I know this has been discussed in the past re. team games, but this is particularly egregious in a game like spades. When a high ELO pairs with a low ELO, their win odds go DOWN SIGNIFICANTLY in reality. using a tie in the calculation is bogus. I saw a game where a single high ELO paired with a low ELO against two opponents with low ELOs. The high ELO won, but got 2? (4?) points, his P got 50+, and both opponents got no change at all. It's quite possible that the high ELO is the only reason the partner won...and the partner hasn't earned such a huge jump
There may be no perfect solution without a ton of programing issues, but a couple of suggestions that might not be so difficult:
Use the team average against the team average, instead of individual ELO scores.
Don't count the partner in the calculation at all. (IE: calculate as if a 3p game, ignore the partner for ELO adjustment)
There may be no perfect solution without a ton of programing issues, but a couple of suggestions that might not be so difficult:
Use the team average against the team average, instead of individual ELO scores.
Don't count the partner in the calculation at all. (IE: calculate as if a 3p game, ignore the partner for ELO adjustment)