FM15: 17/18 Dortmund: Always the bridesmaid, never the bride

The new season promised to make up for the disappointments of the year before. We made very little impact domestically or on the continent, and despite finding ourselves in the Europa League, I was determined to go far.

Transfer dealings were performed largely based on my attribute & role rating system I mentioned in my previous posts. Five of the signings were made for the first team, with four having a strong recommendation from the system; whilst the other was a keeper (Timo Horn). I’m not quite sure why I didn’t set up the system to cater for goalkeepers, but it doesn’t so there we go. Something to add for future.

Selling a few key first teamers gave us a fairly large transfer chest. One of our top performers in 16/17 was Douglas Costa, but a move to Manchester City turned his head, and the club was more than happy with the £23m fee. Max Kruse (£14m), Walter (£8.5m), Milos Jojic (£9m), Jonas Hoffman (£5m) and Erik Durm (£9.5m) were also shipped out after the system insisted they were all pretty rubbish. Leo was then sold for £6m after the opportunity to sign the high potential Lucas Boye arrived, and Kevin Grosskreutz was going to stay until a £15.25m offer arrived from Leverkusen; too good to ignore for a 29 year old.

This warchest meant we could sign whoever we wanted. But no Champions League football meant that even despite the cash, some players chose to go elsewhere. I’ve never been the type of manager to splash the cash unless it’s a player I badly want, and the top target was one of my old Southampton players: Eder Alvarez Balanta. With a £22.5m and £95k p/w deal agreed, the Colombian was due to sign until interest leaked and Premier League champions Arsenal hijacked the deal. Not like I brought you to England and made your name, you b<>astard.

With Balanta off the table, it was a number of smaller deals that took my fancy and I relied on the system to give me value for money.

The Role Rating System: The good

Signing a new right back was the most important piece of business, and the system really came up trumps. With no regard for position, we eventually settled on Hirving Lozano for a cut-price £12m from Napoli.

Lozano prior to purchase:

And after the season with us:

I was delighted to find Lozano, because he fits both the FB-A and CWB-A roles incredibly well. On the latest version of the system, his attributes at point of purchase give him a 7.20 as a FB-A, and 7.14 as CWB-A. These attributes have since improved slightly, and he’s now competent in the position, leading to a number of excellent performances. He became a key part of the team, playing as a right-back in home games and on the right wing in tricky away games.

The other major signing was Dutch central midfielder Jordy Clasie. With Gundogan as the only deep playmaker in the squad, there was a pressing need for reinforcements. Clasie came at good value, again £12m, but also a strong Regista role recommendation from the system: 7.53. His 70k p/w wages are fairly high, but his transfer fee provides excellent value, and his signing means the midfield has plenty of different tactical options.

Our striking options were given a major re-jig over the Summer. Our three top scorers of the previous season were sold, and I piled my faith into Ciro Immobile. At the beginning of the Summer I was ready to sell the Italian, but the system was keen on him as an Advanced Forward. It also told me that he was actually pretty crap at almost every other role, so Lucas Boye was signed as the man who would drop deep from the initial two-man forward line. As first choice striker, and only ever playing as an AF-A, Immobile had an astonishingly good season. In his first three seasons at the club, he had scored 33 goals in 77 games in all competitions. This year he managed 39 goals in 46 games. I was more than ready to get rid, but the system told me otherwise and my faith was more than rewarded.

The Role Rating System: The bad

There was only one major error from the system, and it was fairly large.

Whilst perusing the transfer market, I saw Marouane Fellaini listed for a measly £2.5million. I didn’t expect to want to sign him, but plugged his attributes into the system regardless. Out came a completely unexpected 8.51 rating as a Target Man. At such a low fee, I gambled and we signed him, to use in the Wide Target Man – Support role.

Unfortunately it didn’t really work out. Whilst Fellaini is *really* good at fulfilling the basic tasks of a target man, he couldn’t do much else. In attack, our fluid pressing system was often disrupted. We needed to use an extra man going forward because he was never really potent enough, and he’s not nearly talented enough to use as a luxury player like that. Despite getting us up the field excellently, as he almost always won the aerial challenges after keeper Timo Horn picked him out, he couldn’t counter attack. His acceleration is very poor (8), and this meant play bypassed him.

When we did come up against a low-block, he was pretty handy. The left back was told to maraud down the wing, with Fellaini dropping inside. However, despite asking him to stay narrow, he never really got in the box for the headers, preferring to tuck in on the edge waiting for second balls. The only time he went for a header in the box was when he attacked the back post when a cross came in from the right. He did score a number of goals though, finishing with a non-penalty goal (NPG) per 90 minutes, of 0.39. His aerial ability also gave us something different in the Europa League final, where his headed equalizer in the 92nd minute gave us hope before we lost 9-8 on penalties.

I’d like to keep Fellaini around. He’s so very different from anything else and he’s handy to have. But he’s never going to be Plan A, or even Plan B. That seems to have pissed him off a bit as he’s asked to leave. It’s quite unfortunate but I’m not going to risk upsetting team morale for a Plan C.

The Rating System: How does it work

It’s pretty simple really. For each role, there are four different levels of importance. The most important attributes are given a weight of 2, the second most: 1.5, third: 1, and least important: 0.5. Some attributes have no weight at all.

So an example, the Regista – Support role looks like this:

R-S (10, 4.5, 4, 1.5 = 20):

First Touch
Passing
Technique
Composure
Vision

Flair
Off the Ball
Balance

Anticipation
Decisions
Positioning
Teamwork

Tackling
Aggression
Concentration

Adding up all of these attributes (considering the amount of attributes in each level of importance), we get a weighted total for the number of attributes. Despite there being 15 attributes listed, the weighted total is 20 because not every attribute is worth the same.

So the formula looks like this:

For Jordy Clasie, in the Regista – Support role, we get:

This gives us a nice round 15 out of 20, or 7.5 out of 10.

Now we come to the issue of comparing roles. It’s much easier to do well in specialist roles like Limited Full-Back (weighted total attributes level of 14.5) than in a role which requires lots of different attributes, like Complete Wing Back (weighted total attributes level of 21.5). I used the standard deviation and mean of the total squad’s ratings for each role to calculate a new set of ratings. However, I found this was slightly *too* harsh on the ratings for specialist roles like LFB. So the end results have a mix of both.

The first set of ratings, for example, give Jordy Clasie a 7.12 rating as a Limited Full Back. The second set of ratings consider how easy it is to be good at that role, and weight it down, giving him a score of 6.90 for that same role. The end rating that I use when discussing and comparing players uses these two ratings and finally decides he’s a 7.06 as a Limited Full Back.

The end value numbers roughly compute to this:

9.5: greatest of all-time
9.0: all-time great
8.5: world class
8.0: star
7.5: Champions League calibre
7.0: first division top half
6.5: first division mid-table
6.0: first division bottom half

(so you can see why I got a bit excited at Fellaini being rated an 8.51 target man!)

I’m sure I’ve lost a lot of people in that explanation, but people were asking how I get to these ratings so I felt obliged to answer. If I’ve not made any sense, then please do contact me on Twitter and I’ll try to be a bit more clear. Also if you want extra detail on how I calculate that second set of ratings, I’m also happy to help.

A copy of the spreadsheet can be downloaded here:
https://mega.co.nz/#!0xsSzIhJ!haa56Qgb9Tqh9PG0pukpG3JEXH2HRwIaB2Uj42IpdNY

All you need to do is input player names and attributes and your away.

So, ummm, back to the game

The season itself went pretty well. As the title suggests, though, we didn’t find glory. We came 2nd in the Bundesliga and 2nd in the Europa League. To be honest, we were never really in with a shout in the league. But we went on a fantastic run near the end of the season, winning nine straight league games. Bayern then lost their first game of the year, and it was pretty close. A 1-0 loss away at Gladbach finished us off, though.

Our poor start meant it was too difficult to catch Bayern. Difficult to justify the disappointment too; a team who loses 8 league games really doesn’t deserve to win the league.

Bringing the youngsters through

We’ve got a number of promising youngsters who’ll break through into the first team in the coming seasons. For screenshots on the best ones, go here: http://www.footballmanagernow.co.uk/forum/football-manager-2015/fm-general/the-scout-hut/showcase-your-newgen/8833-mc-andriy-yakovenko

But here’s a screenie of my favourite. A nice note to end on.

Advertisements

14 comments

  1. noikeee

    This system sounds very well thought, but do you input the attributes manually from each player of your squad, and each possible signing into the Excel sheet? That’s tons of work…

    • 13s

      No.

      Get the squad view up, with the attributes up. Then use the FM tab to Print Screen as a webpage. Then import the data into Excel.

      You’ll probably need to do this separately for Technical, Mental and Physical attributes as they don’t all fit onto the squad view at the same time (at least, they don’t for me).

  2. Ose

    Are you worried that your role rating system might make you miss some of the advantages of having a generalist as well as specialists for roles. Also how do you account for two-footedness?

    Having said that I can see massive advantages in your scouting system helping you to be bold in your transfers, something which I find hard once a player has a couple of good seasons with me.

    • 13s

      No, I don’t. Because I get ratings for every single role, I can more than appreciate a player who is good at a range of roles. And if they provide good value, I’m more than happy to have them in the squad.

      I’d say the system actually helps in that regard. I spend less time pondering which other positions a player could be retrained into and it gives me more time to look at details like footedness & languages spoken that I might have disregarded previously. Having one number out of 10 for each role makes it much easier for me to comprehend the +s and -s of each player. Maybe that’s just me, but too much information inevitably means I miss, forget or disregard stuff. This way, it happens much less.

  3. alanantaise

    Hey

    Thanks for this great post and the spreadsheet to boot. Just wondering if the reference mean and stdev values you use for each position are calculated across your team only or across the top n leagues in europe. Presumably if i want to perform the same type of comparison for my team i would need to plug in my teams mean and stdev values?

    • 13s

      Yeah, they’re from several of my teams. I could do all the top European leagues but it’d take ages.

      If you want star ratings relative to your team than you could plug in your own values, yeah. Doubt it’d end up as out of 10 values though. More like 12 or 13.

  4. Mike_E

    Great post. Thank you!

    Nice tip about using Print Screen to export data to a web page. I didn’t know you could do that. I guess you use the same approach to pull data for your spider graphs?

    Thanks for the spread-sheet! I spent some time picking through it to make sure I understand how it works. (As an aside, it imports into Google Spreadsheet with no problems, which is handy if you don’t have Excel). Could you elaborate a little on how you picked the weights for each position? I noticed, for example, that for ‘FB-A / WB-A’ you including Finishing and Long Shots at the same level as Marking. I would never have thought to do that.

    One minor correction…’DM-D / BWM-D / A-D / CM-D’ is divided by 21.5, but the weights only sum to 21.

    Thanks again,

    Mike

    • 13s

      Cheers Mike, glad you liked it. Also glad you spent the time to have a look at the spreadsheet. Given there’s so many row of stuff, I’m sure I will definitely have made more than a few errors!

      The weights for each position are partially the attributes the role suggests (if the game suggests a particular attribute then I think I’ve given it *at least* a 0.5 weighting). It’s also based on my many years experience of playing FM and the needs for each position. Having played almost exclusively with attacking Full Backs or CWBs since FM13, I know how many chances they can get in the attacking end. They really do form a vital part of the attack, and I know how frustrating it is to have an attacking FB that can’t finish. Obviously it’s just personal preference and others may feel differently.

      Maybe the difference between Full Backs and CWBs should be the finishing (I still value long shots for FBs), but I quite simply couldn’t be bothered to separate them just for that.

      & cheers for correcting the error; I appreciate it.. As I say, lots of stuff on the spreadsheet, bound to be more in there. Ta.

  5. Michael

    Hi, I didn’t see either Complete Forward or Poacher listed in the “roles” section. Is this because you don’t use them in your system? Also, both my DMs were listed as being better as FBs, and both my FBs were listed as better wingers. Do you think there’s a bias in the methodology? Thanks,

    • 13s

      Hi Michael,

      I never ever use a Poacher; I use an Advanced Forward instead. There is CF-S/A in there, though.

      & yes, I do think there’s bias in the methodology. But unfortunately it’s very, very difficult to get anything precise and truly accurate unless I used much, much more data. And quite frankly, I just can’t be arsed.

      I’d argue that DMs often make excellent Limited Full Backs (which I imagine is your problem right now?). Think of Javier Mascherano playing RB.. no-one is getting down that flank.

      As for your FBs.. what are their strengths & weaknesses? If they’re not focused on crossing & dribbling then it’s likely they could make decent DMs.

      They’re not absolute comparisons, and I don’t think they ever will be. It’s just meant to open your eyes to the possibilities available. There’s times I’ve completely ignored the spreadsheet because I back my own management, but other times it’s been really handy and I’ve identified completely new roles for players.

      Even if you don’t end up retraining or re-using your players, it makes you think about their strengths, weaknesses and suitability for different roles and that can only be a good thing.

  6. Michael

    Thanks for the quick and thorough reply. I’m a big nerd, so the thought of being able to reduce football to numbers was a messianic vision (they’ve done it in baseball, but it’s such a boring game). The list certainly inspired me to look closer at the attributes, rather than relying on coaches reports’ stars, so that’s a positive. I also think the list would be perfect for judging players in the same position e.g. Balanta vs. Pique. I play Barcelona so I appreciated the Maschereno comment. For Barcelona, it said both Maschereno (by a lot) and Busquets (by a little bit) would be better as FBs, and Alves (by a lot) and Alba (by a little bit) would be better as wingers. I have three fantastic AP-S in Xavi, Iniesta, Rakitic, neither of which are good in more defensive roles. So in addition to Suarez/Messi/Neymar, that’s six attackers. Maybe I should sacrifice one for an extra defender, or a link between defense and midfield. I’m now going round and round in circles trying to set up the best tactics for my players, which is exactly what you were hoping to inspire – thinking like a real manager!

    I appreciate opening up a whole new world for me in FM.

    (Someone’s going to comment that Barcelona’s not a real challenge – I haven’t played since FM10).

  7. Dehsyi

    1. Could you divide Poacher and AF. I think they are sligthly different in use.
    2. How do you set values in BW30:DG30 and BW31:DG31 cells?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s