Ok, a while back I posted some thoughts in the help forum to assist some players on how to simply predict the end-bloom ASI of youngsters. "pompey fc" summarised it up in the english forum, and seeing that the thread keeps popping up every now and again there I thought this might be of general interest. It is not in any way revolutionary, however apparently a lot of people have trouble with it (seeing at the amount of pm I get about it :). Hopefully some of you will find it useful, if so feel free to copy it, translate it etc.

Basic prediction for outfield players
So you have your player. He has 14 different skills. The sum of his current skills we label his "Current Skill Sum"; i.e. strength+stamina+pace+marking+ ....+ set pieces.

So to predict his skill sum at the end of bloom you do the following:

Predicted skill sum = Current Skill Sum + (0.1* Bloom-TI * weeks left of bloom)

The Bloom-TI is the average of the TI he have had sofar in his bloom, and the "0.1" stems from the fact that 1 TI-point corresponds to 0.1 skill increase (most of the time; more on that later)

So let us make an imaginary example:

Imagine J.Doe, 18y5 old. He just started blooming on his 18th birthday with an average bloom TI of 22. Thus he blooms until he is 20y11, meaning he still have 30 weeks of bloom left.

J.Doe have 7 in all his skills. This means his current skill sum is 98 (7*14).

Now what is his predicted skill sum?

Predicted Skill Sum= 98 + (0.1*22*30) = 164

So in the most simple calculation he will end up with 164 skills points. This however is not that accurate since he will loose training in various ways. We will deal with that a little later. First let us explore how ASI and Skills points are related.

Skill points vs. ASI
The ASI of a player is calculated from the amount of skill points a player has. ASI is exponential which means it grows very fast at higher ASI's. So a player with 100.000 ASI do NOT have that many more skill points than a 50.000 ASI player. Anyway here is the list of how many skill points corresponds to a given ASI.

Skills ASI
63 30
69 50
73 75
76 100
79 125
81 150
84 175
85 200
87 250
89 300
92 350
94 400
97 500
100 600
102 700
104 800
106 900
107 1000
110 1200
114 1500
119 2000
127 3000
132 4000
137 5000
144 7500
150 10000
157 12500
159 15000
167 20000
172 25000
176 30000
185 40000
194 50000
209 80000
217 100000
230 150000
240 200000

The way it was done was by taking 20 players within a small range of ASI, and then take their average skill sum. I.e. 20 players with ASI 950-1050 for the 1000 ASI point. At the highest ASI's (80.000+) it was not possible to find 20 players with almost the same ASI just like that, so the uncertainties are bigger there. However if you are even looking in that area your player is absolutely world class, so you should not complain :).

Our J.Doe had a predicted skill sum of 164 skill points. According to this table that means his end-bloom ASI will be between 15.000 and 20.000 ASI. However as said before this will not be very accurate. We need to correct for loss of training.

Loss of training correction
A player will loose training in the following ways:

- When he turns 19y0 he will not get double base-training anymore. That is why most players TI drop between 2-7 points on their 19th birthday. Thus for players early bloomers this might be important when predicting ASI.
- When a player reaches above 15 in a skill he can "loose" TI when training it. For skills between 5-15 1 TI corresponds in average to 0.1 skill increase. For skills above 15 1 TI corresponds in average to 0.05 skill increase. For very high skill (18+) even more training is lost.

Both of these effects will make a players TI become lower, than originally predicted. That is why it is necessary to correct for it. How to do so is the big question.

The most simple way is to simply make a rough guess about how much the player looses. My normal approach is to say that the player looses 10% of his skill sum.

So in this case our J.Doe example end up with a predicted skill sum of: 90%*164 skill points= 147.6 skill points. Thus a more realistic guess is that J.Doe ends up with around 10.000 in ASI (compared to the 15-20.000 predicted before).

What is fundamental to understand is that there is no set way to do it. It is all an approximation/guess.

Another way to do it, which I use on my own players is to not use their bloom TI when predicting. Instead I subtract some amount of their bloom TI. How much depends on i) the fluctuations of their TI ii) their pre-bloom base TI iii) my experience/intuition.

In the example of J.Doe, who had a bloom-TI of 22 I would normally just take off around 5 of it in my calculation.

With this system his skill sum would be: 98+ (0.1*17*30) = 149 skill points; also around 10.000 in ASI.

Goalkeepers
The table with skills vs. ASI does not readily apply for goalies. However there is a quick fix so that it does :)

You just calculate the skill sum of the goalie in the same way as for the outfield player. This number you multiply with 1.27, and the result you read of in the outfield player table.

So if you have a goalkeeper with a predicted skill sum of, for instance, 150, what you have to do to get the ASI is the following:

150*1.27=190.5

So you read off 190 in the ASI table and find that it corresponds to between 40.000 and 50.000 in ASI.

But where does the magic number 1.27 come from some of you might wonder? Probably only the ones of you interested in theorycrafting though :)

It seems that ASI is not really calculated from skills sum, but rather from average skills. And since goalies and outfield players have different amount of skills it makes a big difference.

Goalies have 11 skills, outfield players have 14 skills. 14/11= 1.27.

Skill Sum (GK) / number of skills (GK) = Average skill
Skill Sum (Outfield) / number of skills (Outfield) = Average skill

Since ASI is probably done from the the average skill it means that:

Skill Sum (GK) * (number of skills (Outfield)/number of skills (GK)) = Skill Sum (Outfield)

Thanks it's very useful :)

It's complicated, but very usefull. I have put in on my website when you first posted in somewhere on the forum:

http://www.trophymanager.info/infoEN.htm

You're too good for the 4th division Caz :) I knew all this, but you took the extra mile to write so much so others know as well.

By the way I noticed you always promote and always finish second... wierd.... give us a championship for a change!

excelent info....when i understand better, it will be very helpful

I stuck it on a spreadsheet and banged it on our website. It's handy to have it browser-based:

http://irelandtm.webs.com/ASICalculator.htm

I just copy and paste (text only) the skill stats from the player's profile, then enter in TI and weeks left at that TI.

Only thing is microsoft is tricky getting interactive spreadsheets to work with Firefox. However there is an add-on for Mozilla:

https://addons.mozilla.org/en-US/firefox/addon/1419

See details here:

http://irelandtm.webs.com/newbieguide.htm#251865176

There is a link for a downloadable version of the spreadsheet there, if you still have problems

http://irelandtm.webs.com/asicalculator.xls

also, cheers to Caz for the formula.





Last edited: 2008-10-16 11:33. Edited 8 time(s) in total.

thx for the kind words. I hope it will be useful for a lot of people.

@crimson

Yeah I finished 2nd the last four seasons :(...the strange thing is that the teams which ends 1st have never been able to promote again the season after :)

@moloneys cojones

Cheers for putting it into excel! Couldnt be bothered to clean up my own excell sheet, which is quite messy by now :) (and to make sure it worked in MS office) :)

It's not strange it's theorycraft as well... happened twice to me ^-^

And to calculate the final ASI is not better with the list ASI / TI and then apply this "training corrector" ?

If the ASI of J.Doe with 18´5 years is 500 (98 skills +/-) and the average bloom TI is 22 for the 30 weeks of Bloom, the final ASI of J. Doe with 21´11 years will be exactly 15,507.


It is much simpler than anything you've written.


Knowing the ASI at the beginning of the bloom y the TI average only need a list of correlation ASI / TI and in 10 seconds you'll know the final ASI of any player.

Last edited: 2008-10-16 01:28. Edited 3 time(s) in total.

Nice work Caz...


:-)

@ canicas

I do think we are talking about the exact same way of doing it...otherwise im not sure I understand you completely....

However in the case of J.Doe im pretty sure he will not end at 15.000 ASI, since his training will decrease.

The average TI of 22 is something you know only from his first 6 trainings. So in the end after his bloom if you calculate his real average bloom TI it WILL be lower than 22, due to lost training etc.

Why would you want to know someone's ASI at the end of their bloom?
:P

nice work

It's imprtant to note that this gives an indication only. TI does tend to decrease gradually - I have found - as the bloom progresses.

It is most useful however for deciding to sell or keep a player at the start of his bloom. It can help answer the question: will it be worth more to me to sell now, sell after bloom or retain the player for the team?

thanks a lot! this is very interesting and it helps us noobs immensely! how unselfish of you!

But then again ASI really means nothing at all (as im sure most experienced manager would agree on).

It sometimes saddens me to see the amount of money people throw after players just because their ASI is high. My best defender has 2.000 in ASI, but his skills are perfectly fitted for his position (as shame he stopped blooming a year before I thought he would :( ).

In that regard im actually fiddling around with constructing a variable like TrExMa ratings, but instead the idea is that it should incorporate how "efficient" the skills are distributed...I imagine it could be useful for finding talents.

I have ine question:
to calculate de skill sum i need the weeks left until the end of the bloom right?
but how can i calculate this?

A player blooms for 3 full seasons = 36 weeks.

Thus if your player started blooming at 18y0 and is now 18y5, then he has bloomed for 6 weeks, and thus have 36-6=30 weeks left.

Absoluely, but then you have no idea at the beginning how a player's skill points are going to be distributed amongst his skills.

I have about 4 youths at the moment who's hidden talent is setpieces - I'd far rather they were increasing in areas that they are poor on, but what can you do?

Thanks Caz.. This is really helpful..!
Y translated to spanish and put it in our coutry's Help Forum!

Not if a player blooms straight from when they are drafted then its only 35 weeks and on rare occasions 23 weeks.

nope derry. A player ALWAYS bloom for 3 seasons...however they can bloom before you pull them. So the players you talk about are players who have in effect bloomed 1 and 13 weeks before being pulled. You can recognize the latter of them starting usually with high ASI compared to their potential.

But you are right that when calculating their ASI you might get a nasty surprise if your player had a year bloom before being drawn :)

Oops yes its my fault, I keep thinking players when there drafted are aged 17.1 years old, lol. But just thought I would let some managers know that a few players have already lost a years worth of bloom. Dont worry there a trained by the best coaches on lvl 10 TG so in a way its a good thing.

It is not good laugh at what is ignored, a league win in New Zealand does not automatically convert anyone into a good manager

I´m not meaning to know the final ASI of a youth from your team, but from a player on sale and still in his bloom.

On my opinion, many people would be interested in knowing the final ASI for a potential buy. It´s like playing poker with marked cards… in a legal way though.

By having the ASI/TI list you can know the TI of any player on sale with deadline between Tuesday, 5 A.M. and Sunday, 5 A.M. And by knowing that you can see as well roughly his final ASI.

That allows you to do without any doubt transfers like these:


http://trophymanager.com/showprofile.php?playerid=8291500

Bought for 10 million, sold by 164 millions after 18 weeks

http://trophymanager.com/showprofile.php?playerid=8066510

Bought for 50 million, sold by 150 millions after 9 weeks


And I have more

So are you still maintaining that nobody would be interested in knowing the final ASI for a player?

I am quite sure he was joking... Being sarcastic :)