PowerPlay Magazine

Hello again fellow managers,

I present here the fourth instalment of my investigation into the soccer sports academy. You can find the previous articles here: first, second, third. As always, I continue to ask for your help by uploading HERE. Once again, please do not falsify data and be sure to fill in all entries for each player. With the usual preamble complete, I will turn once again to the reason you all are reading.

I am going to continue to wait before diving into the investigation on staff importance. Instead, I will expand on the last instalment in terms of player overall ratings expected for a given academy level. The final plot in the article

shows a clear linear trend with higher academy levels producing better players, as you would expect. What comes as a surprise that there is only a small change from one level to the next in terms of average OR you can expect when levelling up your sports academy. When you think about it, is it worth spending nearly $240M to go from SA level 13 to SA level 15 to bump your average pulls from 273 OR at level 13 to 301 OR at level 15? Let’s break this down a little further.

Chances are if you have level 13 SA you’re going to have something comparable in training. Assuming 1.3/day you can make up that 28 OR in 3 weeks and save yourself the $240M investment. Sheesh! That was easy! Why build the SA? It is a waste of money! Hold your horses here folks, it is quite the opposite.

The academy is more than just prospect production, it is also a means to boosting training 12.5% (2 weeks of 16) of the season for selected individuals. That in itself is worth it to me. But in terms of cranking out prospects, all I’ve been talking about thus far are “averages”. What is the “average” player you can expect to pull at a given academy level? It is a valid question to ask and the first step towards nailing down the mysterious beast that is our sports academy system. It is, however, much more complicated than that.

Return once again to the previous article this time to the Gaussian fits on the level 8 and level 14 data. You’ll notice the width of the fit is much broader at a higher academy level. This tells us something and I suspect many of you have noticed it already. As you progress up the SA levels you keep getting crappy players, players whose overall rating you could have expected out of SA level 5 for example. Well, these broad distributions work both ways, you can pull the weak players sure, but you also can tap into the regime of star prospects, and here is where I think the sports academy really earns its keep.

What I did next was convert the width of the Gaussian fits into a probability distribution. I won’t get into the mathematics or the theory but basically it is the same approach as my article on how lucky you were with your day 1 handball rosters. I wanted to see, for the various levels of academy, what were the odds of pulling a 300+, 350+ and 400+ OR player. The results were pretty spectacular. Please be mindful of the scale on the vertical axis, it is not the same in each of the the plots to follow.

When it comes to pulling 300+ OR players, it is logical enough that level 15 will pull them around 55% of the time given the average expected pull is 301 OR. From there the probability drops off reliably until level 10 upon which statistics become poor. To point out an example, upgrading from level 10 to level 12 will move your average expected OR pulled from 231 to 259 at a price tag of 56M or so. It may not seem worth it but looking at the plot immediately below. Moving from level 10 to level 12 will increase your odds of pulling a 300+ OR player from ~8% to ~25%. Given a range of 2-3 players pulled per week at level 10 and 2-4 at level 12+, assuming a median value for each facility you could expect 3.2 300+ OR players per season from a level 10 academy and 9 per season from a level 12 academy. That is a big difference. Let’s now proceed up 50 OR and see how things change.

Looking at the 350+ OR player pull probabilities, the trend is very much the same but now much steeper. The odds of a 350+ OR player at level 10 is about 0.4 per season while level 12 is 2.9 per season. One thing that is becoming very apparent is that while the average OR we can expect may only increase by 14 from one level to the next, the odds of pulling a star player increase dramatically. This is especially emphasized in the 400+ OR plot described next.

The last plot I am going to show you is the exclamation point on the take-home message. Looking at the probability of pulling a 400+ OR player, level 12 is now down to ~1% which corresponds to one 400+ OR player every two seasons. Even level 13 is only slightly better at around one per season. However, there is a huge difference from 13 to 14 and from 14 to 15. Level 14 has odds at around 7.5% per player or 3.6 per season while level 15 will net you 6.7 400+ OR players per season (~42% odds per pull assuming 3 players offered). Sure, on average they are only 14 OR better, but that 148M and change is a difference of three 400+ players per season. Which, depending on the time of the season (early brings more money) should net you a nice buck depending on Qs and C/L. After a couple seasons, should you choose to sell some of these players, those 3 per season more than lvl 14 could fund your lvl 15 SA on their own – which comes with the training camp bonus of more players and more training over level 14.

That concludes the fourth instalment. I look forward to comments and questions you may have. I hope you all continue to submit your players as this really is a community effort. I hope this information is helpful and feel free to make your own interpretations to benefit your team however you see fit.

Until next time,
-Scott