Jump to content

rjpageuk

Members
  • Content Count

    2,977
  • Joined

  • Last visited

Recent Profile Visitors

2,945 profile views
  1. rjpageuk

    Football Thread 2018/19

    You are moving the goal posts. You stated that Mane is possibly the most wasteful striker in the league to back up your point that he shouldn't be in the team of the year and compared him to Son/Kane and now you want to talk about goals per minute? What does that have to do with how wasteful a player is? Yes Kane has scored at a higher rate in terms of goals/min than Mane, but he also takes a lot more shots.
  2. rjpageuk

    Football Thread 2018/19

    18 goals from 80 shots (and xG of 14.39) for Mane vs 17 goals from 102 shots (and xG of 16.12) for Kane says otherwise.
  3. rjpageuk

    Football Thread 2018/19

    Tarkowski and Mee have been the first choice centre backs for Burnley all season this season too though so what has changed? It is definitely true that Pope had an extraordinarily good season last season (in a similar way to De Gea). By way of comparison Burnley conceded 39 goals last season with an xGA of 52, and this season they have conceded 57 with an xGA of 55. The basic argument presented against using xG was simply that Burnley had similarly "lucky" stats the season before too and so it was more likely a problem with the model. At the time I posted an article from statsbomb showing that if you used a longer timeline this didnt hold so true, and I notice they have now posted a new article (on both Burnley and Man U from earlier in the season after only four games, but in fact the stats have continued in pretty much the same way since then to now) - https://statsbomb.com/2018/09/are-manchester-united-and-burnley-just-regressing-to-the-mean/. My explanation is just that the level of variance is so high that you cant even expect things to be close to levelling out over a season. Even 38 games is not a lot of data - you might use this to question the value of a system where it cant even be relied on over the course of an entire season but it always comes back to the point that it is the best we currently have access to, and that even if it is subject to huge variance it is still a lot more reliable than the actual results in predicting the future.
  4. rjpageuk

    Football Thread 2018/19

    You say it is clear from watching the game but that's not how the narrative goes either in terms of how the match analysts / commentators explain the game or in terms of how most fans post about how a game went. Most fans are not aware of how lucky or unlucky their team has been in any given season or even on a match by match basis. Your natural bias remembers events where your team missed chances (or increases their quality) and forgets about events where the other team missed chances. Compare the posts in the relevant threads here to say Man U's performance vs Arsenal to their performance vs PSG. Almost every fan and match commentator / analyst has completely results oriented thinking. The last time I brought up xG on this board was with relation to explaining how lucky Burnley were during last season and that they should have expected to be caught up in the relegation battle. This was rejected by the result Burnley fan (forgotten your handle sorry), and in general was not well received so I dropped it. This season they are back to performing pretty much as you would expect in xG. Last season one of the other big stories xG wise was how lucky Man U were particularly with regard to goals conceded. When I pointed this out it was also badly received - De Gea is just a world class shot shopper - of course he is stopping all these shots. This season they are back to performing pretty much as you would expect in xG. You cannot expect your brain to well assess probability without help because it doesnt.
  5. rjpageuk

    Football Thread 2018/19

    Have a look at caley graphics on twitter which has xG for a lot of games and some detailed explanations (e.g. https://cartilagefreecaptain.sbnation.com/2015/10/19/9295905/premier-league-projections-and-new-expected-goals#methoderology).
  6. rjpageuk

    Football Thread 2018/19

    Yes but there is no game that it "should come in". The luck happened in the past and is not relevant for future predictions. No it doesnt, and yes it does matter. You expect the best players to perform above xG over a decent sample size. Gotters - there is always resistance to posting xG data in the football threads. I have given up. I have no idea why when it comes to sport people use a different methodology to assess results or predict the future than they would in pretty much every other field.
  7. rjpageuk

    Liverpool Football Club Thread

    xG for the game was 2.2 - 0.5 which is about as good as it gets for a game of this magnitude. I think saying that Bayern frustrated our attack is just results oriented thinking - we just missed our chances. I do think the second half performance is a little bit of a worry as we have a number of times already this season seemed to run out of momentum/steam in the second half. I guess in Europe the psychology changes as conceding a goal when it is 0-0 is pretty disastrous.
  8. rjpageuk

    Liverpool Football Club Thread

    Saw this article linked to on RAWK and liked it, so wanted to share it here: https://www.football365.com/news/the-2018-19-premier-league-title-race-expose-the-fraud
  9. rjpageuk

    Liverpool Football Club Thread

    I think it is very harsh to say Lovren and Trent performed poorly. Sure pretty much every chance Man City created could be tracked back to Lovren, but Man City are a world class side, and they had the lowest xG they have had all season at Home. Our entire team were the reason for that including the defenders. Both their goals were excellent finishes from low % scoring opportunities [EDIT: Man City's xG was at 0.3 on the 80th minute, ours was 1.2, their chances after that came with us chasing the game]. I think the small things went against us last night, and a draw was a fair reflection of the game and would have been an excellent result for us.
  10. rjpageuk

    Super Smash Bros. Ultimate

    In online quick play mode if you get your GSP high enough (currently around 2.6m, but is constantly increasing - might be top 5% of players?) you play in "elite" mode. It is a pretty major task to try to do this with every character.
  11. rjpageuk

    Super Smash Bros. Ultimate

    Good games guys, we got there in the end!
  12. rjpageuk

    Super Smash Bros. Ultimate

    If you make a public arena you need to tell us the id not the name. You can find the id by clicking on the ring in the top left corner
  13. rjpageuk

    Super Smash Bros. Ultimate

    I thought you could search by lobby name but you can only search by lobby id which is apparently not the same thing. Definitely go back to the friends idea!
  14. rjpageuk

    Super Smash Bros. Ultimate

    Yesterday I went for a public lobby with a password as I figured then you dont need to bother adding people as friends etc., not sure what the downside of that is over having a friend only lobby?
  15. rjpageuk

    Super Smash Bros. Ultimate

    I am in! You can set it up though
×

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Use of this website is subject to our Privacy Policy, Terms of Use, and Guidelines.