Marko Arnautovic January update I'm confused

[Rockywhu]

I expect better to be honest. Whoever is updating the stats for West Ham have got this completely wrong.

[forameuss]

He's on a fairly high value contract for a long time, in a high rep league, at a not particularly low rep club.  Value isn't something that is entirely set by the researcher, it's a function of many things.

[Neil Brock]

Yeah as forameuss has said, it's calculated by a number of factors across the game, including market value, most recent transfer, contract, league/team/player rep etc. It's not something hard set within the data and will change over time depending on other factors. 

[saihtam]

I dont think he wanted to point out the value, but more about stats and how they are considerably low side at some places.

[Neil Brock]

Ah right. Can't see any which are obviously wrong, but if the user wants to raise it, best place to do so is here - 

 

Please remember to follow the guidelines below:

- State what you think is specifically wrong with a particular piece of data.                         
- State what you think the data should be.                        
- State reasons/proof for your suggested corrections/improvements.    

[Wolvesaywe]

Never realised how bad his stats were. 18 flair and 7 decisions! Not to mention fairly average mental stats. I'd sell him soon as you can but yeah he has long on his contract so probably that's why his value is so high.

[TheBigBoss]

9 finishing is incredibly harsh

[hoppo1982]

620 Career Games. 62 Goals.

9 for finishing isn't that harsh imo 

[penza]

320 - 62 goals... not 620 games, he is only 29 and he is a winger thats not bad

[Svenc]

26 minutes ago, penza said:

320 - 62 goals... not 620 games, he is only 29 and he is a winger thats not bad

Yeah, needs to put in context.

It's also still an incredibly simplistic way to look at it. Depending on how a player is utilized, he will get different kind of shots from different positions. SI are teamed up with STATS. Whilst they also publish nonsense and should know why it's nonsense (the magic of small sample sizes and how over more attempts, say 100+, everybody tends to regress towards the mean), flawed and indefinite models such as these are only going to be more refined in the future. Research is a community effort mainly, but perhaps in the future there can be a combination. Less  opinion, more facts. That's a benefit also for the research, as they can better defend themselves (it must be pretty tiresome getting attacked over and over).

Bit off-topic, but it's a fascinating area of added info that has opened up in the past half decade or so, when all that was initially available to the general public was league tables, goals scored and goals conceded. If FM has it accurate, purely finishing skills (and thus the finishing attributes) themselves mostly won't make a league difference to the goal tally too. What makes the best forwards score significantly more goals is mostly that they get far more opportunity in consistently better position and space, with finding space naturally being a skill itself (both by the players movement, as well as the assists provided by players surrounding him). There is evidence that things work like that in-game too, no less in the many user reviews ranting about how they fail to get that "20 finishing composure decisions" forward scoring consistently -- whilst others get lesser rated players doing just that. The higher rated player may put the odd additionally difficult chance away (Ronaldo's bicycle kick against Buffon). But that guy too mainly scores this much as he averages by far the most attempts in all of Europe, from good positions too.

[santy001]

Most of the stats for Arnautovic bear my work. 

He is not the best finisher nor the most composed player, he's hardly been a 1 chance, 1 goal man throughout his career. He keeps fashioning chances, keeps getting past his man with the ball or into the box and goals can follow.

But I worked hard on making Arnautovic something of an enigma, I've said it before but there's 3 approaches that best fit Arnautovic and representing him in FM. An incredibly consistent but low ability (besides a couple high technicals and physicals) player in the 120-ish range. Doesn't really suit him.

A 140-150CA player (which he was when he departed for West Ham, and remember that's a good rating) which gives greater freedom to represent what Arnautovic is as he peaks and troughs through his career. 

A 170+CA player (which I would argue there is definitely merits for) but with lesser consistency, and being a little more critical on his mental attributes. It's incredibly difficult to hit this sweet spot without Arnautovic becoming either a massive flop you can never get anything out of or being a godly player everyone will try to pick up as West Ham would be easier for the bigger clubs to get to agree to sell. 

Arnautovic hasn't actually been changed that much on the surface in FM down the last 4-5 years, but while I can't speak for the West Ham researcher, what I did as he improved was to improve the hidden mental attributes. 

Come the winter update period, he wasn't actually in the best of form. He'd had a tough start to his time at West Ham and was starting to come good but it did look a lot like he had been a bit of a waste for West Ham coming into the Christmas period.

[Svenc]

25 minutes ago, santy001 said:

Arnautovic hasn't actually been changed that much on the surface in FM down the last 4-5 years, but while I can't speak for the West Ham researcher, what I did as he improved was to improve the hidden mental attributes.

Speaking about hidden mental traits: back then there was that bug in the AI manager's conversation history screen. It meant that you could follow the most private conversations between the AI manager and his player. I used to follow Arnautovic on every save as apparently partly due to his traits, he had a bit of trouble with every new boss he came across. Things never kept quiet, at one point there was a row over something, that soap opera never failed to kick in. Wonder if that's still happening with this frequency... or  can at all happen. Also thinking about possible legal issues, back then he had a rep of being a bit of a troublemaker. Probably one for @Neil Brock Brocky.

[forameuss]

15 hours ago, santy001 said:

Most of the stats for Arnautovic bear my work. 

He is not the best finisher nor the most composed player, he's hardly been a 1 chance, 1 goal man throughout his career. He keeps fashioning chances, keeps getting past his man with the ball or into the box and goals can follow.

But I worked hard on making Arnautovic something of an enigma, I've said it before but there's 3 approaches that best fit Arnautovic and representing him in FM. An incredibly consistent but low ability (besides a couple high technicals and physicals) player in the 120-ish range. Doesn't really suit him.

A 140-150CA player (which he was when he departed for West Ham, and remember that's a good rating) which gives greater freedom to represent what Arnautovic is as he peaks and troughs through his career. 

A 170+CA player (which I would argue there is definitely merits for) but with lesser consistency, and being a little more critical on his mental attributes. It's incredibly difficult to hit this sweet spot without Arnautovic becoming either a massive flop you can never get anything out of or being a godly player everyone will try to pick up as West Ham would be easier for the bigger clubs to get to agree to sell. 

Arnautovic hasn't actually been changed that much on the surface in FM down the last 4-5 years, but while I can't speak for the West Ham researcher, what I did as he improved was to improve the hidden mental attributes. 

Come the winter update period, he wasn't actually in the best of form. He'd had a tough start to his time at West Ham and was starting to come good but it did look a lot like he had been a bit of a waste for West Ham coming into the Christmas period.

This post should be stickied and shown to anyone who comes on and complains about Johnny Bloggs having 15 for finishing when really he should have 16 because he scored a good goal once.  A good researcher will put a hell of a lot of thought into why a player is rated like he is, and I don't think a lot of people appreciate what goes into it.  

[hoppo1982]

17 hours ago, penza said:

320 - 62 goals... not 620 games, he is only 29 and he is a winger thats not bad

My mistake. Still 320 games and 62 goals isn't bad. Finishing of 9 seems slightly low. Not overly harsh however. 

[Svenc]

57 minutes ago, hoppo1982 said:

My mistake. Still 320 games and 62 goals isn't bad. Finishing of 9 seems slightly low. Not overly harsh however. 

Ideally, you can't take a look at games and goals and draw much conclusions about somebody's finishing ability. Maybe due to a playing style or how he is utilized by his managers, he doesn't even get that many finishes. Maybe he does, but has them from far more difficult positions and phases of play.

There is a right back playing in the Bundesliga who is "ridiculed" ever other week or so, as despite a decade of top level football, he hasn't scored a single goal. That is true, but he also doesn't overly have much opporunity to do, and when he does, it is rarely from positions and spaces where anybody would be expected to score much goals. Meanwhile Dortmund's back Schmelzer, who has played a decade in the league also, did score a handful of goals in the  same space. As he does so, he never makes similar spotlight, despite he is taking far more shots (and may actually have far worse finishing, as maybe he should have scored more goals due to this).

Thinking about it, researching must be quite hard. It's very easy to fall for cliche, several perception bias and more, as unlike managers, we simply don't get to see players doing drills day in day out that may make it easier to draw something more conclusive. This applies to penalties in particular -- most are simply converted by anybody, and they aren't that regularly taken.

[hoppo1982]

3 minutes ago, Svenc said:

You can't take a look at games and goals and draw much conclusions about somebody's finishing ability. Maybe due to a playing style or how he is utilized by his managers, he doesn't even get that many finishes. Maybe he does, but has them from far more difficult positions and phases of play.

There is a right back playing in the Bundesliga who is "ridiculed" ever other week or so, as despite a decade of top level football, he hasn't scored a single goal. That is true, but he also doesn't overly have much opporunity to do, and when he does, it is rarely from positions and spaces where anybody would be expected to score much goals. Meanwhile Dortmund's back Schmelzer, who has played a decade in the league also, did score a handful of goals in the  same space. As he does so, he never makes similar spotlight, despite he is taking far more shots (and may actually have far worse finishing, as he may score more goals due to this).

You also can't assume you have more knowledge of a player than the scout who has assessed him because said player is having somewhat of a purple patch playing in a league much more visible to the majority of the world.

 

[Svenc]

23 hours ago, hoppo1982 said:

You also can't assume you have more knowledge of a player than the scout who has assessed him because said player is having somewhat of a purple patch playing in a league much more visible to the majority of the world.

 

Plus the respective managers, who get to view players to test and drill their skills day in day out. That is ultimately the challenge in researching, probably. Whilst there is now more info available to the general public than ever, including the number of matches broadcast and far more data than just matches and goals -- there is afaik only a select few of SI research who work for clubs, let alone the clubs they research. Most of them are football fans like any of us, but ones dedicated to improve the database. What will be interesting if the research process will be the exact same in ten years time as it is now.

Some overlap has already started to happen, as there is an exchange going on between SI and STATS with their databases, plus the exposure SI has gotten via Sky Sports. It's one of the most fascinating areas of the long-term development process imo, video game data crossing over into the actual football industry -- and the same may one day happen vice versa. Both are ultimately being interested in getting their numbers as accurate as can. Or in SI's case, same as their match and transfer engines, accurate enough for their manager fantasy to be deemed believable (and their customers not crying foul because Arnautovic would be rated completely wrong).

Speaking about metrics used by clubs and Arnautovic:

 

http://www.optasportspro.com/about/optapro-blog/posts/2017/blog-expected-assists-in-context/ This is a metric that played a role in Brighton bringing in Pascal Groß for a measly 3M€, who last season was relegated with Ingolstadt, but ranked big time here. He's one of the best assist givers in the EPL this term , and some instrumental in their likely survival. Naturally nobody relies on all that data alone. But it can help identifying possible targets, actual strenghts and actual weakness -- and as clubs eventually spend real money on a player, it may help them to form more than gut felt opinions about how much they are willing to part. And as this season shows right at the top, opinion even by professional pundits can be hugely wrong. By January, everybody was talking about the massively decline of Ronaldo, having scored but 4 goals in La Liga. Metrics had identified that this was unlikely to last, as he had just as much opportunity than ever, and was hugely unlikely to keep on scoring this few. Unless he was suddenly the worst forward in the history of football ever, which seemed unlikely.

[steakfaced]

On 4/25/2018 at 23:30, santy001 said:

Most of the stats for Arnautovic bear my work. 

He is not the best finisher nor the most composed player, he's hardly been a 1 chance, 1 goal man throughout his career. He keeps fashioning chances, keeps getting past his man with the ball or into the box and goals can follow.

But I worked hard on making Arnautovic something of an enigma, I've said it before but there's 3 approaches that best fit Arnautovic and representing him in FM. An incredibly consistent but low ability (besides a couple high technicals and physicals) player in the 120-ish range. Doesn't really suit him.

A 140-150CA player (which he was when he departed for West Ham, and remember that's a good rating) which gives greater freedom to represent what Arnautovic is as he peaks and troughs through his career. 

A 170+CA player (which I would argue there is definitely merits for) but with lesser consistency, and being a little more critical on his mental attributes. It's incredibly difficult to hit this sweet spot without Arnautovic becoming either a massive flop you can never get anything out of or being a godly player everyone will try to pick up as West Ham would be easier for the bigger clubs to get to agree to sell. 

Arnautovic hasn't actually been changed that much on the surface in FM down the last 4-5 years, but while I can't speak for the West Ham researcher, what I did as he improved was to improve the hidden mental attributes. 

Come the winter update period, he wasn't actually in the best of form. He'd had a tough start to his time at West Ham and was starting to come good but it did look a lot like he had been a bit of a waste for West Ham coming into the Christmas period.

This is possibly the best post I've ever read on here. Wonderful insight, and cheers. Your hard work hasn't gone unnoticed.

[hoppo1982]

21 hours ago, Svenc said:

Plus the respective managers, who get to view players to test and drill their skills day in day out. That is ultimately the challenge in researching, probably. Whilst there is now more info available to the general public than ever, including the number of matches broadcast and far more data than just matches and goals (SI can access some also via their STATS partners) -- there is afaik only a select few of SI research who work for clubs, let alone the clubs they research. Most of them are football fans like any of us, but ones dedicated to improve the database. What will be interesting if the research process will be the exact same in ten years time as it is now.

Some overlap has already started to happen, as there is an exchange going on between SI and STATS with their databases, plus the exposure SI has gotten via Sky Sports. It's one of the most fascinating areas of the long-term development process imo, video game data crossing over into the actual football industry -- and the same may one day happen vice versa. Both are ultimately being interested in getting their numbers as accurate as can as their business depends on it -- this stuff is being accessed by clubs also who pay good money in an attempt to get any small advantage they can. Or in SI's case, same as their match and transfer engines, accurate enough for their manager fantasy to be deemed believable (and their customers not crying foul because Arnautovic would be rated completely wrong).

Speaking about metrics used by clubs and Arnautovic: http://www.optasportspro.com/about/optapro-blog/posts/2017/blog-expected-assists-in-context/ This is a metric that played a role in Brighton bringing in Pascal Groß for a measly 3M€, who last season was relegated with Ingolstadt, but ranked big time here. He's one of the best assist givers in the EPL this term , and some instrumental in their likely survival. Naturally nobody relies on all that data alone. But it can help identifying possible targets, actual strenghts and actual weakness -- and as clubs eventually spend real money on a player, it may help them to form more than gut felt opinions about how much they are willing to part. And as this season shows right at the top, opinion even by professional pundits can be hugely wrong. By January, everybody was talking about the massively decline of Ronaldo, having scored but 4 goals in La Liga. Metrics had identified that this was unlikely to last, as he had just as much opportunity than ever, and was hugely unlikely to keep on scoring this few. Unless he was suddenly the worst forward in the history of football ever, which seemed unlikely.

Not gonna lie. Didn't read that. Too long. Ill take you on your word ! 

[Svenc]

2 hours ago, hoppo1982 said:

Didn't read that. Too long.

Too bad, but understandable (put a few of the extra tidbits in spoiler tags). It says a few about the pitfalls of judging players on their appearance numbers and goals. If you ever have the time, it may also improve your FM. You as well as AI can funnel all balls to low rated  finishers who score regularly as they have tons of finishes and appear in lots of space. Say due to their movement traits -- as well as the players surrounding them. Scoring  goals is not an indicator of superior finishing. It is an indicator of a player regularly appearing at the end of successful attacking moves.

 

All the while under the leadership of simply different managers, Salah either barely gets firing...



... or is on fire indeed. 

[Rockywhu]

Looking Back on this proves his finishing was too low 

[themadsheep2001]

5 minutes ago, Rockywhu said:

Looking Back on this proves his finishing was too low 

He's not got a crystal ball. Doesn't prove anything because he was rated for at the time. Not for the future. 

[herne79]

27 minutes ago, Rockywhu said:

Looking Back on this proves his finishing was too low 

In FM18 his attributes would have been based on his previous seasons at Stoke.  In his final Stoke season he scored 6 goals in 32 league appearances.  That doesn't prove his Finishing was too low.  And scoring goals in FM is about much more than just Finishing.  Read santy's (he's the Stoke researcher) highlighted post above to understand better about the work put in to arriving at Arnautovic's attributes for FM18 and before.

For FM19 the attributes will have been reviewed by the West Ham researcher as Arnautovic has now been at West Ham for a whole season and will be based in part on the 2017/18 season.  Not this season.  Last season he scored 11 goals in 31 league appearances, which is a decent enough return given that he's not always a front line striker and West Ham's mediocre season.  How will that be reflected in FM19?  We don't yet know.  As a fellow West Ham fan though I personally wouldn't say Arnautovic is a great natural finisher in the same mould as say Chicharito for example.  That's not to say he can't score goals - he absolutely can - and he has great technical ability and physical presence on his day, but being able to score goals once in a while isn't the same thing as being a natural finisher.

When FM19 is released and you've had the chance to look at Arnautovic's attributes, feel free to raise issues in the Database and Research forum that was linked for you above.  If you do, please make sure you read the guidelines there first as simply stating "I think his Finishing is too low because he scored 11 goals last year" won't be enough to get it changed .

[santy001]

The West Ham researcher will rate him as he sees fit of course, but Marko Arnautovic scored 11 from the wings for Stoke a couple years back, from what I gather last season he was essentially leading the line and still only managed to get 11. 

Checking his stats it seems in premier league football he's as likely to miss a big chance as he is to score it. With 37 goals and 37 big chances missed according to premier league stats. He averages around 32% shots on target, and he generally scores around 3 in 10 of those. 

Last season Hernandez on the other hand was getting 55% of shots on target, and he was scoring 5 out of the 10 he was getting on target. 

The big difference? Marko Arnautovic had 71 shots at goal, for that 11 goal return. Hernandez had 29 shots at goal for an 8 goal return. As I said without any stats at the time to back me up earlier in this thread. Arnautovic is a player who keeps fashioning chances. Marko Arnautovic has arguably only ever been that Hernandez level of consistency in front of goal away from being a world class player. 

- - -

One thing that really brings the debate to the fore though, how is he actually playing in game? Bear in mind that Marko Arnautovic can have any ratings you envisage, but if in the game the player you're getting is someone who is just rifling in 20 a season from long range - well that's not Marko Arnautovic. If you've got a player who is unable to beat his man, and in every 1v1 against a defender loses the ball - well that's not Marko Arnautovic. If you've got a player who's tracking so far back that it's frustrating you because he's coming way too deep to help with the defending, well that's not Marko Arnautovic either. 

Just arbitrarily saying his finishing isn't good enough, without any kind of in-game or general context is meaningless fluff. The whole aim of the research process is to represent a player in FM as accurate to their real life counterpart as we can. When you're not remotely putting forth anything about how its actually going in the game how can we react?

[bangkonggedek]

I'm curious though, was Arnautovic's attributes distribution for the most part carried over from the one given by the Werder Bremen Researcher back when he played for them? Because the erratic distribution of points does bear a striking mark of what you'll typically find on German-based players, especially from the earlier editions of FM.

I do agree that his Finishing seemed too low, because last season just for an example his shot accuracy was even slightly better than Mauro Icardi who were widely regarded as one of the best finishers around. That said, I do understand that Finishing is actually one of the more difficult technical attribute to measure though, and the stats we've had from official matches can't fully reveal the whole story.

And surely he's can't be that slow in anticipating situations?

EDIT: Hmm, interesting that the stats offered by santy001 seem to be hugely differing from the ones that I've seen, one of them was from Squawka, who reported his shot accuracy to be something around 65% while Hernandez had only something like 55%.

[DominicForza]

On 25/04/2018 at 12:30, santy001 said:

Most of the stats for Arnautovic bear my work. 

He is not the best finisher nor the most composed player, he's hardly been a 1 chance, 1 goal man throughout his career. He keeps fashioning chances, keeps getting past his man with the ball or into the box and goals can follow.

But I worked hard on making Arnautovic something of an enigma, I've said it before but there's 3 approaches that best fit Arnautovic and representing him in FM. An incredibly consistent but low ability (besides a couple high technicals and physicals) player in the 120-ish range. Doesn't really suit him.

A 140-150CA player (which he was when he departed for West Ham, and remember that's a good rating) which gives greater freedom to represent what Arnautovic is as he peaks and troughs through his career. 

A 170+CA player (which I would argue there is definitely merits for) but with lesser consistency, and being a little more critical on his mental attributes. It's incredibly difficult to hit this sweet spot without Arnautovic becoming either a massive flop you can never get anything out of or being a godly player everyone will try to pick up as West Ham would be easier for the bigger clubs to get to agree to sell. 

Arnautovic hasn't actually been changed that much on the surface in FM down the last 4-5 years, but while I can't speak for the West Ham researcher, what I did as he improved was to improve the hidden mental attributes. 

Come the winter update period, he wasn't actually in the best of form. He'd had a tough start to his time at West Ham and was starting to come good but it did look a lot like he had been a bit of a waste for West Ham coming into the Christmas period.

Thank you for the insight into your decision making.  I'm sure I'm not the only long time player who would like to see more behind the curtain.  

 

"Football is about opinion."  I'm glad they introduced editor so we can implement our own opinions of players into our own games. 

A big thanks to all data researchers who put their passion into their work...so I can have the joy of passionately disagreeing with you!  

 

I'd love to see more posts like this, thanks santy001

[sporadicsmiles]

23 hours ago, santy001 said:

One thing that really brings the debate to the fore though, how is he actually playing in game? Bear in mind that Marko Arnautovic can have any ratings you envisage, but if in the game the player you're getting is someone who is just rifling in 20 a season from long range - well that's not Marko Arnautovic. If you've got a player who is unable to beat his man, and in every 1v1 against a defender loses the ball - well that's not Marko Arnautovic. If you've got a player who's tracking so far back that it's frustrating you because he's coming way too deep to help with the defending, well that's not Marko Arnautovic either. 

This is the one thing everyone who has read or commented on this thread needs to take away as a fundamental message.

[Rockywhu]

On 14/10/2018 at 11:19, themadsheep2001 said:

He's not got a crystal ball. Doesn't prove anything because he was rated for at the time. Not for the future. 

It does actually and he wasn't even updated in January to what he should be when he was simply outstanding during December of last season and onwards.

In fact his finishing actually went down from the previous update whilst he was scoring goals as a striker before January.

Just cold solid facts there..

I can actually point out what's wrong currently wrong with his stats as it stands for FM19.

He was more Composure in front of goal his heading isn't great but definitely better than 8 and his decision making should be over 8 and he needs more strength.

 

Also 12 long shots is quite harsh when you actually look at the goals he has scored in the past and most certainly for West Ham.

 

 

 

 

 

 

 

 

 

 

[Rockywhu]

On 15/10/2018 at 12:54, sporadicsmiles said:

This is the one thing everyone who has read or commented on this thread needs to take away as a fundamental message.

Nothing like irl I can assure you of that 

[Rockywhu]

On 14/10/2018 at 12:58, bangkonggedek said:

I'm curious though, was Arnautovic's attributes distribution for the most part carried over from the one given by the Werder Bremen Researcher back when he played for them? Because the erratic distribution of points does bear a striking mark of what you'll typically find on German-based players, especially from the earlier editions of FM.

I do agree that his Finishing seemed too low, because last season just for an example his shot accuracy was even slightly better than Mauro Icardi who were widely regarded as one of the best finishers around. That said, I do understand that Finishing is actually one of the more difficult technical attribute to measure though, and the stats we've had from official matches can't fully reveal the whole story.

And surely he's can't be that slow in anticipating situations?

EDIT: Hmm, interesting that the stats offered by santy001 seem to be hugely differing from the ones that I've seen, one of them was from Squawka, who reported his shot accuracy to be something around 65% while Hernandez had only something like 55%.

Agree anticipation 7 is nowhere near right😂

[herne79]

2 minutes ago, Rockywhu said:

Nothing like irl I can assure you of that 

As a fellow West Ham fan, I disagree.  But that doesn't make me right and you wrong any more than it makes you right and me wrong - it's just subjective opinion.  It is however nice to see he's been reviewed for FM19, but that's just me speaking as a fan.

Anyway, this is going round in circles.  As mentioned above, the first thing you should do is see how the player actually plays in game.  Once you've done that (and as has also been mentioned above), if you have issues with data you can raise them in the West Ham data issues thread.  Please note you will need to supply supporting evidence, not just your opinion.

I'll close this thread here for now.