I recently wrote about the distribution of players at different ratings in the Splinterlands modern ladder (post). This is a follow-up to that post where I have tracked the development of the ratings over most of the last season. The data gathering approach is explained in the previous post.

The ladder development

I sampled ratings on the following days

• February 18.
• February 21.
• February 24.
• February 27. (Before EOS).

This is how the ratings develop over time (top is the earliest date, and then increasing dates downward).

Considering first bronze and silver, we see very little change over the season. In gold there is a steady decrease of players, while diamond has an increase of players over the season. These are the players that advance from gold, as well as players that begin playing later on in the season, and therefore have valid ratings only on the later dates.

Champion has a big increase of players over the time-frame we are looking at here. However, they pile up right above the limit for Champion league. Only a select few accounts are able to continue climbing in Champion. This is likely because when you enter champion, the rating limit of +-1000 for matchmaking is no longer valid - you can match against anyone in champion. Therefore, they get stuck at 3700 since they will begin matching against the top players. There is clearly a barrier at 3700-3800 rating. Do we want the ladder to behave like that?

Ladder climbers

Over these 8 days, the average rating gain of all players was 290. The median rating gain was 260. Note that the actual rating gains are likely going to be higher than these, since I only consider rating changes between the days I have sampled.

The highest climbers were able to gain above 2000 rating over this period. Without specifying account names, here is the rating gain of the top ten:

Climber no	Rating gain	Final rating
1	2360	3260
2	2280	2980
3	2280	3300
4	2060	2460
5	1940	5140
6	1800	2920
7	1700	5500
8	1700	2800
9	1620	5840
10	1460	3560

You will notice that the highest climbers start somewhere down in silver or gold, and climb up into diamond. These are likely alt accounts or accounts that have been inactive or playing wild. (Consider for a brief moment what that implies for the new leaderboard rules). We also have some of the top players on the top climbers list - these end up at around 5k rating.

The distribution of rating changes looks like this:

We can see that most players were able to climb between zero and two league tiers over these 8 days. Only a minority of players actually dropped rating.

Final words

I hope you enjoyed this ladder analysis post.

Please also check out the spltools app. If you are interested in posts about the spltools python package, you can find them here:

spltools update posts:

9. Updates to the spltools battle analyzer
8. spltools launches as a streamlit app
7. spltools functions and classes for battle data
6. spltools setup guide
5. Best brawler code
4. spltools update: Classes for Brawl data
3. spltools update: Reward chest classes
2. spltools update: Checking which set a card belongs to
1. Initial spltools commit

---

If you have not yet joined Splinterlands please click the referral link below to get started.

Join Splinterlands


Best wishes
@Kalkulus