The current method of calculating ELO for Spades seems broken in how it treats your partner's ELO.
Imagine that you are at a table where your opponents have the same ELO as you, and imagine these two scenarios:
(1) Your partner has a much _lower_ ELO than any of you.
(2) Your partner has a much _higher_ ELO than any of you.
Doing well in scenario (1) is arguably more difficult task than doing well in (2), so one would expect to gain more (or lose less) ELO for the outcome of (1) than (2), but the opposite is true.
Because Spades always records a tie between the two partners, in scenario (1) you always record a tie with a lower-ranked player, causing a net loss of ELO, whereas in scenario (2) you always tie with a higher-ranked player, causing a net gain of ELO.
Is there a different result other than "tie" that can be used to score the relationship between partners?
Are there any other partner games on BGA that score the component with your partner differently, that Spades could use, instead?
Imagine that you are at a table where your opponents have the same ELO as you, and imagine these two scenarios:
(1) Your partner has a much _lower_ ELO than any of you.
(2) Your partner has a much _higher_ ELO than any of you.
Doing well in scenario (1) is arguably more difficult task than doing well in (2), so one would expect to gain more (or lose less) ELO for the outcome of (1) than (2), but the opposite is true.
Because Spades always records a tie between the two partners, in scenario (1) you always record a tie with a lower-ranked player, causing a net loss of ELO, whereas in scenario (2) you always tie with a higher-ranked player, causing a net gain of ELO.
Is there a different result other than "tie" that can be used to score the relationship between partners?
Are there any other partner games on BGA that score the component with your partner differently, that Spades could use, instead?
