Recommend
4 
 Thumb up
 Hide
6 Posts

Star Realms» Forums » General

Subject: Interesting stat on my personal in-app matches rss

Your Tags: Add tags
Popular Tags: [View All]
Brian Quinn
Ireland
Athlone
Westmeath
flag msg tools
badge
Avatar
mbmbmbmbmb
Hi All,
I've been recording every game I've played since the 24th of September - 865 to date and I ran some interesting queries on them today.

The first was around the level of players I was matched against. I broke my opponents into three tiers.

Give my average level during that time was: 11

Beasts = Level 16 or above (5 or more above me)
Peers = Level 7 to 15 (within 5 levels of me)
Noobs = Level 6 or below (more than 5 levels below me).

And the results:

Tier # of Games
Peers 479
Noobs 265
Beasts 202

I'm sure these stats are affected by the time of day I request games (given I live in GMT), plus my actual level, but I found it pretty interesting and figured I'd share.

More random stats to follow

BTW my IGN is: bdq
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Scott Heise
United States
San Jose
California
flag msg tools
badge
Avatar
mbmbmbmbmb
I'd be interested to see how your win% compares against each tier of opponents.
6 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Sándor Kolok
Hungary
Budapest
flag msg tools
badge
Avatar
mbmbmbmbmb
HomerJr wrote:
I'd be interested to see how your win% compares against each tier of opponents.

Well you did not ask me but I have a little bit of a related data (with not too much predictive power ).

In my early days of online SR I took on a personal challenge of challenging all the members of the 'Star Realms Online Player List' (it is a list here on BGG which has become kind of irrelevant by now, it still only has 50+ members while the FB group has swollen over 1000 members).

All in all, I played 44 games with opps btw L1-L22. My overall win% was 61% (27-17) and found no relation btw W/L and opps' L (fun fact: I lost to an L1 opp and beat the L21 ). But this is misleading in more ways than one I think.

- the sample is small
- most of my opps could be said/hypothesized to be from roughly the same 'skill-level' (only 2 of them had 20+ L and none of them 30+)
- I would categorize my skill level at that time as sg like 'not-entirely-beginner-very-low-level-intermediate'

So if I'd repeat this in a more sensible way, like challenging the top 100 in the app and, say, those ranked btw 900-1000 and then compare the results I guess that my win% would be significantly different in these 2 cases. Even more so if we tightened the numbers (like top 50 or 25) although then we'd have the small sample problem again obscuring the results (as we all know anybody can beat abybody in a single game of SR).

But even this way on the other end of all trade rows would be me, with my current skill level, and that would put a hard limit to the generalization of the results.

As I'm a pretty big stat-head I may as well do sg like this still some time.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Frank
France
Paris
flag msg tools
badge
Avatar
mbmbmbmbmb
HomerJr wrote:
I'd be interested to see how your win% compares against each tier of opponents.

Scott, you didn't ask me either but I keep track of my games results (score and opponent's level) since my 3001st game, so this may interest you.

Here's my win% per level-10 tiers for the next 1000 first games:

Level 1-10 -> 267 wins / 384 games (69,53%)
Level 11-20 -> 226 wins / 386 games (58,55%)
Level 21-30 -> 79 wins / 142 games (55,63%)
Level 31+ -> 47 wins / 88 (53,41%)
Total-> 619 wins / 1000 games (61,90%)

This includes all online, challenges and tournament games. About half were played with B prefs (65,01% win%), the rest mostly Vanilla (58,44% win%) with a few other (62,86% win%). I reached online Level 40 a dozen games before the 4000th match.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Scott Heise
United States
San Jose
California
flag msg tools
badge
Avatar
mbmbmbmbmb
FrUnit7 wrote:
HomerJr wrote:
I'd be interested to see how your win% compares against each tier of opponents.

Scott, you didn't ask me either but I keep track of my games results (score and opponent's level) since my 3001st game, so this may interest you.

Here's my win% per level-10 tiers for the next 1000 first games:

Level 1-10 -> 267 wins / 384 games (69,53%)
Level 11-20 -> 226 wins / 386 games (58,55%)
Level 21-30 -> 79 wins / 142 games (55,63%)
Level 31+ -> 47 wins / 88 (53,41%)
Total-> 619 wins / 1000 games (61,90%)

This includes all online, challenges and tournament games. About half were played with B prefs (65,01% win%), the rest mostly Vanilla (58,44% win%) with a few other (62,86% win%). I reached online Level 40 a dozen games before the 4000th match.


Impressive! surprise Good to see some actual numbers over a large data set. Your data seems to jive with the general belief that the game is not as random as some believe (as evidenced by your impressive 70% win rate against lvl 1-10), and that somewhere around lvl 15-20 is where the "elite tier" starts (as evidenced by the plateauing of your win%).

Thanks for sharing!

And also congrats on reaching level 40!!! (and for passing me up on the lifetime leaderboard )
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Frank
France
Paris
flag msg tools
badge
Avatar
mbmbmbmbmb
Thanks. I agreed with your analysis.

Luck is still a big factor in the game, particularly when playing against level 15+ or in single-match tournament.

But IMO it also contributes to the success of the game as even low-level players can still play a high-level with a fair chance to win.

1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.