Pokémon TCG: Sword and Shield—Brilliant Stars

Some Ratings Confusion

Status
Not open for further replies.
Bolt

Bolt

New Member
Forgive me if this is in the wrong place, but EternalFire pointed something out to me.

At Saturday's PA States, I finished 2nd, losing to Pablo G twice. Once during round 6 of Swiss and then in the top 2. At first, the results were reported with me not having lost any points for those losses. It said that he was rated 1600 for both matches, and I lost 0.00 points for them. I wasn't sure what was going on, but it left me rated 1811 after the entire tourney...about a 60 point raise.

Last night, the error/glitch was fixed, and I dropped to 1773.3, about 40 points lower. Now I have no issues with his rating being corrected, although I'd like to know what exactly happened...but here's something else.

EternalFire's round 1 - 27.74 change on 210.46 rating difference

EternalFire's round 3 - 24.19 change on 124.51 rating difference

and for me, it's like this

Bolt loses to Pablo G
1773.56 loses to 1736.48, 19.91 change on 37.08 rating difference

Bolt loses to Pablo G
1793.52 loses to 1751.11, 20.19 change on 42.41 rating difference

I'm willing to be that I'm missing something, but none of that makes sense to me. A 210 point difference shouldn't be anywhere NEAR a 42 point difference, right? This happen to anyone else?

Shine light pls
 
Last edited:
Josef, can you not use me as an example plz. Thanks,
This also would be a question better ask if you email PUI/POP rather than post here,
JMO,
Drew
 
The points you stand to win or lose is different depending on wether you win or lose.

If you're 200 points above your opponent in a 36-K event, you can win only 9 points, but you stand to lose 27.



See link in my sig for more.
 
I got that, but like...two lower rated players win. One's a difference of 200 points and the other of 40 points, what makes the point changes so similar?
 
Yes, I only gained 25 points for beating someone that was 300+ higher than me.
But when I beat someone that is only 100+ Higher I get 22 points. Seems like I should get more or less in some games.
 
Since Bolt sent this to POP via email as well, I figured I'd post my reply to his email here to help shed some light. I'm pretty sure I got this right, but I'm totally not a math/ratings wiz, so I may be off somewhere. :wink:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The difference in ratings and who won has everything to do with it. Lawrence beat two opponents with significantly higher ratings than he had. You lost twice to an opponent who had very similar rating to yours.

To overly simplify it, you apply your Win Expectancy to the K-Value to determine the number of rating points you will LOSE if you LOSE the match. If you WIN the match, you GAIN a number of rating points equal to the difference between your Win Expectancy and 100%.

Round 1:
Lawrence - 1603.80
Opponent - 1814.26

Due to the difference in ratings, Lawrence's win expectancy is only about 22.95%. Because he won that match, he gained the difference between 22.95% and 100% (77.05%) of the K-Value. Since the K-Value was 36 for the tournament, he gained 27.74 points (36 x 0.7705 = 27.738). Had Lawrence lost the match, he would have only lost 22.95% of the K-Value (8.26 rating points).

Round 3:
Lawrence - 1611.14
Opponent - 1735.65

Due to the difference in ratings, Lawrence's win expectancy is only about 32.81%. Because he won that match, he gained the difference between 32.81% and 100% (67.19%) of the K-Value. Since the K-Value was 36 for the tournament, he gained 24.19 points (36 x 0.6719 = 24.1884). Had Lawrence lost the match, he would have only lost 32.81% of the K-Value (11.81 rating points).

Basically, he got a ton of points for winning those matches because (ratingswise) he was the underdog, and managed to win.

Your case is a little different...

Round 5:
You - 1795.54
Pablo - 1717.47

Due to the difference in ratings, your win expectancy is about 61.05%. Because you lost that match, you lost roughly 61.05% of the K-Value. Since the K-Value was 36 for the tournament, you lost 21.98 points (36 x 0.6105 = 21.978). Had you won the match, you would have gained 38.95% of the K-Value (14.02 rating points).

Round 9:
You - 1793.52
Pablo - 1751.11

Due to the difference in ratings, your win expectancy is about 56.08%. Because lost that match, you lost roughly 56.08% of the K-Value. Since the K-Value was 36 for the tournament, you lost 20.19 points (36 x 0.5608 = 21.1888). Had you won the match, you would have gained 43.92% of the K-Value (15.81 rating points).

As you can see, the numbers would have been dramatically different had Lawrence lost those two matches, rather than winning them.

I hope this clears up your concerns.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
Thanks for clearing up the confusion, it just seemed kind of weird to me that the numbers weren't more proportionate to the ratings difference...but I am indeed set. Thanks F_S and Liesik!
 
ah, so thats how it works...
btw, nice meeting you for once mike (fl states, i was the guy in the bandana)
 
Status
Not open for further replies.
Back
Top