Jack Joyce

Mentality changes are hard to explain comprehensively, since we made hundreds of ME changes every year and to some extent most of them will have some effects based on mentality (on either a player or team level). Generally, we went into this year with the aim of making defensive mentalities more viable - previously it always felt that attacking mentalities were more or less the only way to go if you want to succeed as a manager, and we really wanted to rebalance things better and review the effects of mentality holistically. There's been many, many changes to attacking movement, passing, risk assessment - but also to tracking back as well, attacking players are less consistently able to track back when they've committed themselves to an attack (obviously their fitness levels and work rate affect this too). This leads to more of a risk/reward approach to attacking football, whereas before it felt like teams could over-commit in attack but not always feel the cost of that when being countered. Sorry its a bit vague, but we really do make hundreds of changes to the match engine every year and its really hard to fully explain everything!

We covered it a bit in the match blog that went out, but this particular post I copied over from when I explained it the other day in the private beta forums. I'll fix the formatting!

Positions is always a difficult topic, especially in the modern game where positions/roles are so fluid - it almost just becomes a game of semantics! Previously we had a problem - a CM in a 433 is fundamentally very different to a CM in a 4231, they're really not the same position. e.g. you might see Mason Mount as a CM in a 433 and not bat an eyelid, but if you saw him at the base of a 4231 you'd wonder what the manager is thinking! However, in game the AI managers were constantly using players like Mason Mount as CMs in formations where you'd typically expect a double-pivot to be deployed e.g. 4231, 523, 5221. We reviewed all the possible solutions and felt this was best for the realism of the team selection system, fundamentally players in the base of a 4231 are pivots, and these changes reflect that. It does mean that some players needed their positions adjusted by researchers to reflect these changes, and if you see any issues where this still hasn't happened, please report it to the researchers via the forums. You'll probably find that most CMs will end up also having at least some ability at either DM or AM, but this is expected. It does mean that there's a lot less formations used by the AI managers with CMs - but there's still some e.g. 442 diamond and 442 flat which remain. But overall we feel like this has benefits for not only tactical balance and team selection for AI managers, but also for AI shortlisting - since now when they look to recruit a player to play at the base of a 4231 they're not even considering a player like Mason Mount for it. It's worth noting that when you use DMs with no CMs in the team, then those DMs will position higher - and the same for CMs when there's no DMs, they'll position a bit deeper in certain phases.

Comparing FM-generated stats with real-life data providers is rarely a good idea - FM has its own set of definitions and requirements that very often differ from RL providers. Think of FM as having its own in-house data provider, with their own way of defining stats. When data providers track e.g. a clearance they have to assume the intention of the player when doing so, in FM we know the exact intention of every player at all times. Another reason why they differ is we have to work them out purely through code, whereas real-life data providers tend to use people who often have to make subjective calls on what e.g. a 'mistake', 'big chance' or 'dribble' is. FM stats should make sense within the gameworld and be an accurate reflection of what happened in the match - we don't design them with the intention of them just matching up 100% with the statistics you're seeing outside the game. We design our stats based on what we think is most accurate and suitable for our simulation. It does mean that some stats (OPPDA, interceptions, clearances, dribbles) will vary from what you see from the likes of OPTA, but they should be an accurate reflection of team + player performance within the gameworld.

Interesting, could be a gsync or freesync issue of some kind. If you're able to, would be great if you could make a bug report with a video recording of your screen while it is flickering (literally recording your screen with your phone tends to work best).

Sometimes the simplest possibility is the reality - in this case I think what you've found is some inconsistencies with how the data is being set. I agree that you wouldn't expect jumping reach and aerial reach to be wildly different, although it may vary a little bit at times. In the match engine, aerial reach is used to determine how high they can reach to claim crosses, but also how far they can reach to make a save. I would suggest that you post on the research forums for players where you've found a large inconsistency, so that the researcher can have a look. For newgens - typically the newgen attributes are balanced based on the starting DB to a certain extent, but if you're finding issues where these two attributes are vastly different then it'd be a bug to report in the bug forums.

It's worth posting a feature request here: https://community.sigames.com/forums/forum/680-football-manager-feature-requests/

Blocked = passes, crosses and shots Shots blocked is just shots. Sounds like the tooltip for blocked is incorrect which would be a UI bug.