has scored in every home match this season. Common home score lines for Watford include 1-0 and 2-1, suggesting a strong likelihood of scoring at least 2 goals.
QPR's Defensive Weakness: QPR concedes an average of 1.25 goals per away game. They’ve failed to keep a clean sheet in 8 of 9 away matches this season. QPR struggles with goal scoring, failing to score in 50% of their away matches.
Watford's Defensive Strength at Home: Watford has kept a clean sheet in 50% of their home matches. QPR's away scoring rate (50%) is relatively weak, making it less likely they will score.
Recent Form: Watford has been defensively solid in recent home matches, conceding only 4 goals in 8 home games this season. QPR's recent form includes a lack of scoring threats, especially against stronger defensive sides.
Likely Score line: 2-0 Watford: Expected to score 2 goals based on their strong home attacking performance. QPR: Unlikely to score, given their poor away offensive record and Watford's solid home defence.
Obviously, we drew 0-0 so fed in three lots of stats into the same LLMs and got a detailed explanation why it happened - in summary:
Watford's failure to convert despite dominating possession and creating more open-play chances may indicate inefficiency in their attacking execution. QPR's ability to create from set pieces without converting suggests a missed opportunity to exploit their style against Watford's aggressive defense.
Comments: This match highlights a classic example of how inefficiency in finishing can overshadow tactical strategies and result in a frustrating draw for both teams.
I think what you're describing here is a pretty complicated set of models which learn from past performance to try to predict future performance. On any dataset in the world there are flaws to that concept as certain environmental factors don't get repeated and / or can't be predicted - like COVID for instance.
With football, so much changes week-to-week; team selections, weather, player availability, game schedules, etc etc. Let alone how much changes year on year. The QPR team of 2011 for instance would beat this one roundly and squarely about 98 times out of 100, Luton went from non-league to Premier League in 10 years, Brentford, Brighton, Forest have all been wildly different teams to they are now this century. All that considered, any model would probably largely ignore older data in favour of more recent points which would correlate far more accurately with current performance so there would be little point in gaining the data in the first place.
Companies charge a LOT of money to build data models of this type for corporations so I'd be a little surprised if any open source AI platform would just regurgitate it exactly how you want it. AI in this instance would be very useful to collect the data and probably build the algorithms to predict the future but the entire model as a whole would likely score itself very poorly in its ability to predict what happens next. No model on planet earth would ever have predicted us to win at Burnley in 2023 or Leicester last season. it would probably largely just confirm any intuition any one of us could have rather than predict outlying events and when they'll occur.
I think what you're describing here is a pretty complicated set of models which learn from past performance to try to predict future performance. On any dataset in the world there are flaws to that concept as certain environmental factors don't get repeated and / or can't be predicted - like COVID for instance.
With football, so much changes week-to-week; team selections, weather, player availability, game schedules, etc etc. Let alone how much changes year on year. The QPR team of 2011 for instance would beat this one roundly and squarely about 98 times out of 100, Luton went from non-league to Premier League in 10 years, Brentford, Brighton, Forest have all been wildly different teams to they are now this century. All that considered, any model would probably largely ignore older data in favour of more recent points which would correlate far more accurately with current performance so there would be little point in gaining the data in the first place.
Companies charge a LOT of money to build data models of this type for corporations so I'd be a little surprised if any open source AI platform would just regurgitate it exactly how you want it. AI in this instance would be very useful to collect the data and probably build the algorithms to predict the future but the entire model as a whole would likely score itself very poorly in its ability to predict what happens next. No model on planet earth would ever have predicted us to win at Burnley in 2023 or Leicester last season. it would probably largely just confirm any intuition any one of us could have rather than predict outlying events and when they'll occur.
"Queens Park Rangers (QPR) will be playing against Bristol City at Ashton Gate on Saturday. Both teams have had contrasting starts to the season, with Bristol City currently in 11th spot in the Championship table and QPR aiming to extend their unbeaten run to five matches.
Bristol City has only one win in their last five second-tier clashes and has struggled with inconsistency, which threatens their hopes of competing for a top-six spot. However, they have been relatively strong at home, being unbeaten in seven of their nine home outings this season.
On the other hand, QPR has enjoyed back-to-back successes at their Loftus Road base, easing past Norwich City and newly-promoted Oxford United. Despite this, it has not been all smooth sailing for QPR, who were booed off by some sections of the crowd at half-time in their last match.
Given Bristol City's home advantage and QPR's recent form, it could be a closely contested match. Predictions suggest that both teams might not score, and there could be under 2.5 goals in the game. However, QPR's recent unbeaten streak gives them a fighting chance."
Like anything AI, you have to fact check everything. Like the incorrect umber for unbeaten games in the 2nd sentence.
"Queens Park Rangers (QPR) will be playing against Bristol City at Ashton Gate on Saturday. Both teams have had contrasting starts to the season, with Bristol City currently in 11th spot in the Championship table and QPR aiming to extend their unbeaten run to five matches.
Bristol City has only one win in their last five second-tier clashes and has struggled with inconsistency, which threatens their hopes of competing for a top-six spot. However, they have been relatively strong at home, being unbeaten in seven of their nine home outings this season.
On the other hand, QPR has enjoyed back-to-back successes at their Loftus Road base, easing past Norwich City and newly-promoted Oxford United. Despite this, it has not been all smooth sailing for QPR, who were booed off by some sections of the crowd at half-time in their last match.
Given Bristol City's home advantage and QPR's recent form, it could be a closely contested match. Predictions suggest that both teams might not score, and there could be under 2.5 goals in the game. However, QPR's recent unbeaten streak gives them a fighting chance."
Like anything AI, you have to fact check everything. Like the incorrect umber for unbeaten games in the 2nd sentence.
Also very true. I jsut asked Chat GPT what is the biggest nation England have knocked out of a World Cup in a match played outside of England...
The biggest team England has ever knocked out of a World Cup in a knockout game outside of England is **Argentina**. This happened during the **1986 FIFA World Cup** in Mexico, in the quarterfinals.
In that famous match, England defeated Argentina 2–1, with Gary Lineker scoring the winning goal. However, the match is most famously remembered for Diego Maradona's controversial "Hand of God" goal and his brilliant solo "Goal of the Century" shortly after.
This game remains a landmark in England's World Cup history, as Argentina was one of the tournament's powerhouses, and Maradona was arguably the best player in the world at the time.
Also very true. I jsut asked Chat GPT what is the biggest nation England have knocked out of a World Cup in a match played outside of England...
The biggest team England has ever knocked out of a World Cup in a knockout game outside of England is **Argentina**. This happened during the **1986 FIFA World Cup** in Mexico, in the quarterfinals.
In that famous match, England defeated Argentina 2–1, with Gary Lineker scoring the winning goal. However, the match is most famously remembered for Diego Maradona's controversial "Hand of God" goal and his brilliant solo "Goal of the Century" shortly after.
This game remains a landmark in England's World Cup history, as Argentina was one of the tournament's powerhouses, and Maradona was arguably the best player in the world at the time.
To think people are being made redundant and replaced with this level of analysis.
Chairman of the Junior Hoilett appreciation society
Asked ChatGPT to recommend some signings for January from abroad, since that is what we'll be doing. It recommended us getting Ilias Chair back from Anderlecht on loan! What does it know that we don't?
Asked ChatGPT to recommend some signings for January from abroad, since that is what we'll be doing. It recommended us getting Ilias Chair back from Anderlecht on loan! What does it know that we don't?
I asked Copilot who gave me this:
Queens Park Rangers (QPR) are looking to strengthen their squad in the January 2025 transfer window to boost their chances of staying in the Championship. Here are a few players they are reportedly targeting:
Rodrigo Muniz (Fulham) - The Brazilian striker has struggled for game time at Fulham and could bring much-needed firepower to QPR’s attack1. Devante Cole (Barnsley) - With 12 goals in 20 League One appearances this season, Cole could help solve QPR’s goalscoring issues1. Kieffer Moore (Bournemouth) - The experienced striker has a proven track record in the Championship and could be a valuable addition1. Scott McKenna (Nottingham Forest) - The centre-back could bolster QPR’s defense, having previously been linked with the club1. Joe Gelhardt (Leeds United) - The young forward is another potential target to enhance QPR’s attacking options2.
If you are looking for how likely we are to win the next game etc, the work is already done for you. Bookies spend BILLIONS having more data than anyone else (hence Brighton/Brentford advantage in recruiting).
Those odds and backing them into an applied probability, are the closest thing you're going to get to a prediction.
Of course, football isn't a game played on paper...
This is all predicated on the flawed notion that the future can be predicted from the past. It can't, especially for a single event where the margin of difference is usually very fine.
Seems to me what people are calling AI at the moment isn't AI at all, it's just a glorified internet search and not a very reliable one at that.
And bookies are a red herring. They don't profit from being good at predicting results, they profit from making the market and employing a buy/sell spread, like stockbrokers do.
Copilot is optimistic, though I'm not convinced by the word 'suggest'.
Queens Park Rangers Football Club, commonly known as QPR, is a professional football club based in White City, London. They currently play in the Championship, which is the second tier of English football. QPR has a rich history, including winning the League Cup in 1967 and the Championship in 2011. They have also been runners-up in the First Division in the 1975-76 season and reached the final of the FA Cup in 1982.
Recently, QPR has shown improvement under the management of head coach Marti Cifuentes. For instance, they secured a victory over Oxford United with two goals from Sam Field, which helped them move away from the bottom five of the Championship. The team has also achieved four consecutive clean sheets, indicating a strong defensive performance.
Overall, while QPR has had its ups and downs, they have a dedicated fan base and a history of notable achievements. Their recent performances suggest they are on an upward trajectory.
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
