Wot I Think: Football Manager 2016

I've won titles and cups, and I've been sacked following a relegation that was more surrender than battle. I've found young stars at bargain prices, and built a team of overpaid ego-machines who are ready to down tools as soon as the going gets tough. After tasting success and failure, I'm ready to tell you wot I think about Football Manager 2016 [official site].

If I were to review Football Manager, it'd consistently rank in my favourite games of every year. But reviewing a specific version of Football Manager means comparing the game to itself. I've spent 130 hours playing the game and more than I care to count considering which elements Sports Interactive have improved, and which elements seem to have taken a step backwards.

It's important to win and lose, and to experience as much of life at the top and bottom as possible, before judging a management game. If there's no joy in the struggle and no challenge following a couple of victorious seasons, the model has failed. At its heart, the Football Manager series is concerned with the hope and sense of expectation that Sean Bean identified as the heart of Sky Sports SoccerBall Actual Football way back when. If the simulation becomes predictable, whether through inadequate behind the scenes number-crunching or the existence of mechanics that players can easily exploit, then it does not represent sport. And, more damning still, it'll quickly become dull.

My failure came at the helm of Bury FC. Under my stewardship, my local club, newly promoted to League One, were sent packing, back to the foot of the foot-to-ball league system. That I held onto my job at all is astonishing, given that the media had predicted a top six finish. We ended up 23rd, though only one point from safety. In all three cup competitions we entered, we were knocked out in the first possible round, most painfully by arch-rivals Rochdale in a 5-1 FA Cup loss.

Relegation was a mere bump in the road, however. Sadly, the road was headed straight for a cliff. Budget cuts left me with no transfer budget and my reliance on young loan players to fill important positions left me with a thin squad. As soon as they returned to their host clubs, I was left with a group of players that could charitably be described as veterans, and few of my affiliate clubs had a great deal of interest in sending their promising young stars to Bury now that we'd been buried.

For my second career, I decided to take the reigns at Manchester United. Before actually putting myself in the hotseat, I did some scouting. Figuring it'd give me a better idea as to how the team worked, as well as an insight into the AI, I simulated six seasons, spectating twenty United matches in each.

Louis Van Gaal's tactics proved to be as predictable and tedious as they are in real life at the moment, although he was far less successful. Sacked by March, with the team 11th in the Premier League table and knocked out of Europe in the group stages of the Champions League, his insistence on playing two defensive midfielders left the team unable to break down opponents. At the time of his departure, United had scored less goals than anyone but the bottom two in the table and were on a four match streak of nil-nil draws.

Bizarrely, he was rehired as Director of Football during the summer, operating alongside Carlo Ancoletti, who reinvigorated the club, finishing 6th after his first quarter-season, and then 2nd, 1st, 2nd and then 1st again. Tactically, Ancoletti abandoned wingers and hired an upcoming backup player for almost every position, rotating the team and getting the best out of young players.

It's good to see some sensible squad-building, as well as to see AI managers showing specific tactical approaches, traits and blindspots. In the past, the AI has often neglected squads, either buying too many players for one position or failing to replace ageing or outclassed players. Simulating a couple of decades into the future reveals robust squad-building, with occasional collapses that feel credible and add excitement.

It may lack back-of-the-box features but Football Manager 2016 is a step up from last year in all the ways that matter to a long-term player. It seems trickier, which is a good thing in this case. The difficulty isn't artificial - although the general opacity of the simulation and its response to human input can make it seem that way. It's in the need to be a reactive manager rather than finding a solid tactic and applying it to every problem.

I found success as Manchester United's manager, eventually. Two rocky seasons in which I scraped fourth position and Champions League qualification were followed by title success in one year and a Champions League win in the next. I signed flops on the scale of Di Maria (Ayoze Perez, what was I thinking?) and brought my current captain and most spectacular individual to the club for £8.75m. That's Sime Vrsaljko, my recommendation for anyone with a few million to spend and in need of a powerful attack-happy wingback.

Football Manager is ever a game in need of interface improvements, not because the interface is poor, but because it's a game that contains so much data that new ways of organising that data cannot be a bad thing. I was put off by some of the changes this year – tactical screens have blocky graphics rather than the tidy shirt-shapes of previous years, and the graphical embellishments on the team instructions screen seemed to add noise rather than clarity.

I've become accustomed to the changes now but I'm not convinced they add a great deal, the latter in particular. One thing I'm wholly unconvinced by is the conversation system, particularly interviews and press conferences. I've been an apologist for FM's media dealings for a long time but the lack of improvements this year is a sign of stagnation. The press conferences and media judgements have to be there if the game is to simulate modern football, but they're so formulaic as to be busy-work rather than character work.

The match engine is improved from last season though. I think. Complaints elsewhere suggest I've been somewhat fortunate in my experiences (and it'd take a thousand hours or more to form a complete judgement, such are the possibilities) but over 120 hours, I've seen all kinds of weirdness, but nothing more than I'd see across a season of actual football.

The biggest complaint I have is that my defensive line seems to leave a larger gap than I'd like between themselves and the midfield even when my instructions suggest they should do otherwise. Careful, consistent marking seems to be beyond a lot of these players, which is believable at lower levels but seems absurd when playing in the world's best leagues. Perhaps that explains the AI's reliance on defensive midfielders?

This year, the learning curve seems steeper. Not in the sense that there are new tactical complexities in terms of player roles and instructions, but in the sense that the successful brands of football seem more idiosyncratic. Most of my success has come with an attacking 4-2-3-1 and that's mainly because it's a reactive formation, allowing the team to go into a defensive shell or to dominate the width of the pitch in attack. Given the right players, of course.

As Graham and I discussed, the new features are barely noticeable this year. Well, the player-created avatar is noticeable but it'd make a stock Bethesda potato head character look like the world's most handsome digital portrait, and doesn't even add much in the way of flavour to the game. The Create a Club feature is a neat addition, though it'd be good to have a more friendly interface given that it's effectively a shortcut to achieve what the editor already could.

In most areas, the opacity of the game is its strength, provided you're as interested in the stories it tells as in the way that it caters to your input. Many games are dogs, bounding around, providing and demanding companionship. Football Manager is a cat: aloof, superior and often disinterested in your attentions. While this year's offering might not do a great deal to grab your attention, spend enough time with it and you'll find that it purrs.

Football Manager 2016 isn't the best of the 3d match engine era – I think that'd go to 14, which consolidated lots of ideas into a redesigned tactical system – but the apparent improvements to both long- and short-term AI make it a much more involving and reactive experience. In my kinder moments, I think of this release as 'no surface all feeling', an edition that has carried out some much-needed housekeeping and improvements to existing infrastructure, forgoing the marquee signing that might excite the fans.

The lack of improvements in areas that have stagnated, most notably dealings with the media and team talks, is frustrating though. There are seeds of good ideas in the drift toward an RPG-like system of relationships and stats, but they're slim and seem half-conceived in some areas. In fact, where the game is improved it may well benefit in Touch (formerly Classic) mode more than in the full-fat simulation. And for the first time, I'm considering spending my time there, and in the entertaining new multiplayer draft mode. I probably won't stick with the multiplayer until Christmas, let alone next season, but at least it's something new to sink my teeth into before settling into the usual decades of toil.

Football Manager 2016 is out now.