pauly15 Posted November 22, 2018 Share Posted November 22, 2018 I found the last analysis project I did taught me a lot about the game. Player traits are something that I'm unclear on, so I thought I'd turn my attentions there this season. I’ve read this Player Trait Guide to start with: Traits don’t increase ability, they increase frequency of an action. Traits can work well in combination (e.g. shoot with power & shoot from distance) Traits that go against PIs and TIs will be used less often (I have my doubts) Traits that are encouraged by PIs and TIs will be used more often. Secondly, before analysing it, my impression is that the rate of PPM use is extremely high. There’s a video on the frequency of ‘shoots from distance’ here. --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- I have two players for each position, who play the same role. I'm going to try to observe what affect different Player Traits have on style and outcomes. It is not a scientific test (that would involve duplicating players and playing the same match 20 times), my 1st and 2nd team have different skill levels, attributes and make-ups. The purpose is not to say at the end “X trait is better than Y”, but rather to analyse how they work. I'm continuing on with the same tactic: First 11: Second 11: Complete Forwards (Silva as more of a f9, Zambujo more to get in behind): Silva Comes deep to get ball Arrives late in opposition area Runs with ball through centre Plays 1-2s Zambujo Beats offside Trap Cuts inside from both wings Places shots Inside Forward, Left (Chiesa more of a threat inside the area, Marin more of a threat cutting in): Marin Runs with ball often Places shots Cuts inside Tries tricks Chiesa Beats offside trap Plays 1-2s Gets into opposition area Gets forward whenever possible Cuts inside Inside Forward, Right (Tete more inside the area, Kangin just outside): Kangin Tries killer balls Runs w/ ball through centre Cuts inside from right Tries tricks Plays 1-2s Tete Moves ball to left before dribbling Cuts inside from right wing Plays 1-2s Gets into opposition area Gets forward whenever possible Beats offside trap Mezzala, RCM (Gravenberch more of an Iniesta, Bekir more of a Kaka): Gravenberch Looks for pass instead of scoring Plays 1-2s Refrains from long shots Plays way out of trouble Bekir Dictates Tempo?! Tries killer balls often Tries to play out of trouble Arrives late in opposition area Runs with ball through centre CM, MCL (Tonali more of a Verratti, Julio Cesar more of a... Lampard?): Julio Cesar Plays short simple passes Plays 1-2s Plays way out of trouble Arrives late in area Tonali Comes deep to get ball Doesn’t dive in to tackles Looks for pass rather than score Half Backs (Sabbe more of a Xabi Alonso, D'amico more of a Busquets): Sabbe Dictates Tempo Does not dive in to tackles Refrains from long shots D’amico Comes deep to get ball Dictates Tempo Plays way out of trouble Stays back at all times CWBs, Left: Pellegrini Runs with ball down left Gets forward whenever possible Marks tightly Plays 1-2s Hugs line Verbeek None CWBs, Right: Martin Gets forward whenever possible Hugs Line Plays 1-2s Knocks ball past opponent Refrains from long shots Kristensen Short simple passes Gets forward whenever possible Runs with ball down right Long flat throw Hugs line Dives into tackles Refrains from long shots Central Defenders, Defend (will be interesting to see the affect of 'marks tightly' here, they have similar attributes): De Ligt None Edilson Marks Tightly Plays short simple passes Central Defenders, Cover: Gonclaves Refrains from long shots Breton Plays short simple passes Sweeper Keepers: Lafont Avoids using weaker foot Uses long throw to start counter Ixtaurraga None Link to post Share on other sites More sharing options...
pauly15 Posted November 22, 2018 Author Share Posted November 22, 2018 Game 1, first thoughts: My AML running w/ ball through centre / cutting in, provided time for my CWB to get forward into an advanced position, whereupon the AML played a good pass to him. Having "tries killer balls often" on my AMR (IF-a) was good, as he played killer passes from advanced areas. Because we play very short, the ball gets to the AMR-a after a build up, so he has numbers forward to hit. I was pleasantly surprised by the overall PPM rate (i.e. not overly excessive). Off the ball traits for strikers simply didn't happen. We're gonna have to wait for the patch on that one. Oddly, my striker with drop deep to get ball, made far more runs behind the defence than my striker with 'likes to beat offside trap'. Marin (AML) "runs with ball through centre" worked well, giving my CWB(r) time to get forward. My AMR tried killer balls in decent areas / situations, however the lack of striker movement is terminal for this PPM combination at the moment. For "likes to beat offside trap", the guide above lists: Decisions, Off The Ball, Anticipation, Acceleration, Pace & Agility as important attributes. Finally, Refrains from long shots didn't stop Gravenberch hitting a first time shot in this situation: Odd as Gravenberch has 17 for decisions, and all relevant PIs & TIs discourage this. This doesn't do much for my belief in the statement "PIs and TIs that discourage the trait will see it happen less often". He also tried this "pass rather than looking to score": I will now wait for the upcoming patch before continuing, I think it'll change the ME significantly so it makes sense to wait:) Link to post Share on other sites More sharing options...
robot_skeleton Posted November 22, 2018 Share Posted November 22, 2018 1 hour ago, pauly15 said: Off the ball traits for strikers simply didn't happen. We're gonna have to wait for the patch on that one. Oddly, my striker with drop deep to get ball, made far more runs behind the defence than my striker with 'likes to beat offside trap'. On this note, why is it mandatory for deep lying forward, trequartistas to move into channels? Even false nines too, or support duty comelete forwards? Link to post Share on other sites More sharing options...
pauly15 Posted November 23, 2018 Author Share Posted November 23, 2018 On 22/11/2018 at 18:08, robot_skeleton said: On this note, why is it mandatory for deep lying forward, trequartistas to move into channels? Even false nines too, or support duty comelete forwards? Yeah too many roles with that. Especially annoying is the f9 one- suggest you post a bug report / feature request about that to add to mine. Link to post Share on other sites More sharing options...
pauly15 Posted November 23, 2018 Author Share Posted November 23, 2018 So! Early impressions, but first a primer on what I'm trying to do with my system. What I want is: - to build up short, with lots of passes. - to work the ball into central attacking midfield and have my AP attempt killer balls from that area. - When the ball has been worked to the advanced playmaker in central areas, I want players in front of him making advanced runs (CWBs, IFs and out course the striker). What this requires is: 1. Players being close enough to one another to present for short passes (e.g. if you have a CM(s) in MCL and an IF(a) on the left, the IF will be too far advanced to receive a short pass from the CM and your 'many passes through midfield' & 'time for players to get forward' idea won't work. The CM(s) will be stuck for options and launch it as the AP won't always be available. 2. Nobody to launch it or play fast / direct balls forward. 3. Traits that slow things down a little, giving players that time. So far, I've found the runs with ball through centre trait fantastic for my IFs on the way to achiveing the above E.g. Not a perfect example... but... without the 'runs with ball through centre' trait, the CF wouldn't have had time to make the deeper run. Also it improved the angle in this instance. Basically it works well because dribbling across (horizontally at times) from the wing into the centre takes time. Link to post Share on other sites More sharing options...
pauly15 Posted November 23, 2018 Author Share Posted November 23, 2018 The next really interesting thing to present itself is RE striker performance. My strikers play as CF(a). Keep in mind this role should see a balance of them getting in behind, drifting out wide, holding up the ball etc. They press higher, and are both relatively suited to the base role: decent strength, dribbling, off the ball, stamina etc. The two strikers with essentially opposite traits: Zambujo gets in behind, Silva comes deep / arrives late in area / runs w/ball through centre / plays 1-2s. My assistant claims Zambujo is a full star better, but he's performing far worse. Keep in mind, my system is a short passing, patient one, reliant on moving forward with fluidity, and lots of passes through midfield. You can see Zambujo's average rating is far lower, and the stats back up why. - Less shots per 90 mins - Lower shot on target % - 71 mins per goal for Silva, 336 mins per goal for Zambujo (and Silva plays in the harder matches). - Lower key passes - Lower chances created - Lower interceptions - Lower offsides! - and we even concede more as a team when he's on the pitch. Trying to figure out specifically why these stats occur, I'll look at some analysis. Silva on the left from one game, Zambujo on the right from another. (Slightly confusing as in some images Silva is attacking to the left of picture, Zambujo to the right (and in others not). And I also gave Zambujo #1;)) When you look at these, keep in mind the following: a CF(a), will naturally do a bit of everything (get behind, move into channels, hold up the ball). An important point is that a CF(a) will naturally try to get behind the defence. They will do less of one really important thing in my setup: connect with the midfield. So when thinking of a Trait for the CF(a) role to make it work with my short passing game... the question is, what trait can I give to make the role closer to what I want from the player? Heat map & average positions: When Silva plays and "drops deep to get ball" look at the deep red on the heat map. With Zambujo the whole map changes and our play is less... focused? Passes received: Neither receive a lot of the ball, but Silva far more. The only ball Zambujo receives is from out wing backs. Touches: Ignore the green box. This is Silva's "touches" stats. As you can see, Zambujo has far less touches in central areas. Interestingly, Silva received far more passes from my CM(s) rather than my AP(a). Headers: Silva wins more headers despite being significantly worse in jumping reach, bravery, strength and heading than Zambujo. No idea why- perhaps points to a hidden problem with Z. Shots: Silva is having a variety of shots, perhaps due to his 'arrives late in area' trait. Zambujo only gets poacher's chances. My conclusion to all this is: 1. With the CF(a) role, Silva's traits suit my system more. 2. I may have to change Zambujo to a DLF(a), to try to get more of what I want from him (connecting with midfield), and let his traits make up for the lack of forward runs in that role. 3. We play better as a team with Silva in, for obvious reasons. Link to post Share on other sites More sharing options...
robot_skeleton Posted November 23, 2018 Share Posted November 23, 2018 You have two very different strikers in terms of PPMs which is great! I love what you're doing! I have been looking for a striker who drops deep, I might check out Silva now 😁 Regarding the tactic, this is very interesting since you use a striker on attack, for possession systems I would always use support ones, but in FM19 I often find myself pushed too deep and cannot come out, I reckon you don't face this issue too often even in CL? Link to post Share on other sites More sharing options...
pauly15 Posted November 23, 2018 Author Share Posted November 23, 2018 I've trained the PPMs and the players from very young. I find the opposite- as Ajax I'm often playing against teams parked in their area. The main reason though is that this system was developed from the beta where playing a f9 or dlf was effectively useless. New ME is much better. Link to post Share on other sites More sharing options...
summatsupeer Posted November 23, 2018 Share Posted November 23, 2018 I would say a players trait makes a tactical instruction be used less. I would also say other tactical settings and attributes can have an effect: 1. Creative freedom allows players to deviate from the tactical instructions which I think will allow them to do what they want more often so what there traits are. 2. The players mental attributes should play a factor, especially decisions. I think a player with excellent decisions will recognize a better option over there preferred action more often than a player with same trait but lower decisions. All theory and gut feeling. Link to post Share on other sites More sharing options...
pauly15 Posted November 23, 2018 Author Share Posted November 23, 2018 Time for a comparison of my two CM (s) players, who have some different traits. First, for those wanting context: My CM(s) in LCM have some contrasting traits. Let's have a look at what affect they might be having on passing and attacking stats after 10 games. Julio Cesar Plays short simple passes Plays 1-2s Plays way out of trouble Arrives late in area Looks for pass rather than attempting to score Tonali Looks for pass rather than attempting to score Doesn’t dive in to tackles Plays way out of trouble Julio Cesar is great, and Tonali is more suitable to a DLP(s) role than a CM(s) role. The stats are interesting: The passing stats surprise me a little: Despite having "plays short simple passes", Julio Cesar has: a lower pass accuracy higher key passes, assists, & chances created lower # of attempted and completed passes (despite having played more minutes) Their attributes perhaps explain the above. Tonali has higher vision, flair, passing and composure attributes. Cesar has higher agility, pace, stamina, technique, decisions and off the ball attributes. As @Rashidi wrote earlier, attributes are relative and their combination is important, as is how they interact with a Player Trait. At first glance, it would seem that a combination of Cesar's physicals plus his very high decisions (18) and technique (17) attributes are giving him opportunities to play good passes. The goalscoring stats are less surprising. Tonali is more of a DLP: He's having fewer shots per 90 minutes More of those shots are on target (composure 17) Cesar's 'arrives late in area' combined with his attributes makes him more of a goal threat. (the only time Tonali & Cesar played in the same 11) Some analysis. Cesar on the left from one game, Tonali on the right from another (both playing CM) Average position & heat map: Interesting heat map showing Cesar's hot-spot on the edge of the area, and Tonali's deeper. This could be 'arrives late in area' related, or just a physical thing. Completed passes and chances made: More completed passes (and passes received) for Tonali, but Cesar with some key passes in his game. Touches: Another interesting one showing the amount of touches Cesar has on the edge of the area. Very different maps! This one is far more difficult to distinguish between than the striker example posted above. It may be that Cesar's physicals (+3 stamina, +5 pace, +5 agility) are saying more than any of the traits. That said, the 'arrives late' is certainly not doing him any harm with our CWBs getting up high. I'll likely replace Tonali for a more appropriate CM(s) next year. Link to post Share on other sites More sharing options...
pauly15 Posted November 24, 2018 Author Share Posted November 24, 2018 Next up, my AP(a) role in RCM. Gravenberch Looks for pass instead of scoring Plays 1-2s Refrains from long shots Plays way out of trouble Bekir Dictates Tempo?! Tries killer balls often Tries to play out of trouble Arrives late in opposition area Runs with ball through centre Quite similar attributes except Gravenberch's superiority in crossing, jumping and teamwork. I'm tempted to raise the passing directness for the AP(a), as I want him to be the one playing through balls. I'll test it out at some point. I'm worried he'll play longer laterally rather than vertically. The stats will be slightly out as Gravenberch is one of my free kick & corner takers (and we're good at corners). They're also strange in general and require further analysis: - Despite 'plays killer balls often', 16 for vision, passing, off the ball... 17 for flair... 18 for decisions... 16 for first touch... 17 acceleration... 15 for stamina... Bekir has 0 assists from 1140 minutes in all competitions!? - Despite very similar dribbling attributes and Bekir having the 'runs w/ ball through centre' trait, Gravenberch has an average of 1.43 dribbles per game to Bekir's 0.95 - Bekir has more attempted and completed passes per 90. Before looking at the team analysis screens, the only suggestion I can make is that Granberch's +4 teamwork is having an affect. That, or Gravenberch is playing at his peak. Bekir on the left, Gravenberch on the right: Heat map, average positions with / without ball, and shots: Passes completed & key passes: No key passes from Gravenberch. Passes received: Team heat-map and passes from CM(s) to AP(a): This one is interesting. Oddly, we only had 38% possession in the game on the right, so perhaps not the best one to analyse. I'll look at another two, more similar games. Link to post Share on other sites More sharing options...
pauly15 Posted November 24, 2018 Author Share Posted November 24, 2018 Bekir on the left, Gravenberch on the right. Both games were dominant wins. Heat map, average position with / without ball, and shots: Passes completed & key passes: Passes received: Team Heat-map and CM(s) - AP(a) passes: I think Bekir's relatively average stats relate to his team-work attribute of 9. How would you assess this? As for the PPMs... they don't appear to be having the desired effect / influencing their play in the way that I'd hope. Link to post Share on other sites More sharing options...
BadanieLuck Posted November 24, 2018 Share Posted November 24, 2018 Great read again! I think you won't be able to compare the behaviour of players based on single games. It just depends too much on the opposition you face, their quality and how they play and how your team performs as well. But the statistics are very interesting. I see it like this: 1. Bekir and Gravenberch seem to be very similar players, this is showcased by their passing stats and they also create the same amount of chances/90 which are pretty much the same. The difference is, Gravenberch is taking your freekicks and corners and might get his 10 assists (and shots) from that. Looking solely on the passing stats they perform pretty much on the same level. 'Tries Killer Balls More Often' might not have much effect as they play as AP(A) anyway (who will naturally do that). 2. The dribbling stat though is very interesting. Maybe it comes down to the role again as AP(A) has the 'Dribble More' active and Gravenberch follows what he is asked to do more because of his higher teamwork, while Bekir decides on his own what he's going to do next. Also take their hidden attributes into consideration (consistency etc.). But it does seem like the player traits aren't as effective or occur as often as expected. Link to post Share on other sites More sharing options...
pauly15 Posted November 24, 2018 Author Share Posted November 24, 2018 22 minutes ago, BadanieLuck said: I think you won't be able to compare the behaviour of players based on single games. It just depends too much on the opposition you face, their quality and how they play and how your team performs as well. Agreed! Watching the games and focusing on one player and their traits was yielding more than trying to understand things from the stats and match analysis screens. 22 minutes ago, BadanieLuck said: But the statistics are very interesting. I see it like this: 1. Bekir and Gravenberch seem to be very similar players, this is showcased by their passing stats and they also create the same amount of chances/90 which are pretty much the same. The difference is, Gravenberch is taking your freekicks and corners and might get his 10 assists (and shots) from that. Looking solely on the passing stats they perform pretty much on the same level. 'Tries Killer Balls More Often' might not have much effect as they play as AP(A) anyway (who will naturally do that). Yes! good point. I'll do a round of analysis on roles which have the instruction included, by putting a player with and without the relevant PPM in. 22 minutes ago, BadanieLuck said: 2. The dribbling stat though is very interesting. Maybe it comes down to the role again as AP(A) has the 'Dribble More' active and Gravenberch follows what he is asked to do more because of his higher teamwork, while Bekir decides on his own what he's going to do next. Quite possible- teamwork is sometimes partly defined as 'willingness to carry out your instructions' right? I've always thought of it as just tracking back but it's clearly much more. I find the "arrives late in area" trait interesting. If you look at the CM(s) and AP(a) match analysis screens, the two players with the 'arrives late' trait (Cesar and Bekir) occupy and receive passes on the edge of the area significantly more than those who don't (Gravenberch & Tonali). So... that particular trait is possibly having a stronger affect on movement than 'tries killer passes' is having on chance creation. Anyway, early days... Thanks for your reply though, good food for thought:) Link to post Share on other sites More sharing options...
yonko Posted November 25, 2018 Share Posted November 25, 2018 On 21/11/2018 at 23:08, robot_skeleton said: On this note, why is it mandatory for deep lying forward, trequartistas to move into channels? Even false nines too, or support duty comelete forwards? On 23/11/2018 at 02:27, pauly15 said: Yeah too many roles with that. Especially annoying is the f9 one- suggest you post a bug report / feature request about that to add to mine. This is a sore subject for me. It annoys me that SI still can't get right how a F9 role should play. And now they have adde the annoying moves into channels to CF-S and DLF-S too. At least they got the HB and the splitting CBs right in a Back 4 with DM. I hope they find solution for more interactive play between CMs and attacking players in the final 3rd. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.