Baker’s boys

It never hurts to look in on an old PES from time to time, to refresh one’s memory about what’s great about the series – and to see that really and truly, what we have now might not be so bad after all.

Yesterday I dug out my Xbox360, untouched since roughly mid-2014. I plugged it in and fired it up. It had a FIFA14 disc in the tray. I have no recollection whatsoever of playing FIFA14 on the Xbox360.

I popped in the PES6(360) disc. A commenter, Darryl, had mentioned PES6(360) last week. I was lavish with my praise for this one-off, quirky edition of PES6, which is unrelated to the PS2/PC version, despite the oddly persistent belief of many that it’s the same.

How would it stand up now?

In a nutshell, good but not great. I played one match of my mature ML save. I had a galaxy of stars. Maldini, Mathieu, Messi. The game played fast and furious – and there were no fouls. Seriously, did we only just start noticing no-fouls since PES2015? Has it pretty much always been with us since the anti-fouls hullaballoo that surrounded PES5?

I enjoyed this one match, but all I could think was: PES2018 is so much better than this.

There. I said it. A much-maligned nuPES plays better than one of my most beloved old classic PESes.

And so I fired up PES2018 with renewed sense of purpose and optimism.

A full session followed. I’m all the way through the January transfer window in Season 2 and out the other side.

I added three other players to my roster. Xabi Prieto from Everton. Rubén Castro from Free Agents. And, from my Youth team – Wayne Rooney. He’s only 17, but was famously good at age 17 in real life, so he’s not too bad here right now.

I thought long and hard before pressing the Rooney button, with all my pre-PES2018 talk about House Rules in mind. But the PES2018 AI seems to have liberated me from needing House Rules. For now.

My full post-window squad:

I’m shaping my First XI and squad around players who represent a feeling of classic PES individuality to me.

There are quite a few players on that screenshot who are ‘nothingy’ types of players – just bundles of pixels with a name attached. But most are at least above the waterline on the individuality front.

(I was going to do a player-by-player individuality assessment in today’s post, but ran out of time. Will do it in Friday’s now.)

Shelvey and Richarlison are still too new to pass judgement, but seem worryingly anonymous so far. This individuality semi-House Rule of mine means that if they don’t shape up in the rest of this season, I’ll probably move them on at the end of it, or this time next season at the latest.

Certain players will likely stay with me for the duration of this Master League.

L BAKER is a little star. Dare I say it, there are strong shades of Shimizu with him. Baker handles like Shimizu would have if he’d been able to shoot.

Here’s a decent mid-range finesse goal from him:

Scoring goals isn’t really his main game, so far. He’s a versatile midfielder who links play – not a stopper (he doesn’t play at DMF), but a maker.

I’m intrigued to see where Baker’s story goes – that’s the essence of Master League, right there. In real life, Lewis Baker seems to be drifting from loan to loan, and trying not to be the next James Wilson. In the much-realler life of PES, he seems destined for legendary status.

Finally today, a PES2018 quirk.

Look at this offside decision, and see where the resulting free kick is taken from – both are wrong:

The still-new-fangled ‘active play’ thing means Veldwijk being offside in the first phase (0:16s) would have ‘worn off’ by the time he runs across to touch the ball several seconds later.

Debatable, perhaps, but what isn’t debatable is the ludicrous position the free kick is taken from. Never seen anything like this in PES before.

Updated: 24th October 2017 — 11:02


  1. Paul – not conceding in this PES to this AI can be as rewarding as scoring.

    And after yesterday’s penalty discussion, I spotted this in the wild:

    Do we think this kind of thing will be fixed?

    JS Hutt – 4-4-2 diamond did great for me in Championship Manager 5, I think it was – the first edition that had the 2D circles.

    Cook/Mr Noodles – I’ve started going long-ball a lot more, as with the penalty from yesterday where I basically just launched one hopefully. The mechanics of the game are now such that the AI players take a bad/clumsy touch a lot of the time, giving lots of scope for stealing in and creating situations.

    Tommy – the kind of weave-dribbling you do with Sessegnon at around 1:40 is something I’ve found myself doing more on PES2018 than on any other PES, as the AI is so good at shutting down passing lanes.

  2. Paul, Cook – Superstar. Don’t be fooled, it’s not very forgiving. It’s been a hard slog, it’s definitely harder than SS last year. Despite a couple of good goals, Barkley didn’t do it for me and sold him after 6 months. Renato Sanches is rediculous and still young.

  3. Isn’t it amazing how nobody talks about Legend. I personally can’t wait to get there, although I know people describe it as a joke, more like a bonus content, animal-heads-style add-on. My feeling is that the ultimate destiny of this ML is to be played on Legend. PES2012 was like this, escalating slowly up the difficulties.

  4. Great compilation Tommy! SS is definitely harder than last year’s. I was hoping it wasn’t a case of this year’s legend is last year’s SS etc. but I have no such worry anymore! This *might* be the longest running ML I do, as I’ll want to win on Legend. This probably means I’ll for the first time end up at some point using regens.

    On regens, I’ve not touched them. Does my academy just get infinitely bigger & the players in it better and older?

  5. NG – Absolutely. There was a couple of games where I rode my luck, my stand in GK, Rubinho, was obviously trying to prove a point, he pulled off some worldies, and the CPU missed a few chances that it always usually puts away, so I think I was riding a wave of defensive momentum, it won’t last obviously.
    My Goals for tally is paltry, something like 14 goals in 15 matches, but my GA column is one of the best in the league. 1-0 to the Arsenal.

    That foul is absolutely shocking, I’m just lost for words how, in any universe, that can not be triggered by the collision detection model as a foul.

    I have a worrying suspicion that the patch if/when it arrives, will just make the COM slide tackle more. It wont alter the logic that decides when a foul is a foul.
    Its the lazy half-assed approach Konami always default to, so I expect little.

    Jay – No. eventually, usually after the age of 18, players will drop out of your youth ream if you don’t sign them, you don’t keep them forever.

  6. Tommy – Some class goals there. At risk of flogging a dead horse on my part, is that Advanced shooting? Looks like it with some of the power you’re getting.

  7. Keith – Nope, just basic.

  8. Tommy – Ta, good stuff.

  9. I just attempted to submit a long post about Legend difficulty, and it was apparently deleted as spam. I’ll try posting it in two parts, just in case it’s a word count thing. 😉

    NG – I experimented with the Legend setting over the weekend after I encountered a weird glitch that completely destroyed the bespoke 3-5-2 formation that I’ve been using so far this year. And it’s… well, *interesting*. I’d hesitate to describe it as an entirely new type of AI, but there are clearly fundamental differences in playing philosophy and the global ruleset between Legend and the Professional/Top Player/Superstar AI archetype.


    * Legend either tones down assistance on passes, shots and crosses, or applies blanket stat reductions across the board on your players; it’s possibly both at once. Passes that are routine on Superstar are more likely to be underhit, overhit, or inaccurate; they also seem to take slightly longer to power up. For example, a 30-yard drilled ground pass to an unmarked forward has a higher probability to leave the ground and travel just to the left or right of him, increasing the time it takes to bring it under control.

    * AI players enjoy a massive boost in reactions and acceleration; you rarely get a second to put your foot on the ball to allow teammates to run and develop a move, or to wait for someone to get into position. That slight increase in the power-up duration for passes means that, unless you turn away from an incoming marker, they’ll just take the ball from you before your player can swing a leg. This is a huge shock to the system, and probably a hurdle that will send most players straight back to Superstar. One-touch passing helps, but – due to the presumed stat nerfs – the accuracy drops off a cliff after two or three passes. You don’t get to just tap the ball around – you really have to work to keep it.

    * AI pressing is much more aggressive, and they position themselves more effectively to block presumed passing lanes. You really notice the tighter marking. In one-on-one situations where you attempt to pass to a teammate just beyond a marker, you absolutely have to make enough space to stroke the ball past them, or they’ll always get a foot to it. If you attempt a longer pass where there’s an AI defender along its path who might have a chance to reach it, he almost certainly will.

    * AI teams pass the ball much more than on Superstar, especially in the final third. They can still hit a pinpoint delivery to a player running into space, but seem more inclined to do this further up the pitch – and, moreover, after developing the move more convincingly. It didn’t seem to rely on the “transition -> simple pass -> simple pass -> PEAK ZIDANE 40-YARD PASS TO PELE WHILE YOUR DEFENDERS WAVE TO SOMEBODY IN THE CROWD” chance creation model that the AI often favours on Pro, TP and SS.

    * AI players with dribbling traits are a fucking nightmare: you really need to concentrate while marking them, especially in the final third. They’ll quite happily slalom through your defence if you give them the space. It’s most noticeable with wingers, who regularly cut inside and carry the ball into the box if you give them the opportunity to do so. Arjen Robben scored a very Arjen Robben-type goal against me in one game, where he made a blindingly fast (trick) turn inside from the wing to run parallel to the goal along the edge of the penalty area. As I moved a defender to mark what I saw as the obvious pass, he suddenly thumped the ball into the net at the near post with his left peg. In almost 200 matches in PES 2018 so far, I haven’t seen anything quite like that.

    * As the AI plays with such intensity, I experienced something that hasn’t really occurred for me so far this year, and was (in my experience) completely absent from PES 2017: I had spells of being penned back into my penalty area. In these instances, attempting to pass out from the back is suicidal: you fucking *get rid* as soon as you can.

    * Here’s the most interesting thing: Pro, TP and SS incrementally turn up the heat by gifting AI players with increased strength and tackle resistance. Playing on Superstar, the AI wins the vast majority of headers in midfield and attacking positions, simply because you either exploit a split-second opportunity to run a player in front of them with Super Cancel, or they lock themselves into position and will not budge. Even then, I rarely find that any player in my team has recorded more than ten successful headers when I look at the monthly stats screen. I played a few games against the Netherlands on SS and Legend to compare the difference, and found that Legend *completely removes the boost to AI strength stats*. On SS, Dutch forward Quincy Promes either directly won the ball against my centre-backs, or the ball dropped into space after inconclusive contact. But on Legend, he didn’t win *anything* in the air against my central CB. This applies across the board: if your player should have a strength advantage against an opponent, then he will. It’s such a breath of fresh air after weeks of Superstar’s bullshit in this respect.

    I’m not sure if Legend is a parallel AI development track, or simply an example of the devs adjusting FIFA-style ‘sliders’ behind the scenes to mix things up, but it feels like a very different approach to simulating football with the tools at their disposal. The net result is that AI teams are much more proactive, and they don’t surrender the ball as readily. I average around 60% possession against Superstar, but the total was much closer to 50% against Legend.

  10. Pro, TP and Superstar generally play a direct and primarily reactive brand of football that is closer to the PES 2017 model than the majority appear to have realised. As I’ve mastered the art of defending against the magic through-balls and crosses that were so frustrating at first, I’ve started to think that these serve a dual purpose in terms of design: for the AI to score goals, sure, but also to provide an opportunity for the player to win the ball back immediately, and begin their next attack.

    In short, I suspect that the PES lead designers don’t want their core AI levels to have too much possession against players, because it might piss them off. They’re right, in a sense – FIFA’s Legendary AI has historically played keep-ball with dirty tricks to artificially inflate difficulty, which leads to miserably dull matches. Legend is a glimpse at a PES where the AI actively *wants* the ball all of the time, rather than primarily being set up to provide challenging yet surmountable obstacles to the player’s build-up and chance creation.

    But does it work? I’m on the fence. I experienced some genuinely absorbing moments, and I love the idea of the AI bossing possession, but the relentless hounding is wearying when there are hardly any breaks in play to take a breath. I also think it goes too far in nerfing the capacity of your faster players to stretch the play, as AI defenders reel them in waaaay too quickly.

    As with every other mode, it needs fouls, cards and injuries – and, unless they tune it in the next patch, a potential drop to -1 speed to make it less of a whirlwind.

    I might try it again today and see how it stacks up after a few days away from the game.

  11. James, very interesting read and as ever very detailed and insightful.
    It worries me that I find the Insane Pressure and harassing by the AI on TP unrealistic, sounds like Legendary would drive me mad.
    Also nerfing your players stats in order to make a game more difficult is yet another half-assed approach by Konami, it should always be the COM AI that is smarter and better, not handicapping the player to make the COM look better.

    I’ve always been sure that Konami’s approach to scaling the difficulty levels were by giving a blanket across the board % stats boost to COM teams, depending on the difficulty level, ie:

    Playing Burton Albion on Regular, a players stats may be say 65 for shooting, 68 for Speed, and 63 for passing
    that gets a boost as the levels go up, stats get boosted….so….

    10% increase for Professional level:

    Shooting – 71, Speed – 78, Passing – 69

    Top player 20% boost:

    Shooting – 78, Speed – 82, Passing – 80

    Superstar 30% boost:

    Shooting – 84, Speed – 88, Passing – 86

    That’s why on TP weaker teams like Burton play well above their ability levels and are able to ping laser guided passes about, volley from 30 yards, back heel blindly 20 yard intricate passes etc, and why such supposedly lower teams don’t feel any different from the bigger teams, konami’s lazy programming makes the teams all kind of blur into one another, so every match is against Random team X, as opposed to actually feeling the difference between playing a bottom of Div 2 side or a Champions League side.

  12. James – thanks for that in-depth post on Legend. Very helpful 🙂 maybe it’s a case of having an all 90s squad before taking on Legend.

    I’ve found the skill training cards very effective again this year, just in case anyone is still suspicious

  13. Paul – the fact that PES and FIFA alike do not take team form and league position into account is one of my biggest bugbears with football games. It makes the career modes feel less like a campaign, and more like 40-odd isolated matches with no real context other than what your struggling imagination can muster when a 4-0 walloping of table-topping Man United is followed by a 0-2 reverse at Burton Albion. I want career modes to tell me an opponent’s recent form, highlight any players of note (purple patches, blue arrows, notable goal/assist/tackle/pass stats), and then tailor the match AI accordingly. IMO, a combination of adaptive difficulty and contextual difficulty could revolutionise football games, as long as the systems are broadly transparent and fair – and openly signposted. (I could talk about this subject *all day long*.)

    Jay – Better players would help, I’m sure, but I’m positive that Legend also scales back/disables assistance on passing, movement, first touches – pretty much everything that you do. It’s much more technical.

    A few additional points of interest, after dipping a toe further into the Legend pool this afternoon:

    * One of the big issues with Legend (also a problem on Superstar) is the “refractory period” that occurs whenever one of your players is tackled is far too lengthy, and often too pronounced (a stumble, a five-yard stagger, a dramatic collapse). Meanwhile, Legend AI players are straight back at you from the very instant you disposes them – sometimes before you can even power up a pass. The funny thing is, this direct input prohibition also applies to *everyone else on your team*, so you can’t even switch and move another player. I understand why this stun/knockback mechanic is necessary (FIFA has it, too), but it’s way too strong against high-level AI.

    * The sheer speed and inventiveness of AI play (relative to other difficulty levels) means that you fight against the current player change mechanics just as much as their movement and creativity.

    * The AI’s ability to get in and tackle you before you can power up a pass from a standing/walking position is a little suspect, but I’m finding that the AI players that do this the most are their best defenders – CDMs, CBs. That’s not so bad. This mechanic is complicated by the fact that first touches and general dribbling are much more technical on Legend. It really feels like the training wheels are off on this mode.

    * The AI makes far more passing mistakes than in any other mode – but they’re all viable gambits to find a teammate and build a move, and don’t feel like RPG-like random fail states.

    * AI players will shoot from greater distances, and in situations that I haven’t seen elsewhere. However, they’re a bit too fond of passing it into the box from just outside, where a forward will take a touch, move to the side, then make one of those horrible “floaty” PES shots that you always feel that your keeper should save. It’s really hard to defend against these.

    * I’m conceding at least a few fouls (plus additional “advantage” phases) per game, because you *can* just wallop tricky opponents on Legend – they can’t just ride almost every tackle, good or bad.

    * I’m almost certain that the AI is directly reading/reacting to controller inputs – rather than responding exclusively to events inside the simulation – to improve its capacity to mark and tackle. It’s the equivalent of a mate looking at your hand movements during a head-to-head penalty shootout, but you don’t have the real-life option of telling him to fuck the fuck off and hiding your controller under a cushion. Not sure how I feel about it being this blatant.

    * PES 2018 first: in the game I finished just before writing this, the AI’s Verratti was booked for diving.

  14. I’ve found a patch for winning penalties. My wife is a teacher and it’s half term; so she’s had the delight of watching me play a few times – her choice, I’ll add. To her credit she does watch on average 4 real life games a week with me, despite having no particular interest.

    An hour or so ago she said, “so do you get injuries on this? Can your striker get injured for a year?” I rubbed my hands together, licked my lips and gave her a historical account of Master League since PES3 – my first. It transgressed and ended on me mentioning that in game week 32ish of my third season, I was still to get a penalty. I proceed to the next game and promptly get a penalty. I don’t know who laughed more – me or her.

    James – re: headers. Last season I played on TP and my most frequent header of the ball won 40 in 46 games (CB) and second won 31 in 42 games (CF). I’ll look later for how many my header-ers have won this season so far (now on SS, so will be interesting to see the frequency.) I’ll try and remember to note down the monthly figures.

    General question: has anyone noticed the AI playing their sub-goalie against you and having their first choice on the bench or even left out entirely? I’ve had it a good few times this season and remember having it on PES2017. Can’t decide if it’s cocky AI resting players against little old me, or some konamical error. (Yes I just made that up.)

  15. More good highlights James!

    The stumble is a bugbear of mine. I’ve had a few dives against me. Including ones where I’ve been raging thinking I’ve conceded a penalty, just to see it overturned. Has anyone dived themselves with any success?

  16. How on earth does that work? Mrs Turf is a teacher too and simply writes me and my games off as a 13 year old boy who didn’t grow up.

    If only she had watched she’d have seen them dancing on the streets of lorient as the mighty orange secure promotion with two games to go. I can still win the title but only if other results go my way. Thus far I have a record of 19/8/9 which I feel is rather generous for a top three spot. It would have been much better were it not for the screw job month where I couldn’t buy a win.

  17. James – me too.
    Ea and Konami would benefit greatly from Employing a football loving data scientist.
    Master league, or FIFA’s career mode could be absolutely amazing, immersive affairs, with even a few more basic common sense decisions and just a sprinkling of innovation.
    I have so many ideas for a career mode, specifically ML that i’d Be confident of directing a team of devs to produce something special.

    I applied for a job at a Finnish game developer company recently as a game data analyst. Had a Skype interview today with the chief technology officer and the way he was describing the data and the game events and how they use it to intelligently predict user behaviour and as a result balance the game mechanics accordingly, was very eye opening.
    This is a 20 person startup who produce mobile Games, the thinking comes as standard for them.
    Yet Konami are a world famous multi million dollar publishing giant, and yet their thinking is so dated and reactive it’s untrue.

  18. A quick look at konami’s website buries the idea of them being the plucky little guy.

    Will definitely be third after the top two won. I’ve decided I will only react to job offers, I won’t seek an empty slot by searching leagues and resigning. I don’t mind a season in the French first division, especially as I have a good side now but should a premier league or similar come knocking I may be off.

  19. Turf – are you referring to the 2017 to date net income for just konami’s digital media division of £173million ?

    Small team with limited resources my ballbags.

  20. I was looking at their casino gaming operation Paul – they were estimating about $250m profit on that by end of march 2018….virtually a one man cottage industry, no wonder they save money on repetitive commentary!

    Odd thing I don’t think I’ve ever noticed afore happened in my close season. I’ve got no money as I’ve signed up a load of good regens on high salaries and my matches only brought in 368k last season. Consequently I was very interested in lorient’s youth team. It was bad, 59 overall and below. And every player was French. Does that normally happen? Surely not otherwise how would we ever get the likes of Toni, forlan and company if it only offers same nation players as the home team.

    Anyway only French teams came in for me, so no move.

  21. Paul/Turf – Konami has done a great job for its shareholders over the last few years, there is no way these excess profits in other divisions should be used to subsidise its small declining console gaming business!

  22. Jay – read that when it first came out about 5 years ago, just sounds like Murphy at the time, making excuses and doing anything other than taking ownership for a product that has been in steady decline over the years.

    Konami as a business make enough money, the old argument of small team with limited resources is complete rubbish, and I’m often embarrassed for people when they try and wheel that argument out.
    Abbeyhill’s point is mute too, any company with a solid and proper business ethos would invest overall company profit into any of their extended businesses, thats the whole point.
    To make hundreds of millions via casino gaming, arcade machines, mobile downloads (Yugi-Oh game has 55mil downloads!!!) and not support one of your core, best known products (PES) and let that team constantly roll out half finished, under-par, dated products is business suicide, but probably just more of an indicator of how little the upper echelons of Konami management think about PES as a product.

  23. Paul – well it clearly isn’t business suicide if you look at Konami’s group profits and share price for the last few years. The decision to focus on mobile gaming, trading cards, casinos, gyms etc where they do have a competitive advantage rather than console gaming has been a great success. I agree it would be nice for us if they invested more in PES though!

  24. Similarly, I’m not criticising konami’s business decisions, its the idea that they are a small developer struggling against the might of ea’s licensing steamroller that is clearly bunk.

    It’s something that comes up when I do presentations on my work – that forest green rovers, the ‘eco club’, are an example of what can be done at a lower level of football. They’re owned by a multi millionaire ffs!

    Think I’m going to have to head into season three with 21 players, three of whom I have listed. I have 500 quid in the bank.

  25. Abbeyhill – I meant suicide where PES Productions (as a business) is concerned, because failure to invest, has resulted in releasing sub-standard products, which is why its in the state its in today.

    PES 17 sold 40 times less copies than FIFA 17, PES 18 was outsold in the UK by NBA 2k18!!
    a basketball game, a sport which isn’t even readily available on british TV, outsold a football game in a country where football is the national sport, and on tv 6 times a week minimum, says it all really.

  26. Any attempt to position Konami and/or PES as the ‘plucky underdog’ is just embarrassing, and whoever tries it has zero credibility. It’s just an aspect of the PES-FIFA argument, akin to PS3-Xbox360 debates, Blur-Oasis debates, etc. Trying to establish a fandom position, with one’s own side as the publicly downtrodden but secretly superior force. It’s nauseating.

    James – great Legend analysis, looking forward to it even more now. I’ve seen AI players booked for diving a couple of times – once in the demo, once in the full game – and have a vague memory of seeing it happen once in PES2016 or PES2017.

  27. Paul – yes I do agree with all that. Hadn’t realised the sales discrepancy with FIFA was now as much as 40x. In that sense Konami’s shift of focus to multiplayer has been a failure too

Comments are closed.