While checking my PredictIt account this morning, I saw that the market was valuing Sanders' chance of winning the presidency at 10%.
(I took this screenshot after purchasing my shares, which moved the market, so the image is slightly out of sync with the narrative.)
This seemed high to me, so I took a look at the conditional probability of Sanders winning the presidency given winning the Democratic nomination,
p(Sanders prez | dnom), the same thing I did for my initial bet against Sanders.
Looking at the Democratic nomination market:
Sanders is at 17% to win the nomination. So, the conditional probability of Sanders winning the presidency, given the nomination, is:
p(Sanders prez | dnom) = 0.1 / 0.17 = 0.588
A 58.8% chance of winning the presidency if Sanders gets the nomination? That still seems quite high.
So I bought some more "No" shares. My initial bet had done quite well during February, but I had been selling off shares to pay for other bets I wanted to take, so this was a welcome chance to re-up on a bet I feel good about.
[reread: 1, edits: phrasing tweaks]