Forum Clock: 2024-11-19 19:24 PST
 


Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
The Longevity Problem
#1
When I joined the SSL, one of the first things I checked was the rosters of the top teams. These teams were obviously very successful, but one detail caught my attention: the seasons in which these players were created. For example, on Reykjavik, 8 of their 13 players were created in Season 7 or earlier. Similarly, on Schwarzwälder, 5 of their 11 players were created in Season 7 or before. Between these two teams, roughly two-thirds of their players are in the regression. Seeing this, I realized that the SSL faces a longevity problem.

I would like to preface the rest of this article by saying that I have no intention of discrediting or criticizing the Board of Directors. They are extremely dedicated people who have put in significant time and effort to ensure the league’s success, and I’m sure they’ve already had conversations about this very topic. I’m writing this to address an important issue that I believe exists, in hopes of sparking more conversations that can help improve the league for everyone involved.

Why is there a Longevity Problem

To address this issue, we first need to examine the factors contributing to it. A good place to start is with the regression system, which currently works as follows:

[Image: AD_4nXd3jF0MbDhc4rDfqdHG5J0M75W91ExXQLhM...6pjE-cGpsw]

Looking at this without additional context, a few observations stand out. First, regression begins relatively late. For example, in comparison to the ISFL (the only other league I’m involved in, so expect several comparisons), regression here starts two seasons later and is significantly milder. However, this delay in itself isn’t necessarily a bad thing.

In real life, soccer players typically have longer careers than football players, so this "delay" might actually make sense. Additionally, their performance decline isn't as steep as that of football players, so the gentler regression scale could also be appropriate.
That being said, this is a sim league, so let’s consider it from a sim league perspective. Take, for example, our friend Roquefort Cotswold. From his draft season up to today, he has earned 173 TPE (note: these numbers are based on the draft stream and the current index, so they may not be perfectly accurate).

Since 173 is a weird number for math so let's use 150 as our seasonal average, and using this average we can make a rough estimate for how much TPE a max earner should have at every point in their career. Now, let’s run some numbers.
Assuming a player can earn 150 TPE each season, then after 8 full seasons, they will have accumulated 1,200 TPE. Adding in the starting TPE (350) and the academy season (another 150), a player would have a total of 1,700 TPE going into their 9th season.
Now it’s time to tackle regression. After the first round of regression (-170), you’ll drop to 1,530 TPE, but with another 150 TPE earned, you’ll be back up to 1,680—almost the same as before.

Following a second round of 10% regression (-168), you’ll be down to 1,512 TPE, but earning another 150 brings you back up to 1,662. Through these two phases of regression, the player loses a minimal amount of TPE.

As we head into our 11th season, regression shifts from 10% to 15%. However, this change has minimal impact. After this regression, we’re down to 1,412 TPE, but after earning another 150 TPE, we’re back up to 1,562.

Now, in season 12, regression jumps to 20% (the same percentage as the first regression stage in the ISFL). This takes us down to 1,249 TPE, but with 150 TPE earned, we’re back up to 1,399.

In season 13, regression hits at 25%, bringing us down to 1,049 TPE, but earning another 150 TPE takes us to 1,199.
Season 14 comes with a hefty 30% regression, dropping us to 839 TPE, marking the first time we dip below 1,000 TPE. This is significant, as I’d argue that a player with 1,000 TPE is in the “great” range, meaning we stay competitive until at least our 14th season. In comparison, ISFL players are automatically retired after their 13th season and stay at peak competitiveness for a shorter period.

However, our career isn’t over yet. After earning 150 TPE again, we’re back up to 989.

From here, regression becomes constant at 40% each season. This means in season 15, we drop to 593 TPE, then return to 743 after earning.

In season 16, we fall to 446 TPE and return to 596. In season 17, we drop to 357 and go back up to 507.

At this point, if you haven’t retired, I don't even know what to say (looking at you, Canadice)

For those of you who are visual learners, here’s a table showing a player’s TPE history:

[Image: AD_4nXcGEgA4jauZZ3DareUipbMxC05C4rKfxq04...6pjE-cGpsw]

The rows highlighted in green represent seasons where a player stays above 1,000 TPE. From this, we can conclude that a player remains over 1,000 TPE for a full 9 seasons. This doesn’t make sense from a sim league perspective. It results in teams being too strong for too long, which significantly reduces parity—ultimately making the sim league experience less enjoyable. It’s also important to note that this assessment was made using 150 as the benchmark for the season, when in actuality it can be much much more, making a player’s stretch of dominance even longer. Furious Chicken is a perfect example of this. As a Season 7 player entering his 11th season, Chicken has just experienced a 15% regression and is now at 1,760 TPE. If Chicken begins to follow our model of earning 150 TPE per season, he will be able to remain above 1,000 TPE until his 15th season.

The regression scale is, at least in my opinion, too weak and starts too late. However, this problem is exacerbated by another factor: the length of seasons.

The average season in the SSL, including the offseason, lasts around 2.5 months—shorter when there is no WSFC tournament and longer when there is one. In comparison, the ISFL season lasts about 1.5 months for both the regular season and the off-season. In my opinion, the longer SSL season makes it harder to retain active players. When we look at both the season length and career length together, the longevity problem becomes much clearer.

Assuming a player chooses to retire once they become below average for a minor league team, they would retire at the end of their 16th season. This means they would spend a total of 17 seasons (1 academy season + 16 regular seasons) on a single player. With each season lasting around 2.5 months, the player's total career would last approximately 42.5 months, or 3 years, 6 months, and about 15 days. In reality, this number could easily be much larger, creating a clear commitment problem. New players might see this long career length and feel it's too big of a commitment to make. Similarly, players who recreate and have enjoyed the league might feel that starting another player would take up too much time.

I know the SHL also has long careers, but a key difference between the SHL and SSL is the size of the user base. The SHL has a much larger pool of active players, making long careers more feasible for them. The SSL simply just doesn’t have that yet. Maybe once the league grows more having longer careers would become more reasonable, but a the moment I feel like that just isn’t the case.

So How Do We Solve it?

I wish I could provide a simple, straightforward solution—something I feel could be an easy fix—but unfortunately, I can’t. However, I believe any changes that need to be made come down to two main factors:

Season Length

This one is particularly challenging, as I don’t see much room to reduce the season length. The offseason is already very short, and the matchweeks are packed with matches. So, for this aspect, I don’t have any suggestions, but perhaps you, the reader, might have an idea.

Regression Scale

To me, this seems like the best approach for shortening player careers. By adjusting the regression scale to create a steeper decline earlier, players’ careers would become shorter and more competitive. This change would improve balance by encouraging faster turnover, preventing teams from holding onto top players for too long, and allowing more teams to challenge for the title. It would also drastically reduce the time a single career encouraging more new users to create.

The End.
I want to reiterate that I have no intention of criticizing or disrespecting the Board of Directors in any way. My goal here is simply to share my thoughts on what I believe is an important issue that could improve the league.
Reply

#2
I don't see the connection between career longevity and lack of parity. The problem isn't long careers per se but certain teams having a higher share of players in their prime. If parity is your concern tinkering with contracts and the salary cap is probably the most direct fix.

The way a career plays out does feel different than other sim leagues. I'm not sure that's necessarily a bad thing, but I could see how it would be daunting for a new or re-rolling player.
Reply

#3
(2024-11-12, 09:55 PM)Wiggli Wrote: I don't see the connection between career longevity and lack of parity. The problem isn't long careers per se but certain teams having a higher share of players in their prime. If parity is your concern tinkering with contracts and the salary cap is probably the most direct fix.

The way a career plays out does feel different than other sim leagues. I'm not sure that's necessarily a bad thing, but I could see how it would be daunting for a new or re-rolling player.

The way I see it, longer player careers reduce parity because they allow dominant teams to maintain their competitive advantage for extremely long stretches, making it harder for other teams to catch up. The dominance of Hollywood FC in the earlier seasons and the current dominance of Reykjavik are clear examples of this in the SSL. The current system creates periods of dominance rather than consistent competition.

In contrast, if careers were shorter, player turnover would naturally increase, forcing even top teams to replace talent more frequently and giving worse teams more opportunities to close the talent gap.
Reply

#4
Nice analysis!

My S1 player is still alive (only one of two players left from that class still not retired) with 410 TPE after the last regression. We will definitely see some of the same issues as the earning opportunities for later classes were a lot more than back when I started and this would be my 18th season.

There are discussions ongoing about this issue, but one of the main hurdles is that there are too many day-to-day tasks that the BoD is covering from different departments which prevents an in-depth discussion about possible solutions and a analysis on the potential consequences of different changes. I'm going to bring the discussion to the BoD about looking into creating committees of interested users that might be able to help out on that.
[Image: 0HMDG8L.png]
Thanks to @sulovilen
Reply

#5
Another small thing to note, is that the 150 TPE a season is on the lower end, as you got:
9 weeks a season,
Weekly 12 TPE:
6 TPE AC
6 TPE Task
average of 2.5 TPE prediction (for 6 weeks of these 9)

Seasonly:
40/30/20/10 TPE Training Camp (pre-regression average IIRC is 25, after regression it is only IIRC 10)
4 different season predictions, 9 TPE on average each (Not available for the academy season)

With the 350 starting TPE, this means a max earner could earn 2070ish TPE before regression (can probably put this up quite a lot, as this is excluding the academy season, excluding rookie tasks, and excluding possibly shop TPE buys).

One thing to however note is that, because of the main competition being a competition without play-offs at the end, that regression hits a bit harder as sure you get (most of it) back before the end of the season, but most games have already been played at that point.
Reply



Forum Jump:


Users browsing this thread:
1 Guest(s)

Powered By MyBB, © 2002-2024 Melroy van den Berg.