Sort:  

Boundary conditions and result testing. It's how my brain works.

I'd suggest if TB1 and TB2 result in no division, you could do TB1 (Over) followed by TB2 (Over) to effectively double your tie breaker options. But only in the case of ties caused by no one tipping under the margin in both.

If there is still a tie from TB1 and TB2 with people tipping under the margins, I think that's a fair tie, where I'd only be concerned about timing of tips and one user riding another user's decisions (or you know, copying their tip). One of the NFL tipping comps I've played here have an earliest tipper wins tie breaker, but it mostly makes me check if my exact tip set already exists when I'm making it. I'm not sure I'm a fan of this, but with open comments, it's possible for people to copy other people's choices. I doubt that's an issue, but it could be.

If there's still a tie, and the tips are different (so both got 7/9 but at least 1 of the games they got were different), it's still a tie even if you use 'earliest post' as a tie breaker for identical tips. You could use a 'rarest correct' tiebreaker for the different tips, prompting people to take chances on upsets, as the biggest upset would win the tie break if things got there.

Coin Marketplace

STEEM 0.18
TRX 0.16
JST 0.029
BTC 76343.25
ETH 3042.42
USDT 1.00
SBD 2.62