Polymarket: https://polymarket.com/@abrownfox001?tab=activity
GitHub: https://github.com/abrownfox0/abrownfox001-twap60-prediction-trigger-system
Telegram: https://t.me/abrownfox001
Part 1 froze the book: BTC, 5 minutes, coin-flip band.
Part 2 is why that band is the whole business.
A 99¢ farmer lives on tiny clips and huge volume.
@abrownfox001 lives on a different equation: buy near 50¢, win more than half of the tickets that actually reach resolution, scratch the rest near flat.
If that math does not clear fees, there is no bot. There is a hobby.
The raw identity
Hold a binary to $1 at entry price (p), with held-side win rate (w):
[
EV = w(1-p) - (1-w)p
]
At (p \approx 0.50):
[
EV \approx 0.50 \times (2w - 1)
]
| Held-side win rate | EV / stake at 50¢ |
|---|---|
| 50% | 0% |
| 55% | +5% |
| 60.7% | +10.7% |
| 65% | +15% |
That +10.7% is the schematic from the public resolved sample — before scratches and before fees.
Scratches are how the live book can look better than raw (w): losers get flattened near entry instead of going to $0. That only helps if scratch spread stays ~0.
Fees change the threshold
On short-dated crypto, taker fees are not a rounding error once you stop pretending every fill is free.
Write the real hurdle as:
[
w_{\min} \approx \frac{p + f}{1 + f_{\text{net}}}
]
You do not need a perfect fee model to use the rule:
- at 50¢, a 1% round-trip leak already eats a chunk of a 55% read
- at 99¢, the same leak can erase the whole clip
- scratches that leak 2–3¢ are just fees with extra steps
This is why the engine refuses 57–62¢ “favorites.”
You paid up, so (p) rose, so (w) has to rise with it. Most short-horizon reads are not that good.
What a legal trade looks like
trusted P(up)
ask still inside 45–55¢
edge after fees >= threshold
clip <= slot cap
else skip
P(up) = 0.56 vs a 56¢ ask is not edge.
P(up) = 0.56 vs a 49¢ ask might be.
The signal estimates settlement.
The book decides whether you get paid to say it.
Why I do not farm 99¢
Convergence EV is small by design. Fees, rebates, and queue position become the strategy.
That is a different account:
- massive volume
- almost no sells
- tiny per-ticket edge
- rate-limit budget spent on sweep size
This account spends the budget on one fair clip and one clean scratch. Mixing the two styles on one wallet makes the public tape unreadable and the fee drag incoherent.
The only number that matters this week
Not one-day P&L.
After fees and scratches, did held-to-resolution tickets still clear the 50¢ identity?
If yes, keep the loop.
If no, the problem is usually fill price or scratch leak — not “we need more markets.”
Part 3: slots you must skip, even when (P(up)) looks pretty.
Not financial advice. Paper or micro-size first.
If you have more questions, please feel free to contact me at any time: https://t.me/abrownfox001
My Polymarket Activity: https://polymarket.com/@abrownfox001?tab=activity
#Polymarket #TradingBot #EV #TWAP #BTC #AlgoTrading #PredictionMarkets
Top comments (0)