Jump to content

It's Baggio

Members+
  • Posts

    35
  • Joined

  • Last visited

Posts posted by It's Baggio

  1. 3 minutes ago, janrzm said:

    It does my head in when someone on here says it happens in real life......But, I'm a Wednesday fan, we took an absolute age to appoint Monk when Bruce left in pre-season.......

    I'm sure there are exceptions and in that case, the circumstances were a little different.

    However, every time I've applied for jobs it takes ages. Clubs sack their managers at the seasons close, then don't do anything until pre-season has started again.

  2. 1 hour ago, knap said:

    With top clubs the tactics on the recommended list should all still work, as I've seen nothing to suggest there are any issues with the set ups. FM 19 was dominated by Long throws and perhaps short corners. For FM 20 it is probably  worthwhile changing these to other set ups.

    If you have problems it is also worthwhile trying Gegenpress settings.

    What/where is the recommended list?

    Sorry if it's a stupid question but I didn't play much FM19.

  3. 9 minutes ago, westy8chimp said:

    Im not personally opposed to the model, but it wont really change anything, still relying on what the ME calculates a chance to be and how good the chance was etc. And then peoples perception and use of that information.

    At least xG has context though, what with it being based on actual shot data. Plus it's actually quantifiable, what constitutes a 'clear cut chance' in the current set up? 50% chance of scoring? More? Less?

  4. 2 hours ago, Svenc said:

    Yees, BUT IT NEVER HAPPENS TO THE SAME TEAM! Except it absolutely does. Ask Zidane last term, Klopp about his last season at Dortmund, Conte in 2015, Wenger when he lost the title race to Leicester, every single opposition of Burnley last term....... and that despite all of those managers managing space on their pitches, and players, and circumstance, rather than the most simplistic numbers on a spreadsheet. Frankly, with any should simplistic outlook on football, it should happen far more often to such players as it usually does. Which is, typically not that often. This is where aforementioned confirmation bias kicks in. If Guardiola had taken a look at all the matches he at all conceded last term, even he would have found that his opposition would have scored from few shots, including 3 shots off 4 against United. However, he didn't concede in all matches, despite his fairly aggressive style of football. Considering he barely conceded 6 shots average, that is only to be expected. Unless SI code an ME where it is possible to go a season without conceding shots, you will concede goals. It doesn't matter when and from where, as each one is a chance of hitting the back of the net.

    Perfectly captures the reason why I've written my post from yesterday, suggesting in-game feedback that would mock users who gawk at nothing but final match stats rather than supporting them in doing so (the final match reports as of FM are based on FM's statistics and suggest similar -- if a team had more shots it was "unlucky", which is neither football nor the ME). Whilst it is 100% viable to primarily have an emotional interest into football -- that still doesn't detract from the fact that the average football discussion on your average FM board makes puppies weep.

    The worst thing is that all this "cheating AI" nonsenes typically distracts from actually issues. F'r instanced, often times such players have a point, even though it's not what they think it usually is. AI going too rigidly defensive too readily, pitches that can be controlled too easily even with below average sides against superior ones; maybe even that ball over the top being a tad too efficient depending on the release etc.  On the occasion, it may be even a marking bug triggered. It also typically distracts from the fact that top teams managed by AI oft face similar issues, as they face defensive opposition every week likewise -- who could be more comptetitive with somewhat more logical match management that wouldn't make them drop the few additionally points. All of which FM's simplistic stats won't reveal.

    Even with more advanced stats, managing numbers still ain't managing. Football and the ME and the AI does not agree, sorry. edit: And SI may well realize that even at the current level AI, it has an edge over most of its playerbase in that way. Even though it only shows in the odd match or two. In other words, would it improve, the perception of the game being inherently "unfair" would only grow stronger. It's time to get rid of the notion that having more shots/possession was inherently in any kind of way managing. Or that havin g umpteenth more shots was inherently dominating anything (the CCC stats was set out to solve some of this a decade ago, but it can't work much the same way for both obvious and not that obvious reasons).
     

     

     

    The game really needs some sort of xG model. It's been used by analysts for a long time and is now commonly used in punditry. Counting shots, as you point out, can be mis-leading and the 'half chance' 'clear cut chance' totals are vague and seem slightly random at times.

    All shots in the game are mapped so some basic xG forumula shouldn't be difficult to implement. This could then lead to a (much needed) revamp of th analysis features whch currently is a bit of a token IMO.

×
×
  • Create New...