Blog Summary: May

During the last weeks I've looked into lots and lots of FML numbers.

Here is a short summary of what I've blogged about. My Voller test team (based on a weighted list of attributes) is currently ranked 15th in the GW. I'm quite happy with that given the fact that I started off as a lower league team from season 1 and also considering all the great managers that are playing in the GW.

The weighted team approach
This is where I described what I did to identify which players to buy.

GK attributes
The complete list I identified for important attributes for GKs. Of course, I have the same list for all the other positions, but I'm not going to hand them out to you. I still think there should be some mystery involved in attributes, and my weighting can only be used as guidelines anyway.

I've also written up a post on flair and one on the most important stats for increasing average rating for midfielders. It has changed my tactical set up to let all my players play with run with ball and creative freedom maxed out!

Further on - the comparison of height vs jumping, Anthony helped me out and made a great graph of the same stats that you can find here.

Feel free to use the comment field if you have anything specific you want me to look into for future posts. Or catch me in one of the GWs I play in: Voller, Vald or Sanchez!

And if you are a fan of the blog remember to add the RSS feed, or add me to your bloglist if you want to read new posts as soon as they are written!

Stamina

I've always assumed that stamina is a measurement for the amount of time a player lasts on the pitch before their condition drops to an unacceptable level.

Could this be totally wrong?

The thing about stamina that I don't understand is that I've found it quite important for all positions on the pitch. And I struggle to make sense of this using the above explanation for stamina.

Could it be that stamina is actually an indicator for how long time a player will last for each individual run?

That would explain a lot to me. Pace 20, stamina 5. The player starts off quick, but after 20 meters he is just drop dead exhausted and can't finish his run towards the goal...

Just a thought, could very well fit into the crazy ramblings category.. :)

Muda, Mura, Muri...

If you are familiar with production practices, Lean Manufacturing, derived from the Toyota Production System is arguably the most succesful and effective way of producing assets for the lowest cost.

I'll show you how I think in terms of stadium building, based on lean manufacturing.

First, I'll identify what my rep will be the coming seasons:

The top one is for my Upper league team, the bottom one for my Lower league team. I assume I'll get a top 3 position each season.
 
The strength of the new rep system is that it is very predictable, we also know when rep changes will happen in the future. So let's introduce the concept Just in Time

What I do is that I identify the time until these rep changes occur. I'll list my rep changes for my team in the GW Vald:

Current: 7.35
Tonight: 9.27
23 days: 11.07
28 days: 12.61
51 days: 14.41
86 days: 16.03

Do you get the idea here? Lean manufacturing defines anything created before these timelines as waste. Anything created after these timelines is of course equally inefficient.

So plan ahead. If your corporate boxes are going to be filled in 28 days, you better make sure that it doesn't take 50 odd days to build them!

By the way, I totally miscalculated my own rep (I blame this post from Jakswans blog), that is why I'm way ahead of what is actually needed for my stadium right now!

Team stats

A couple of days ago I scraped all available stats from the 22583 official matches played in Voller in Season 2.

2.4 goals per match in average. 1.31 YCs per team per match. 0.07 RCs.
330 passes, 17 dribbles, 19 crosses and 41 tackles in average per team per match.
16.9% goals from corners.

Some truly *doh* moments there when I looked into the stats. Like the fact that more tacklings won per match equals less goals conceded.

But I also found some decent findings. Dribbling for instance. The more dribbles a team has per game, the more likely it is to score goals. This does not impact goals conceded however, so I see no reason whatsoever not to let your pawns dribble as much as they can.

An observation I found on passing is that a higher amount of total passes per match does not impact the amount of goals scored, but it is important for conceding less goals.

An increased amount of passes NOT completed is a clear indication of more goals conceded. Maybe not rocket science, but this actually connects the dots quite well for me. I've found that the passing attribute is somewhat important for defensive players to not concede team goals, and this finding supports that theory. So my advice to you, get defensive players that can actually pass the ball!

On the topic of the hot potato of headers, the results are the same. The more headers you win, the more likely you are to score. The more headers you lose, the more likely you are to concede.

Did you expect anything else? :)

Some quick correlations

1.0 means they are always identical. This is based on my data sample for DCs in Sanchez.

Acc vs Agi: 0.72
Acc vs Pac: 0.59
Agi vs Pac: 0.61
Ant vs Cnt: 0.55
Ant vs Dec: 0.54
Jum vs Str: 0.73

Thats it. Some other attributes that didn't quite make it above 0.5:
Cnt vs Dec, Ant vs Pos, Agi vs Sta

Golden Retrievers

It's been a while since I posted anything about my Golden Retriever test team in Voller.

I just won the third run of the EFA lower league, I came second the two runs before.

I'm not too concerned about average rating, I really don't think ratings give the proper view of how my players are impacting goals scored and goals conceded.

My team consists of the following starting xi right now (including respective rank on my weighted list):
Position - AF - Av Rat - Weighted Rank - Natural Position
GK - 16k - 6.89 - crap - GK
DL - 55k - 7.55 - 200+ - DL
DC - 60k - 7.30 - 13th - DC
DC - 350k - 6.89 - 8th - DM
DR - 22k - 6.95 - crap - DR
DMC - 475k - 7.03 - 10th - AM
MCC - 90k - 6.94 - 17th - DM
MCC - 28k - 7.05 - crap - AM
FCL - 375k - 7.21 (0.3 GpM) - 9th - DR
FCC -  110k - 7.27 (0.8 GpM) - 4th - S
FCR - 150k - 7.10 (0.2 GpM) - 40th - AM

As you can see, I still have some positions left to fill. My income is limited though since I'm in the lower league. From a financial point, there is not much incentive for me to sign new players as long as I win my league.

I'm ranked 50th in the ranking system, decent enough knowing that most of the teams ahead of me are playing in the upper league.

Jumping vs height (#2)

Anthony over in Voller made this neat looking graph to view statistics from the last post about jumping visually.

Really a cool graph, I suggest you print it out and put it on your wall! :)

Dribble, pass, score!!

I've just looked into stats for 1200 matches for this MC.

He has a very high average rating, so I thought I'd find out why, and more importantly, what makes the average rating increase for players.

Remember these are still early findings and that I've only looked at stats for this one MC.

What I found is that there are 4 variables that really improves an attacking minded midfielder's match ratings.

These are:
Key passes
Assists
Dribbles
Goals

An easy way to test this is to go to Matches -> Stats for your favourite midfielder. Sort by either one of these variables, count the number of 8.0+ match ratings. Then sort descending on the same variable and count the number of 8.0+ match ratings again.

I'm thinking this could be the reason why flair is so overrated as an attribute for match ratings too. Surely a player with high flair will dribble more?

Yeah, and if you miss heading, it is number 5 on the list.. :)

Top AF country distribution

I'm not going to tell you which GW this list of countries is gathered from (I'm in 4 GWs in total, so you can still take a qualified guess).

Should give an indication of where you want your Academy to be placed to maximise high output (given that high AF = top talent). For me at least one country stood out. Still, I don't think YAs are financially worthwile in their current form, so I won't build anything there.

The list is gathered for top 1125 players with highest AF.

countr	count(countr)
CGO	1
HON	1
KAZ	1
IAM	1
MKD	1
TOG	1
LBR	1
CAN	1
GUI	1
ISL	1
LTU	1
UAE	1
MTN	1
EGY	1
GAB	1
IRQ	1
AUS	1
ALB	1
ZAM	1
SV	1
EOG	1
IRN	2
FIN	2
BOL	2
BEN	2
KOR	2
TUN	2
GRE	2
BLR	2
KSA	2
CIV	2
SVN	2
RSA	3
SWE	3
POL	3
COD	3
AUT	3
MNE	3
SVK	3
GEO	4
USA	4
IRL	4
ECU	4
BIH	4
WAL	5
CN	5
PAR	5
GHA	5
IPN	6
CMR	6
SCO	6
NOR	6
ISR	6
SEN	6
MLI	7
DEN	7
BUL	8
CHI	10
ROU	10
SUI	10
PER	10
RUS	11
CRO	13
MEX	14
CZE	14
UKR	15
BEL	15
COL	17
NGA	18
URU	18
SRB	19
TUR	26
POR	29
NED	39
GER	44
ENG	60
ESP	91
FRA	102
ITA	107
ARG	118
BRA	153

Expectation of the jumping attribute based on height

What a boring title of this post! :D
 
Anyway, I just had a chat with Hristo, the manager of BG United in Voller (he is one of the few ppl that actually added me as a friend due to my blog).

We started talking about his youth and what he could expect for his relatively low graduate DC to reach in terms of the jumping attribute.

And here is what I found out, these numbers are based on 24-29 year old random DCs in Voller.

Oh, and if you use imperial settings, then I suggest you switch over to metric numbers. Tell me again, how many inches is in a foot? :p

Looks like you need to hit 186 cm for your regens if they really are going to make it as a strong jumpers.


The cross-referenced list:

The Magical Mystery World of Flair

I'll be honest. I have no idea what flair is. Really.
A FM definition I found somewhere described it this way: "Flair is a player's unpredictability and willingness to try to pull off spectacular things."

Well. Flair is actually quite interesting when it comes to FML. 

From the numbers I have pulled, flair is in fact the most important attribute for improving match ratings for midfielders, and also the second most important attribute for strikers (after jumping)!

Further on, flair is the 6th most important attribute in terms of match rating for both DCs and DLs.

But then, the really interesting part is that fielding a player with high flair will not necessarily help your team score more goals.

The problem is, that for the life of me, I can't see what part of the match rating flair influences.
Jumping, for instance, is easy, it is the attribute that increases headers won %, which ultimately improves the match rating.

Anyone got a clue, what is the magic thing flair does that I have missed?

Goals scored - All good.

Checked 3 lower league divisions in EFA Voller.

The random sample shows
1'-15': 261 goals (17%)
16'-30': 241 goals (16%)
31'-45': 257 goals (17%)
45' + injury time: 43 goals (3%)

45'-60': 209 goals (14%)
61'-75': 220 goals (15%)
76'-90': 227 goals (15%)
90' + injury time: 55 goals (4%)

Nothing unusual here. In general, more goals are scored in the first half. Not sure how that relates to real life.

No exceptional amount of goals are scored in injury time.

When are my goals scored?

I had a quick look for my Golden Retrievers team in Voller.

When do I score goals, and when do I concede. The clips tab in FML where I got the info from only dates back two weeks, but here are the numbers I found:

Scored goals:
1'-15': 14 (16%)
16'-30': 19 (23%)
31'-45': 12 (14%)
46'-60': 15 (18%)
61'-75': 8 (9%)
76'-90': 16 (19%)

Conceded goals:
1'-15': 3 (10%)
16'-30': 6 (21%)
31'-45': 8 (28%)
46'-60': 6 (21%)
61'-75': 3 (10%)
76'-90': 3 (10%)

I'm not sure how much information I can read into this, except that it seems like I play my best in the first 15 and last 15 of each match.

The most interesting information is that 60% of all my goals scored within the last 15 minutes are actually scored from the 88th minute or later. Even though that almost makes sense if each match includes 4 minutes injury time.

No idea why the results are like this, but I'll note that I *never* sub players unless they get injured. And I've played most of my games without using match plans, so I think it could come down to higher stamina for my players when looking at goals scored by the end of my matches.

Or it could just all be totally random! :)

The secret sauce of GKs

Fact is. I don't think too many people visit this blog. So I'll give you the secret GK sauce to whoever reads this stuff!

The data samples come from two different GWs. I've probably analysed stats for more than 800 GKs in each of the GWs.

The results are the same for both GWs.

This is where you want to say that my numbers are wrong, or that this just doesn't make sense. But the numbers are right. Still, I'll admit. It doesn't really make sense.

The most important attribute for not conceding goals for a GK is Acceleration.

Yes, I'll repeat that.

The most important attribute for not conceding goals for a GK is Acceleration.

I have no idea why. But then again, I just crunch the numbers and act from there. Someone else can probably tell me why Acceleration is so important for GKs.

Here is the full list from my research for attributes for GKs for not conceding goals.

 
If someone can tell me why Acceleration is on top, please let me know..

Patrik Christensson

My newest striker. Identified as 3rd highest GpM player in Voller.

14 goals in 14 matches for me. Got him for £750k.

The other guy caused a bit of disturbance in the forums, so I thought it was better to play him as DR which is his natural position.  :)

On another note, I actually expect my team to be a bit better than it is. I'm ranked 80th-ish in the GW, not too bad for being a lower league team. Still, I've only got 4 players on the team now that isn't top 30 in their position, so I do expect more from my team!

Time will tell. It's hard to assemble a good squad with the limited resources available from not playing in the upper league. I still think the gap is too large depending on which league teams ended up in from season 1, but there is nothing to do except to look forward. At least my GpM indicator doesn't seem too far off.. :)

Peak of attributes..

Low activity in the blogosphere today so thought I'd post whatever (un)interesting stuff I have been up to since last time.

Here is a dump of average attributes for different age groups from the three GWs I've indexed.
Ignore the actual numbers (they include zeros and stuff), but take a look at the deltas and when the attributes peak.

Not much more to say really. At least you can get an idea of when and how attributes increase for youth. The data set is representative enough I guess, but remember that I've indexed players based on attributes. For instance, I've sorted players on jumping with filter u18, and this is probably why jumping has a higher average for younger players.

My newest signing..

GW Voller: Matias Perez - D/M R plays as FCC - £22k wages - 12 goals in 11 matches

He scored 4th highest on my GpM test. Just imagine what his wage would have been if he had been a natural striker!

Conundrum of the day..

If Jumping is the most important attribute for not conceding goals, why is it that Jumping is not the most important attribute for scoring goals?

The times catched up on me!

I'll let you know why I will never become number 1 in any GW.

The reason?
I'm too old. Aged 29!

If I'd been playing this game 10 years ago, I know I would have been all over the market place. I would have pestered people, chatting in the lobby, making myself a name in the GW, always looking for good buys, selling crap players expensively and generally strike good deals all day long.

Nowadays, I can't be bothered. If I want to sell a player, I'll put him up in auction for a reasonable price, and he is picked up within a couple of days. If I want to buy a player, I'll submit a couple of over-the-top offers and hope someone will accept at least one of them.

So that leaves me in a position just outside the very best of managers. I know I make mostly correct decisions in the game and that strategically speaking, I'm probably one of the best managers out there. But it doesn't add up when I'm not at pair in the transfer market. Add the fact that my tactical knowledge is somewhat limited, the conclusion is that I don't think I'll ever make it to the top spot for an extended period of time.

I feel a bit like a footballer in his last years. Even though I now have a much better understanding of the game and I know how everything will play out, I just don't have the youthful dedication that is needed for the final push towards true glory!