2026-06-29 · Kalshi · 31 条相关讨论
<@717963841245282355> <@1066645794108735510> after LA Sen, i just want to say trading open-ended margin tranches is so much nicer than the bounded ones
<@717963841245282355> <@1066645794108735510> after LA Sen, i just want to say trading open-ended margin tranches is so much nicer than the bounded ones
its super intuitive, and it makes it easy to discern what the market thinks the O/U is. i also think for big races it's gonna be more attractive for retail, who will get intimidated trying to guess whether a result will fall between like, 3 and 6
its super intuitive, and it makes it easy to discern what the market thinks the O/U is. i also think for big races it's gonna be more attractive for retail, who will get intimidated trying to guess whether a result will fall between like, 3 and 6
"i think my candidate wins by 10 at least" is more attractive to bet on than "i think my candidate wins by 9-12% exactly"
"i think my candidate wins by 10 at least" is more attractive to bet on than "i think my candidate wins by 9-12% exactly"
i can see modelsharps preferring the bounded ranges because it allows them to deploy capital more efficiently, but that doesn't matter if the retail flow is worse
i can see modelsharps preferring the bounded ranges because it allows them to deploy capital more efficiently, but that doesn't matter if the retail flow is worse
not sure what u mean by deploy capital more efficiently. u can take the same positions in both types of market, so it’s just a question of convenience.
not sure what u mean by deploy capital more efficiently. u can take the same positions in both types of market, so it’s just a question of convenience.
Negrisking vs directional netting
Negrisking vs directional netting
So if your model says a candidate wins by exactly 11%, with narrow bounded strikes, you can probably buy like 9-12% for $.20 or lower (thinking about some recent elections where this was the case). With open-ended strikes, you'd probably have to buy like 10%+ for $.40, 5%+ for $.80, 15%+ N for $.80, etc.
So if your model says a candidate wins by exactly 11%, with narrow bounded strikes, you can probably buy like 9-12% for $.20 or lower (thinking about some recent elections where this was the case). With open-ended strikes, you'd probably have to buy like 10%+ for $.40, 5%+ for $.80, 15%+ N for $.80, etc.
But my view is that increased retail flow would make the latter more profitable overall, since the former isn't profitable if you're the only person on the OB
But my view is that increased retail flow would make the latter more profitable overall, since the former isn't profitable if you're the only person on the OB
the directinoal markets have directional netting so u dont tie up extra capital
the directinoal markets have directional netting so u dont tie up extra capital
Oh then if that's the case, all the better
Oh then if that's the case, all the better
so if the strikes are 9+ and 12+ and the margin is close to 10, u can just add a 10+ strike. with mutually exclusive brackets youd have to make a whole new market
so if the strikes are 9+ and 12+ and the margin is close to 10, u can just add a 10+ strike. with mutually exclusive brackets youd have to make a whole new market
literally just buy whatever bracket rich baris says is good
literally just buy whatever bracket rich baris says is good
I agree, I don’t bet on these, because I know to be able to right price for 9-12% or something would require some real number crunching. Whereas greater than 10% can be done more on vibes and intuition
I agree, I don’t bet on these, because I know to be able to right price for 9-12% or something would require some real number crunching. Whereas greater than 10% can be done more on vibes and intuition
<@940487785029185589> the current problem with directional netting being used even more extensively is the fact all Yeses must precede all Nos, so if it's multi-candidate then it won't work unless you make separate events (which may be annoying to navigate between)
<@940487785029185589> the current problem with directional netting being used even more extensively is the fact all Yeses must precede all Nos, so if it's multi-candidate then it won't work unless you make separate events (which may be annoying to navigate between)
e.g., if I wanted to have Candidate A >3%, >6%, >9% and Candidate B >3%, >6%, >9%, these structurally cannot be in the same event
e.g., if I wanted to have Candidate A >3%, >6%, >9% and Candidate B >3%, >6%, >9%, these structurally cannot be in the same event
buy the cheapest on for best return, and the most expensive one for the most likely winner
buy the cheapest on for best return, and the most expensive one for the most likely winner
so since CO-01 and Gov (D) are relatively close the MOV will probably be MECNET, but as always the specific candidate vote percentages can be directional. but longer term I'll try to figure out a solution with product/eng
so since CO-01 and Gov (D) are relatively close the MOV will probably be MECNET, but as always the specific candidate vote percentages can be directional. but longer term I'll try to figure out a solution with product/eng
i guess directional for vote share and mecnet for mov probably works (unless there is a clear favorite)
i guess directional for vote share and mecnet for mov probably works (unless there is a clear favorite)
can u just construct a new collateral return mechanism for this tho
can u just construct a new collateral return mechanism for this tho
ahhh gotcha. IIRC the way this was handled in the LA Mayor "dashboard" was pretty intuitive
ahhh gotcha. IIRC the way this was handled in the LA Mayor "dashboard" was pretty intuitive