Going back to my project of the ultimate data crypto trader. I've had to re-evaluate the raw trade data and how I get it. "Just track the trades?" is super not easy onchain. I've pivoted to parsing whole blocks at a time for token and eth differences. That way I can account for

The good folk at AMPLE know the value of community Booosting. Co-ordinated efforts will get you to the top of the t.me/BobbyBooost charts in no time. Does your community have what it takes?

Bobby has had an obscure memory leak for the past 6 months that I just could not find. Every AI up until now also failed to find it. Enter Gemini Pro 3.5.. nailed it first go. Gotta love AI. (It was an undocumented web3.py cache thing).

The problem with utility is it actually has value. If it has value you can estimate a valuation… which caps its mcap, on the other hand if something has no value it has no cap, so value can be infinity. Weird.

RWA projects are in a bit of an unfortunate dichotomy category. They're too slow and low yielding for degens, too risky for normies. An almost impossible spot.

x.com/i/article/1955…

Isn't it funny how KOL's are both almost unanimously hated, but also agreed are necessary? Tough gig.

Utility tokens that have a signal/calls utility that only post their winning calls to social media and not their winning % are....annoying.

Good bye Bobby. Hello.... 🤔😉

Seeing "You're absolutely right!" is going to give me PTSD.

Curious coincidence of the day.. the negRiskAdapter NO_TOKEN_BURN_ADDRESS (0xa5ef39) making an appearance in the polymarket API docs as an example proxyAddress. polymarket.notion.site/Polymarket-Dat… Polymarket jonamen.eth

ohhhhh boy. ive swallowed the polymarket pill. bad.

ahh iteration. I've re-downloaded the entire polymarket transaction data (hundreds of millions of rows) set like 4 times now as ive discovered what data i do and do not need.

✅Download millions of polymarket trade related contract events using SQD ✅ Parse events into different buy/sell/transfer transactions ✅ Build a position tracker ✅ Build a wallet stats tracker ✅ Build an aggregated wallet stats to token stats tracker ...oof.been a busy week

Whipped up a live tracker for F1 using API's. This data comes through a solid 20-40 seconds before the live stream... gonna be handy to throw some cash towards polymarket F1 markets...

Don't you love it when trying to diagnose a blockchain parser issue by verifying against the scan website, only to find out the scanner website is wrong (polygonscan in this case) and you're code was right all along?

Naive me: "oh, I just need to track orderFilled events to get accurate polymarket trade data"... 3 weeks, 5 contracts and 12 different event types later .. i think I have it. I think... ugh.

Win rate % is not a great metric for following polymarket traders, due to the fact that (unless you trade out of a losing position) you will lose 100% of your proceeds in a loss. 10 wins of 10% each gets wiped out with 1 loss of 100%, but still shows a 90% win rate.

4 weeks of deep diving into how Polymarket works and getting the raw data from millions of trades shaped into a nice lil package. First back test of my theory returns a 77% win rate. Still some work to do, but nice to see theory playing out.

If you think you can vibe code a proficient polymarket trading bot based off an x post... you're going to have a bad time.